The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
S. H r g . 103-793
HOMEOWNERS’ INSURANCE DISCRIMINATION
HEARING
BEFORE THE
COMMITTEE ON
BANKING, HOUSING, AND URBAN AFFAIRS
UNITED STATES SENATE
ONE HUNDRED THIRD CONGRESS
SECOND SESSION
ON
THE AVAILABILITY, AFFORDABILITY, AND ACCESSIBILITY OF HOME
OWNERS’ INSURANCE, PARTICULARLY IN URBAN NEIGHBORHOODS.
COMMUNITY ACTIVISTS AND INTEREST GROUPS ARGUE THAT MANY
URBAN AREAS HAVE INSURANCE PROBLEMS BECAUSE OF INSUR
ANCE DISCRIMINATION BASED ON RACIAL AND INCOME CHARACTER
ISTICS OF A GEOGRAPHIC AREA—I.E., REDLINING
INSURERS RESPOND THAT THEY PROVIDE COVERAGE BASED ON
COLOR-BLIND UNDERWRITING GUIDELINES AND THEY SIMPLY DIF
FERENTIATE BETWEEN RISKS IN PROVIDING COVERAGE TO INDIVID
UALS AT A PRICE THAT REFLECTS EXPECTED LOSSES; HIGHER
PRICES, PARTICULARLY IN URBAN NEIGHBORHOODS, THAT REFLECT
A GREATER RISK OF CRIME, VANDALISM, OR LOSS
MAY 11, 1994
Printed for the use of the Committee on Banking, Housing, and Urban Affairs
U.S. GOVERNMENT PRINTING OFFICE
84-051 CC
WASHINGTON : 1994
For sale by the U.S. Government Printing Office
Superintendent of Documents, Congressional Sales Office, Washington, DC 20402
ISBN 0 -1 6 -0 4 60 8 0 -8
COMMITTEE ON BANKING, HOUSING, AND URBAN AFFAIRS
DONALD W. RIEGLE, JR., Michigan, Chairman
ALFONSE M. D’AMATO, New York
PAUL S. SARBANES, Maryland
PHIL GRAMM, Texas
CHRISTOPHER J. DODD, Connecticut
CHRISTOPHER S. BOND, Missouri
JIM SASSER, Tennessee
CONNIE MACK, Florida
RICHARD C. SHELBY, Alabama
LAUCH FAIRCLOTH, North Carolina
JOHN F. KERRY, Massachusetts
ROBERT F. BENNETT, Utah
RICHARD H. BRYAN, Nevada
WILLIAM V. ROTH, JR., Delaware
BARBARA BOXER, California
BEN NIGHTHORSE CAMPBELL, Colorado
PETE V. DOMENICI, New Mexico
CAROL MOSELEY-BRAUN, Illinois
PATTY MURRAY, Washington
STEVEN B. H arris, Staff Director and Chief Counsel
HOWARD A. MENELL, Republican Staff Director
GLENN IVEY, Counsel
MARK Kaufm an, Financial Policy Analysis
T im othy P. M it c h e ll, Legislative Assistant
DOUGLAS Nappi, Republican Counsel
JONATHAN Kam arck, Republican Staff Director /Subcommittee on Housing
Edward M. M ala n , Editor
(II)
CONTENTS
WEDNESDAY, MAY 11, 1994
Page
Opening statement of Chairman Riegle ..........................................................
Opening statements, comments, or prepared statements of:
Senator Feingold.........................................................................................
Prepared statement ............................................................................
Senator D’Amato........................................................................................
Prepared statement .............................................................................
Senator Kerry ...........................................................................................
1
9
44
11
45
12
WITNESSES
Congresswoman Cardiss Collins, U.S. Representative in Congress from the
7th District of the State of Illinois................................................................
Congressman Joseph P. Kennedy, II, U.S. Representative in Congress from
the 8th District of the State of Massachusetts ..............................................
Deval L. Patrick, Assistant Attorney General for Civil Rights, U.S. Depart
ment of Justice ................................ .............................................................
Prepared statement .......... .........................................................................
Discrimination in insurance is prohibited by the Fair Housing Act ...
Enforcement efforts ..............................................................................
Roberta Achtenberg, Assistant Secretary for Fair Housing and Equal Oppor
tunity, U.S. Department of Housing and Uihan Development ....................
Prepared statement ...................................................................................
Insurance and the Fair Housing Act .................................................
Risk, race, and insurance ..................................................................
Current HUD initiatives ......................................................................
The need for data disclosure ..............................................................
Risk, race, and insurance, revisited ...................................................
Response to written questions of Senator Bond..........................................
William R. Tisdale, president, National Fair Housing Alliance; accompanied
by: Shanna L. Smith and Cathy Cloud ........................................................
Prepared statement ..................................................................................
Introduction ........................................................................................
I. NFHA’s homeowners* insurance testing project ............................. .
II. NFHA’s testing process .................................................................
HI. Target company and agent selection .............................................
IV. Testing results ..............................................................................
V. Insurance companies, in general....................................................
VI. The Fair Housing Act prohibits insurance discrimination ..........
VII. Why are so few complaints filed? ................................................
VIII. Recommendations .....................................................................
Conclusion ...................................................................................... .
J. Robert Hunter, commissioner, Texas Department of Insurance .................
Prepared statement ...................................................................................
Definitions...........................................................................................
Data collection ....................................................................................
Evidence of redlining—the Texas experience .....................................
Underwriting guidelines .....................................................................
Reporting by census tract vs. reporting by zip code ..........................
Collecting race and gender data .........................................................
Reporting of claims information .........................................................
Automatic sunset provision ................................................................
Review of S. 1917 ................................................................................
Conclusion ..........................................................................................
(III)
3
5
13
46
46
47
17
49
49
50
51
53
54
156
29
54
55
55
56
56
57
57
59
60
61
62
33
63
63
64
64
65
66
66
66
67
67
67
IV
Page
J. Robert Hunter, commissioner, Texas Department of Insurance—Continued
Resoonse to written questions of Senator Riegle.................................... .
Lynn M. Schubert, assistant general counsel, American Insurance Associa
tion .................................................................................................................
Prepared statement ....................................................................................
I. Introduction......................................................................................
II. Definition of redlining.....................................................................
HI. The first step toward a solution ....................................................
IV. Is insurance available? ..................................................................
V. Potential Federal action..................................................................
VI. Affirmative efforts by the industiy.................................................
VII. Other issues ..................................................................................
VHI. Conclusion...................................................................................
Charles Kamasaki, National Council of La Raza...............................................
Prepared statement of Raul Yzaguirre .......................................................
I. Introduction.......................................................................................
II. Impact of discrimination on the Hispanic community ....................
HI. Effects of housing discrimination ..................................................
IV. Home ownership opportunities denied ..........................................
V..Recommendations ............................................................................
Letter dated May 9, 1994, to Senator Riegle ..............................................
Additional Material Supplied
for the
37
67
67
68
69
69
72
73
76
77
41
77
77
78
81
82
83
85
Record
An analysis of zip code distribution of State Farm and Allstate agents and
policies in Chicago..........................................................................................
State of Missouri, Director of Insurance, Jay Angoff ........................................
Prepared statement .....................................................................................
Homeowners* Insurance in Missouri..........................................................
The NAACP Legal Defense and Educational Fund, Inc.....................................
Nationwide Insurance Companies.................................................................... .
Health News Daily, May 11, 1994, FDC Reports ..............................................
Patton, Boggs & Blow, LLP, letter dated May 13, 1994, from John L.
Oberdorfer......................................................................................................
State of Minnesota, Department of Commerce .................................................
158
88
116
123
123
131
133
135
141
145
HOMEOWNERS’ INSURANCE DISCRIMINATION
WEDNESDAY, MAY 11, 1994
C o m m it t e e
on
B a n k in g , H o u s in g ,
U .S. S e n a t e ,
U r b a n A f f a ir s ,
and
Washington, DC.
The Committee met at 10:05 a.m., in room SD-538 of the Dirksen Senate Office Building, Senator Donald W. Riegle, Jr. (Chair
man of the Committee) presiding.
OPENING STATEMENT OF CHAIRMAN DONALD W. RIEGLE, JR.
The C h a ir m a n . The Committee will come to order.
Let me welcome all those in attendance this morning and invite
those standing to find seats. We’re pleased to have everyone here
today. We’re anticipating two of our colleagues from the House of
Representatives coming to make statements this morning. We’ll ac
commodate them—I see Congresswoman Cardiss Collins arriving.
I’d like to invite her to come on up to the table and be seated. I
know our colleague, Joe Kennedy, is also coming over this morning
and will be here shortly.
I’m going to make a brief opening statement and call on Senator
Feingold, who has asked for the opportunity to make a presen
tation to the Committee. We’re delighted that he is here and sitting
in with us, and I’ve invited him to sit up here with the Members
this morning.
After we nave gotten started with the background information,
I’m going to call on our colleague from the House, Congresswoman
Collins, who I’m delighted is here today, and I’ll say more about
that just shortly.
This morning, the Committee will consider whether homeowners’
insurance is as available, affordable, and accessible to all Ameri
cans as it should be, and most particularly those living in urban
neighborhoods.
We’re going to consider not only if insurance companies are re
fusing to sell in certain areas, but if they are also charging more
for insurance coverage or offering restricted insurance coverage
without justification.
This Committee has considered claims of what is commonly re
ferred to as redlining, screening people out for credit or other kinds
of financial services, and we’ve done that on several occasions.
In fact, we have included provisions in virtually every bill that
the Committee has passed during my 6 years as Chairman to en
sure as a matter of law and legal right that credit is available to
all communities.
(l)
2
The Committee included provisions in the Financial Institutions
Reform, Recovery, and Enforcement Act, known as FIRREA, and
amended the Home Mortgage Disclosure Act to address the prob
lems of mortgage discrimination. We amended the Equal Credit
Opportunity Act to require regulators to notify the Justice Depart
ment about instances of lending discrimination.
We have also held hearings to make sure that fair lending laws
are being enforced aggressively and effectively. Since lenders re
quire their borrowers to secure property insurance, eliminating
homeowners’ insurance discrimination is a logical and necessary
progression of the Committee’s efforts to ensure that there is an
adequate and fair flow of capital into distressed communities and
that there, i;n fact, be fair access to financial services by all citi
zens.
In 1968, the Fair Housing Act banned redlining and housing dis
crimination. The Federal courts have also made it clear that the
Fair Housing Act bans homeowners’ insurance discrimination. Nev
ertheless, we still find many serious and terribly troubling exam
ples of discrimination fully 26 years after Lyndon Johnson signed
that piece of legislation, and I remember it well because I was
there as a new Member of Congress at the White House at that
bill-signing.
One insured drew a line around the entire city of St. Louis, Mis
souri, and labeled it “ineligible property.”
In Milwaukee—Senator Kennedy? Congressman Kennedy, why
don’t you iust come right on up and take a seat beside Congress
woman Collins?
In Milwaukee, a district sales manager was taped while criticiz
ing other agents for writing too many policies for blacks and sug
gesting ways to avoid writing policies for black Americans.
This Committee has also heard testimony about the tremendous
shortage of property insurance in Los Angeles. For example, Cali
fornia’s Department of Insurance has reported that 61 percent of
the businesses damaged in the riots after the Rodney King verdict
were uninsured because coverage was too expensive or just not
available. An additional 4 percent said the agent they contacted
would not quote rates in their area.
In Georgia, and here in Washington, DC, insurance regulators
are investigating charges of insurance discrimination.
The State of Texas recently fined Allstate $850,000 for discrimi
natory practices. The Ohio insurance department fined Farmers In
surance for determining rates by zip codes instead of by municipal
ity, which led to underpricing insurance in the suburbs.
Today, we will hear from a panel of experts in this field, includ
ing Texas insurance commissioner, Robert Hunter; Lynn Schubert,
of the American Insurance Association; Wayne Tisdale, president of
the National Fair Housing Alliance, which I understand has some
discouraging, but nevertheless, important results to share with us
from a tester program; and Raúl Yzaguirre, of the National Council
of La Raza.
They’ll be followed by Assistant Attorney General, Deval Patrick,
of the Justice Department and Assistant Secretary, Roberta
Achtenberg, of HUD, who will testify about the Administration’s
enforcement efforts in this area.
3
I want to say that this is an abbreviated opening statement with
respect to the work of this Committee and the counterpart Commit
tee in the House, upon which our colleague serves.
Putting an end to discriminatory financial practices in this coun
try, in the banking system, in the insurance system, and in every
other way, are part of fulfilling the basic doctrine of this land.
The laws that have been written and signed into effect prohibit
many of the practices that are going on today and have been going
on for decades. We’re determined to use every power at our com
mand to bring that to an end.
It doesn’t mean anything to talk about using our normal capital
markets and free enterprise system to get resources into the entire
fabric of our land and into our distressed communities if there are
barriers that prevent that capital from flowing there.
If people can’t get mortgages or they can’t get home insurance
which enables them to get mortgages, the whole system starts to
just malfunction and die right on the spot. We can’t have that.
That’s not what America is about and the law says it shouldn’t be
that way and we’re not going to tolerate those practices.
If we need stronger penalties with respect to enforcement and re
medial action, then those are precisely, the steps that we’ll need to
take. This practice has to stop if we want to have any hope of hav
ing the kind of America that we speak about and hold out as a vi
sion and also that can work fairly and properly on a day-to-day
basis.
Senator Feingold, if I may, I think I’m going to call on our House
colleagues who have come across to let them make their opening
statements. After they have done so, I’m going to call on you as a
witness, in effect, before the Committee, and then other Members
of the Committee for opening statements.
Let me welcome my two colleagues.
I want to just say, Congresswoman Collins, you and I served to
gether a good many years ago as seatmates on the Foreign Rela
tions Committee in the House, the International Relations Commit
tee. I remember that opportunity with special fondness. I’m a great
admirer of you and your record. I’m just delighted that you’re here
to make a statement to the Committee this morning.
OPENING STATEMENT OF CONGRESSWOMAN CARDISS COL
LINS, U.S. REPRESENTATIVE IN CONGRESS FROM THE 7TH
DISTRICT OF THE STATE OF ILLINOIS
Representative C o l l in s . Thank you very much. I remember
those days as well and they’re certainly very fond memories for me,
equally.
Mr. Chairman, and Members of the Committee, I certainly ap
preciate this opportunity to testify today on the very important
issue of insurance redlining.
Over the last year, the Energy and Commerce Subcommittee on
Commerce, Consumer Protection, and Competitiveness has exam
ined redlining practices of insurance companies. At the Subcommit
tee’s two hearings, we heard very disturbing reports about a vari
ety of practices insurance companies use to deny access to insur
ance to the residents of the urban areas.
4
Some may say that insurance redlining is a thing of the past, but
the witnesses at our hearings testified that the practice continues,
in fact, flourishes. Some may say that redlining doesn’t exist or
they’re not sure it’s a problem.
I know it’s a problem. My constituents know it’s a problem. Most
people living in urban areas know it’s a problem. For example,
those at the hearing cannot forget Selwyn Whitehead of the Eco
nomic Empowerment Foundation, who testified about her experi
ence in trying to get liability insurance for her telecommunications
consulting firm in the late 1980’s.
When she identified her firm as a woman-owned firm, of color,
in Oakland, California, she was turned away or quoted premiums
of $8,000 to $10,000 per year. When she called on behalf of her fic
titious white male boss, a Mr. Selwyn Whitehead, the first quote
was for a mere $1,200.
The statistics speak for themselves. Illinois Public Action testi
fied that there are 52 State Farm offices and 32 Allstate offices in
a predominantly white congressional district in Chicago. But, in
the Chicago portion of my district, according to Public Action, there
are only 6 State Farm offices and 2 Allstate offices outside the
downtown area.
I’d like to submit for the record the detailed study prepared by
Illinois Public Action called, “An Analysis of Zip Code Distribution
of State Farm and Allstate Agents and Policies in Chicago.”
ACORN testified that in Chicago, only 51.1 percent of occupied,
single-family units in low-income neighborhoods, and only 57.6 per
cent in minority neighborhoods, were covered by any type of insur
ance, compared to 90-percent coverage in high-income and 87.7 per
cent coverage in white areas.
There is plenty of other evidence of redlining behavior by insur
ance companies themselves. For example, the NAACP has a law
suit pending against American Family Mutual Insurance Company.
That’s the case where the sales manager was recorded as telling an
agent:
I think you write too many blacks. You gotta sell good, solid, premium-paying
white people.
The California Insurance Department recently entered into a
$500,000 settlement with the California Insurance Group after the
company was accused of redlining large areas of San Francisco.
Other witnesses today will provide still more evidence of redlining.
As a practical matter, access to property insurance is a necessity
for mortgage loans and is often essential for access to very small
business loans. Without access to affordable insurance, small busi
nesses in our urban areas cannot prosper, nor generate badly need
ed jobs.
Similarly, access to affordable automobile insurance is often es
sential for residents in the inner cities to keep or to hold their jobs.
My constituents still suffer daily the indignities of insurance red
lining. They want to start seeing some relief.
We, here in Washington, can argue about the perfect bill, but my
constituents want results. We can wait forever for State legislators
to pass the perfect bill, or even any bill, for that matter. But the
people in Chicago and other urban areas want results.
5
We can argue about the perfect bill and let the clock keep ticking
away. Maybe we can even wait until next Congress, or the Con
gress after that, in the hope of the perfect bill. But I feel the envi
ronment for good legislation will not be any better next year and
will likely be worse, if anything. While we argue over the perfect
bill, the people back home don’t have any results.
I would urge this Committee to begin the process of fighting red
lining by supporting appropriate legislation, and yield back the bal
ance of my time, Mr. Chairman.
The C h a ir m a n . Thank you very much. I very much appreciate
your coming over and testifying today. I think you state very pow
erfully what this problem is and the need for us to do something
about fixing it.
Let me say that if anv of you need to leave for reasons of activi
ties on the House side, feel free to do so at any time.
Let me also say to Congressman Joe Kennedy how much I appre
ciate your being here ana your leadership on our counterpart Com
mittee in the House and in this area.
Your family has given great service to this countiy and here in
the Senate, your father and now your uncle are doing so. I have
the particular privilege of occupying the office, here in the Senate
Dirksen Building, that was your father’s office when he was here
as a Senator, and that has a very special meaning to me, as you
know.
We welcome you here. You come very much enveloped in that
family legacy of standing up on these civil rights, human rights,
and equity issues. And so, it’s entirely appropriate that you come
to testify today on this issue where you’ve been giving very impor
tant leadership.
OPENING STATEMENT OF CONGRESSMAN JOSEPH P. KEN
NEDY, II, U.S. REPRESENTATIVE IN CONGRESS FROM THE
8TH DISTRICT OF THE STATE OF MASSACHUSETTS
R epresen tative K e n n e d y . S enator, I v ery m u ch a pp recia te y ou r
w elcom in g rem arks, m ost p articu larly because th ey com e from you.
I, having served on the House Banking Committee now for close
to 8 years, have come to realize how difficult many of the issues
that you have focused on, particularly in your Chairmanship over
here, have been, and how divisive they can be, and how difficult
it is to gain any kind of consensus.
I look around this Committee room and see that there are some
similarities, though we finally did get a Republican to join us on
this issue. But the fact is, Senator, your leadership in taking on a
lot of special interests that surround banking and insurance issues
is something to be very much commended.
I think everybody is going to miss the leadership that you have
shown on this Committee, particularly those of us that struggle
with many of the same issues on the House side.
I want to really thank you and tell you that we’re going to miss
you, but I’m sure you’ll find a way to keep your oar in the water
there, Senator,
The C h a ir m a n . Thank you. Thank you. I assure you, I will.
Representative K e n n e d y . I want to thank the other Members of
the Committee. I also want to pay particular thanks to Senator
6
Feingold for the tremendous work that he’s doing on this issue. The
bill that he has filed here on the Senate side is, I think, one that
can really get to the heart of whether or not these kinds of prac
tices can be documented in a fashion that will allow us to signifi
cantly change the policies of the insurance industries in this coun
try.
All of us, particularly Members of this Committee, are very fa
miliar with how the HMDA data on the banks have enabled us to
get very significant amounts of dollars by the banks to begin flow
ing into inner-city neighborhoods and into neighborhoods of color
around this country.
The light of reality and the light of what is good for this country
will shed itself on this issue if, in fact, we get the data that Senator
Feingold is seeking.
I would urge the Members of this Committee to look closely at
his legislation and I hope that you would support it if it has an op
portunity to come to the Senate Floor and in this Committee on the
way to the Floor.
I want to thank, obviously, all the others for your longstanding
commitment to ending inequality in this country, Senator Kerry,
Senator Campbell, ana Senator D’Amato.
It’s fitting that you have decided to examine the serious problem
of insurance redlining in the property and casualty area. If any
issue deserves your principled leadership, it’s this one.
People are being denied reasonable insurance for reasons having
nothing to do with risk factors, but having everything to do with
their race, their wealth, and tneir neighborhood. We nave in our
country a system of insurance that is separate and unequal. It is
wrong. It’s un-American. It must stop.
If you’re an African-American, a Latino, an Asian, or a gay, then
Nationwide is not on your side. You’re in slippeiy hands with All
state. And, unlike a good neighbor, State Farm is not there. Let’s
just look at some of the facts.
The California Insurance Group gave maps to agents which cov
ered in yellow ink the Africán-American, Hispanic, and gay neigh
borhoods of San Francisco. The company deemed those areas off
limits for the purposes of writing policies.
California’s insurance commissioner, John Garamendi, sued the
company for unlawful discrimination. Ultimately, he reached a
$500,000 settlement and won a commitment from the company to
increase its business in minority and gay communities by $3 to $4
million over the next 4 years.
As you will hear about later this morning, the National Fair
Housing Alliance is about to press charges against Allstate and Na
tionwide for discrimination in four major cities. The findings used
in these cases established dramatic evidence of discrimination.
In Chicago, for instance, testers found evidence of discrimination
as cases were examined.
In Atlanta, where the eyes of the world will be focused in 2 years
from now during the Olympics, only 2 percent of Nationwide’s 155
offices are located in or near a minority area.
In Wisconsin, where the Committee held a hearing earlier this
year, the NAACP has recently filed suit against the American Fam
ily Insurance Company, the State’s largest underwriter of home
7
owners’ insurance, for redlining minority areas of Milwaukee. One
of the company’s sales managers was caught on tape making the
following statement, and I quote:
Very honestly, I think you write to too many blacks. YouVe got to sell to good
white people. Very honestly, black people will buy anything that
{>remium-paying
ooks good right now. But when it comes to pay for it next time, you’re not going
to get your money out of them. The only way you’re going to correct your perform
ance is to get away from the blacks.
The agent who was the focus of those comments has been subse
quently fired.
Mr. Chairman, I’m sure you’ve heard insurers tell you, as I have,
that urban homeowners are adequately served by State-sanctioned
F.A.I.R. plans. In limited instances, that may be so. But F.A.I.R.
plans are far too often poor substitutes for private insurance. In
stead of pooling truly high-risk homeowners, they have become
dumping grounds for everyone living in the city, including the peo
ple who are good risks—the people who live in solid neighborhoods
who take care of their homes. F.A.I.R. plans require people to pay
more in premiums for less coverage.
A recent study by the community group, ACORN, shows that
urban and minority consumers are paying as much as 270 percent
more for F.A.I.R. plans than they pay for the equivalent in private
insurance coverage. F.A.I.R. plans aren’t fair. They are simply a
rip-off.
You’ve probably also heard insurers say that they can’t insure
urban residents because losses are too high. Again, the evidence
suggests that this is simply a flimsy excuse. Missouri’s insurance
commissioner analyzed 12 years’ worth of data collected on white
and black areas of St. Louis and Kansas City. The numbers show
that residents of low-income black areas pay more for homeowners’
insurance and get less coverage than whites of the same income,
even though their losses are less than those of whites.
All in all, Mr. Chairman, the record suggests a nationwide pat
tern of discrimination by insurers with respect to urban and minor
ity consumers. Although HUD and the Justice Department have
begun to seriously examine the practices of the insurance industry,
more needs to be done.
As the GAO recently reported to you, we need more information
about who is getting insurance and not and why. I believe that we
need information to be as detailed as possible, including census
tract, race, gender, and loss data. Such information will help us
shine a light on an industry practice and expose the truth about
redlining once and for all.
Those who have nothing to hide will have nothing to fear. Others
who receive much needed prodding will get that to correct the inde
fensible practices that they pursue.
Some of us may differ about how to best address the issue of in
surance redlining, but I believe that most of us agree that there is
a serious problem and we must get working to resolve it.
The Home Mortgage Disclosure Act and other laws have helped
us eliminate discrimination in the banking industiy. Now it’s time
to expand the fight for fairness into the insurance industry as well.
In closing, let me thank you again, Mr. Chairman, for having me
here this morning. I’d be happy to answer any questions that you
might have, but Yd just urge all of the Members of the Committee
8
to please try to move on this issue. It’s wrong. I think we can do
something about helping people that are denied access to an impor
tant industry in this country.
Than you very much, Mr. Chairman.
The C h a ir m a n . Let me just say, in response to your specific sug
gestion, Senator Feingold has prepared a bill which he’s going to
discuss here momentarily. I intend to work with him to either join
an effort with his bill, perhaps with some modifications, or produce
a bill of our own within this Committee, which might then be com
bined with his bill, but we’re going to move ahead on this.
Representative K e n n e d y . Teriffic.
The C h a ir m a n . Let me iust say two other things of a personal
sort, if I may. I’ll be brief about it.
I remember back in 1967, when your mother, Ethel Kennedy,
came to my office. I was a young freshman Member of Congress,
a Republican at that time, and she was there with some other ac
tivists because she was distressed that there were not enough pub
lic swimming facilities in the District of Columbia for the children
and the young people here in the District at that time.
We’ve had patterns of deprivation of that sort here and other
places for a long, long time. She felt so strongly about it that she
came to lobby on that effort.
I happened to be assigned, Congresswoman Collins, at that time
to the D.C. Appropriations Subcommittee, which was not exactly
seen as one of the choice assignments in the House, but it was an
interesting place. As a matter of fact, it gave me the chance to get
involved in that issue and we were able to do something about it,
and a lot of progress was made.
That family commitment was not just on your father’s side. I can
assure you, from first-hand testimony, it was your mother as well.
Representative K e n n e d y . Thank you.
The C h a ir m a n . I w an t to say som eth in g else because w e’re in a
p eriod
people
com es
seem s
o f tim e w hen th ere’s a lot o f h ostile sh arpsh ootin g tow ard
in p ublic office and, u n fortu n ately, as p a rt o f the m ed ia b e
an en tertain m en t business as m u ch as th e n ew s bu sin ess, it
like th a t m u shroom s.
I was struck, as you were speaking, by just the family history
here. Your Uncle Joe, losing his life in uniform to this country,
your Uncle Jack Kennedy to an assassin’s bullet, and your father
to an assassin’s bullet. The fact that you’ve picked up the flag and
carried it ahead on behalf of your family commitment to public
service in such a distinguished way, that your Uncle Ted Kennedy
is leading the effort here in the Senate, as he has for years and
years and years on health care reform—a tremendous example of
a commitment of life and heart and soul to the pressing issues of
our country and to other citizens within our country.
I think if we need to maybe think a bit and restore ourselves
about examples of things that are right and good and decent about
our system and the people who commit themselves to public serv
ice, we don’t really have to look any further than your family or
your appearance here today on this issue.
It’s just something I wanted to say and I appreciate it. I appre
ciate that effort by everyone who—I met your brother Michael the
9
other day, who I know is involved also as a campaign manager this
year, apparently, up in Massachusetts.
Representative K e n n e d y . Hopefully, he won’t be having to work
too hard.
The C h a ir m a n . I don’t know. He looked to be working pretty
hard the day I saw him.
In any event, it’s a very special part of the example, I think, that
is being set in the countiy and we don’t talk enough about it. We
ought to talk more about it.
Representative K e n n e d y . It’s a little embarrassing to talk about
it in that respect. All I can say is thank you very, very much for
those remarks.
I think that the only thing that Teddy might have disagreed with
was when you initially introduced me as “Senator” Kennedy. He
probably would have taken great offense to that.
T h e C h a ir m a n . W ell, m aybe som e day. I m a y h ave been a little
ahead o f m y s e lf on th at one.
[Laughter.]
Senator Feingold, I’ve asked you to sit up here with the Commit
tee today, rather than to sit as a witness, although you’re now
going to speak as a witness on behalf of your bill. We’re delighted
to have you here. With the support and permission of Senator
D’Amato, I’m going to call on you now and then we’ll go to the
statements of other Members at that point.
OPENING STATEMENT OF SENATOR RUSSELL D. FEINGOLD
Senator F e in g o l d . Thank you very much, Mr. Chairman. I
thank the Ranking Member. I thank you for the courtesy of allow
ing me to sit up here, for the chance to testify, and especially for
holding this hearing on the problem of discrimination in the deter
mination of who has access to affordable, high-quality homeowners’
insurance in America.
I’d also like to thank you, as Representative Kennedy did, for
your past leadership and continued efforts through community
banking and reinvestment legislation to expand financial services
to individuals and communities that have been bypassed in the
past by the private market.
As Representative Kennedy said, on these and other issues, your
leadership will be sorely missed. I’m only sorry that I only had the
chance to work with you, here in the Senate, for 2 years, but I was
well aware, as Chairman of Wisconsin’s State Senate Banking
Committee for 10 years, of your fine work.
I also want to thank both Representative Collins and Represent
ative Kennedy for, of course, taking the lead on this issue. I am
only following in the hard work that they have already done in the
House, hoping to bring the Senate in line in helping with this tre
mendous effort.
Representative Kennedy, it was your willingness to help us focus
on the situation in Milwaukee earlier this year that specifically got
me involved. We were, of course, very troubled by the reports of
Milwaukee being number one, number two, or number three in a
variety of studies showing discrimination in this area.
It’s painful, but the only way to respond is to act. It is because
of your leadership that I got involved, and I thank you for that and
10
thank you for your kind words about the bill, the specifics of which,
of course, you were the inspiration.
I thank you.
During this hearing, the Senate Banking Committee will gather
testimony from a diverse group of individuals and organizations
that will, of course, describe the extent of the lack of access to af
fordable, quality, homeowners’ insurance. This will not come as a
surprise to most people in this room. That’s because we’ve already
had three decades of research, studies, and reports which have
reaffirmed the extent of the problem of insurance redlining.
It’s time that we heed these studies and the three decades of re
search and take concrete steps toward addressing the shameful
practice of insurance redlining, which strikes at the core of the
ability of many Americans to participate fully in our society by
being able to enjoy one very important part of tne American dream,
that of home ownership.
That’s why, upon learning of the efforts of Representative Ken
nedy, Representative Collins, and others, I decided to introduce
S. 1917, the Anti-Redlining and Insurance Disclosure Act of 1994.
This bill would require insurance companies to disclose informa
tion regarding where they write property insurance patterned after
the reporting requirements that are already required of banks and
thrifts under the Home Mortgage Disclosure Act.
Just very briefly, Mr. Chairman, there are three major compo
nents of this bill that make it, I think, meaningful.
First, it is important that any data collection and reporting re
quirements on insurance costs and policies be done at the most de
tailed level that is reasonably feasible. Therefore, I think, in this
matter, it is preferable to require reporting by census tract rather
than zip code, since that allows for more detailed detection and
analysis.
The census tract reporting standard is that which we require of
the banking industry under the Home Mortgage Disclosure Act,
and I think it should be applied to the insurance industry as well.
Second, Mr. Chairman, the collection of data on insurance losses
and claims should also be included in any insurance disclosure ini
tiative. That data would be essential for any proper analysis nec
essary to resolve disputes that arise involving claims that there are
disparities in the price of insurance between different neighbor
hoods or groups of people that are solely based on loss experience
and the associated risk involved, rather than suggesting, as we
must in some situations, that it has to do with prejudice.
Third, and finally, Mr. Chairman, since the collection and disclo
sure of such data will provide affected individuals and Federal and
State regulators valuable information necessary to enforce our Na
tion’s anti-discrimination laws, it is important that it be made
available to the greatest number of communities and individuals as
possible.
This bill, S. 1917, would require that data be collected in 150
metropolitan statistical areas.
Mr. Chairman, we have to place all people of all races and ethnic
backgrounds on a level playing field when it comes to the oppor
tunity to purchase insurance. It’s difficult enough these days to be
11
able to afford to buy a home and keep up the payments. It’s almost
impossible to purchase one without homeowners’ insurance.
Expanding home ownership is critical to any effort at urban revi
talization and we have to remove all barriers, such as insurance
redlining, in order to fulfill any such goals.
So, again, I thank you very much, Mr. Chairman, Representative
Kennedy, for this opportunity.
The C h a ir m a n . Thank you, Senator Feingold, and I appreciate
your leadership. We’ll work with you on this.
Let me now call on Senator D Amato for his opening statement.
OPENING STATEMENT OF SENATOR ALFONSE M. D’AMATO
Senator D’A m a t o . Mr. Chairman, I’d like to join you in welcom
ing our distinguished panel of witnesses which includes Represent
ative Collins and Representative Kennedy. I’d also like to welcome
our colleague, Senator Feingold.
Mr. Chairman, you should be commended, because this Commit
tee has spent quite a bit of time and has done quite meaningful
work under your leadership to ensure that the plight of economi
cally deprived individuals and communities is not exacerbated as
a result of the denial of financial services.
Under your leadership, we were able to pass a bill designed to
eliminate the abusive mortgage-lending practices of unscrupulous
loan sharks, whose high-interest mortgage scams are diverting re
sources directed at the elderly and low-income homeowners.
With respect to the insurance industry and homeowners’ insur
ance, the guiding principles must be sound and objective under
writing practices, not the color of a person’s skin. The fact is that
there should be one standard applied for all.
I look forward to working with you, Mr. Chairman, and with
Senator Feingold in producing legislation that will give us the op>ortunity to monitor carefully, without bringing about undue reguations mat will be burdensome and thereby self-defeating, a sys
tem that will bring about accountability so that we can be assured
that there is that one standard criteria applied on eligibility and
on the ability to pay, within a sound rating system, and other con
siderations tnat would give pause to concern.
This is something that cannot be tolerated. Home ownership be
comes a myth if discrimination precludes affordable insurance, be
cause the two are inextricably linked. You cannot buy a home un
less you have this insurance.
{
T h e C h a ir m a n . R ight.
S en ator D ’A m a t o . I f y ou h ave to p u rch a se a h om e and h ave in
su ran ce th at costs you tw ice as m u ch sim ply becau se o f you r origin
or the color o f you r skin, th at is a bsolu tely un acceptable.
I hope that we can work on this issue together. I know we can.
We’ve done it heretofore on other legislative initiatives—we’ve
worked together in a bipartisan fashion to put together legislation,
and I believe that can be done again.
The C h a ir m a n . Thank you very much, Senator D’Amato. I appre
ciate that.
You’ve been a terrific colleague on these legislative issues as
we’ve been able to work together on them. I appreciate your state
ment today and I’m confident that working together, as we’ve done
12
in every other case, we can find a package that will work. We’ll do
it on a bipartisan basis and incorporate Senator Feingold’s thinking
as well, and that will be a very important piece of work.
Senator Kerry, we’d like your statement next.
OPENING STATEMENT OF SENATOR JOHN F. KERRY
Senator K e r r y . Thank you, Mr. Chairman. Joe’s left. I would
have, obviously, ioined you in your appropriate commendation of
his efforts over the years, and his family’s efforts, which I join in.
This is an important hearing and I think our colleague, Senator
Russ Feingold, is to be commended for pursuing this effort.
It’s really astonishing, when you think about it, that this fight
is going on. I was just reading Deval Patrick’s testimony. I have
to go down to the Committee in which he is currently testifying on
the issue, ironically, of the international convention for the elimi
nation of all forms of discrimination which we have in front of the
Foreign Relations Committee today.
I was just thinking about that. That was put before us by Presi
dent Carter. It was signed in 1965. The Bush and Reagan Adminis
trations didn’t believe in it, so it never came back to Congress.
Now, we finally have it back with a few reservations. That’s what
we’re really talking about here today. It’s an attitude that has per
sisted about enforcement and about what we really care about.
I was just reading the regulation that applies to the Fair Hous
ing Act, which simply says that no one can refuse to provide prop
erty or hazard insurance for dwellings or provide such insurance
differently because of race, color, religion, sex, handicap, familial
status, or national origin. It’s their finding that that constitutes a
violation of the Act.
Now that’s pretty straightforward stuff. I know games played by
some people about the application, but if you read the history of
the struggle between the various circuits and companies contesting
this notion, it is at the core of a problem that is searing at this Na
tion’s conscience.
If you go back to 1832 and read DeToquevile, he observed, then,
that the great unresolved issue in America was race. In 1994, the
great unresolved and, to many people’s thinking, worsening issue
remains race.
When companies not only adopt the bad public policy, but the
bad business policy of not doing business in whole sectors of our
country, they are writing out the whole American dream. They’re
just rubbing it off. I think it’s disgusting. I think it is contrary to
the best notions of business practice.
In my hat, as Chairman of the Subcommittee on Urban and Minority-ownea Business Development of the Small Business Com
mittee, I would say that there’s a huge opportunity there for good
businesses. But the underwriting is critical, obviously, and that’s
what you’re looking at today.
I want to join with you, Senator D’Amato, Senator Feingold, and
others in saying, this Committee should resolve the question of this
struggle between the jurisdictions of our courts and resolve it clear
ly that the Fair Housing Act applies, and set up a fair regimen.
If you look at other countries on this planet, they are creating
banks, whole banks whose lending practices only among the poor
13
have a better rate of return than the Texas banks did to a lot of
rich people in this country.
I hope we’re going to seize this front and center and do the job
that Senator Feingold, Congressman Kennedy, and Congress
woman Collins are asking us to do, and I’m glad you’re pushing for
this and I want to join you.
The C h a ir m a n . Thank you very much, Senator Kerry.
Let me now invite Roberta Achtenberg up to the witness table,
and also, Deval Patrick, who has another appearance this morning
on the Senate side. We’re delighted to have you both here.
Senator K e r r y . You’re already here. I thought you were down
stairs. I’ll be back and forth.
The C h a ir m a n . Assistant Attorney General Patrick, I know you
went through your confirmation process recently. Have you had oc
casions to make other formal presentations to the Senate?
Mr. P a t r ic k . This is one of my first, Mr. Chairman. I had an ap
pearance about a half-hour ago in the Senate Foreign Relations
Committee and I had to go defend someone else’s budget before the
Appropriations Committee about 2 weeks ago.
The C h a ir m a n . I see. So you’ve had a couple of warm-ups for the
main event here today.
Mr. P a t r ic k . A couple of warm-ups, that’s right.
The C h a ir m a n . All right. Very good. Why don’t we hear from you
now and then we’ll go to Ms. Achtenberg.
Senator K e r r y . He’s already prepared to show you his scars.
[Laughter.]
M r. P a t r ic k . I hope I don’t gain a ny n ew ones today.
[Laughter.]
OPENING STATEMENT OF DEVAL L. PATRICK, ASSISTANT AT
TORNEY GENERAL FOR CIVIL RIGHTS, U.S. DEPARTMENT OF
JUSTICE
Mr, P a t r ic k . Mr. Chairman, Senator D’Amato, and other Mem
bers of the Committee, I very much appreciate the opportunity to
appear before you today to talk about this very important problem
of discrimination on the basis of race in the provision of home
owners’ insurance.
We prepared a formal statement which has been presented, I be
lieve, and I’d ask that that be made a part of the record.
The C h a ir m a n . Without objection, it is so ordered.
M r, P a t r ic k . I w ill ju s t sum m arize n ow , briefly, and then give
y o u som e opportunities for q u estion s, i f y o u h av e any.
This problem, through the efforts of this Committee and others,
is gaining increasing attention, as you know. I look forward to
using the full extent of my authority as Assistant Attorney General
for the Civil Rights Division at the Department of Justice to elimi
nate this form of discrimination as all others.
One lesson we’ve learned, from more than 25 years of fair hous
ing enforcement, is that the actions of many players, can have an
impact on the ability to obtain and enjoy housing.
The Civil Rights Division has often challenged direct providers of
housing by taking action against unlawful discrimination in the
sale and rental markets. Recently, we’ve addressed unlawful dis
crimination in the mortgage lending industry and have faced up to
14
the devastating impact that such discrimination can have on indi
viduals and communities.
Discrimination in the homeowners’ insurance industry has an
equally devastating impact and that’s important to see. As the
United States Court of Appeals for the 7th Circuit succinctly noted:
No insurance, no loan, no loan, no house. I believe that’s the point
you were making, Senator D’Amato, and it is as simple as that.
In spite of this obvious practical, real-life link, our efforts to ad
dress discrimination in homeowners’ insurance have been slowed
by claims that such discrimination is not sufficiently related to
housing to evoke coverage under the Fair Housing Act.
The Department of Justice and HUD have consistently, through
both Democratic and Republican Administrations, argued that dis
crimination by insurers of housing violates the Fair Housing Act.
Although we argued successfully in court in 1978 that insurance
redlining is prohibited by the Fair Housing Act, the Court of Ap
peals for the 4th Circuit reached a contrary holding in 1984, which
gave comfort to those resisting enforcement.
This decision, however, did not weaken the resolve of the enforce
ment agencies. On January 23, 1989, HUD published regulations
implementing the Fair Housing Act, stating that, and I’m quoting:
Refusing to provide property or hazard insurance for dwellings or providing such
insurance differently because of race, color, religion, sex, handicap, familial status,
or national origin, constitutes a violation of the Act.
The road to substantive enforcement, however, has not been
smooth. It rarely is in the area of civil rights. The Civil Rights Di
vision received allegations in 1988 that American Family Insur
ance, a Wisconsin insurer, had directed its agents not to sell prop
erty insurance in areas where African-Americans constituted the
majority of the population.
An investigation was initiated, but the company declined to co
operate, arguing that insurance discrimination did not fall within
the Fair Housing Act.
Our investigation was delayed for 4 years by litigation of this
issue, which was eventually resolved in our favor by the Court of
Appeals for the 7th Circuit, as I referred to a moment ago.
Nationwide Insurance Company also claimed the Fair Housing
Act did not reach racial discrimination in homeowners’ insurance
when it refused to comply with a HUD request for information nec
essary to investigate a complaint of discrimination in the provision
of homeowners’ insurance.
The company sued the Secretary of HUD over the issue and in
February of this year, after more than 2 years of litigation, an
order was entered upholding our position that the Fair Housing Act
prohibits racial discrimination in the provision of homeowners’ in
surance.
We have recently been successful elsewhere in litigating this
issue. In practical terms, this is really no longer an issue.
We are pleased that progress has been made in clarifying the
meaning of the Act. The HUD regulations are particularly signifi
cant, and I’ll defer to my colleague on that in a moment, since they
are entitled to substantial deference by the courts.
15
The 7th Circuit concluded that the regulations were valid and
controlling and we believe that sound reasoning should and will be
followed by other courts that are presented with this issue.
Our initial lesson from investigating claims of insurance dis
crimination is that such investigations can be exceedingly complex.
The investigations are sometimes compared to lending discrimina
tion investigations, but several differences make insurance dis
crimination investigations even more subtle.
The Home Mortgage Disclosure Act, HMDA, as it’s sometimes re
ferred to, provides us useful information about the lending prac
tices of depository financial institutions. We know the number of
minorities who apply for mortgage financing, the number accepted,
and the number rejected, and this information is very helpful in
surfacing suspicious behavior and practices.
We do not have comparable information, yet, about the perform
ance of property insurance companies. Also, while a depository in
stitution may receive several thousand loan applications in a given
year, a property insurance company may write or renew several
hundred thousand policies in the same time period. Unlike lenders,
property insurance companies, to the best of our knowledge, do not
maintain records regarding the rejected applications, nor do they
record the race of applicants or policyholders.
For these reasons, it is more difficult to assess claims that a par
ticular company is discriminating. Discrimination can occur with
out an application or a record if an agent, on account of race, de
clines to return a phone call, fails to keep an appointment, or sim
ply tells a prospective customer that the company will not insure
the property because of its condition or location. Given the state of
available records, this type of discrimination might best be detected
by traditional fair housing testing.
Possible discrimination at the underwriting stage must also be
addressed. Our own investigations have revealed that policy-holders in minority neighborhoods might receive insurance only after
being subjected to unorthodox inspections or being required to
make repairs that are not required in other neighborhoods of other
property owners.
We have found that dwellings in minority neighborhoods often
receive inferior insurance coverage that only allows for the repair,
rather than the replacement, of a dwelling in the event of a prop
erty claim or total loss, or imposes a low dollar limit that prevents
its replacement.
We ve also found that some companies charge a higher premium
per dollar of insurance coverage for the inferior repair cost policies
than they do for the more desirable replacement cost policy. Sen
ator Kerry’s point—bad business as well as race discrimination.
Finally, insurance companies can inflict substantial damage on
neighborhoods by refusing, because of the racial or national origin
identity of the neighborhood, to market their products. It’s often re
vealing to examine the number of insurance offices located in mi
nority neighborhoods, as compared to the number located in identifiably white areas.
The marketing of insurance products relies heavily on a commu
nity approach with agents located in close proximity to the areas
they intend to serve. If a company desires to redline an area as off-
16
limits for company business, it can do so without drawing a red or,
in some cases, a yellow line on a map. It can simply decline to open
offices in the area or otherwise to market in that area.
The effect is as predictable as that caused by lending discrimina
tion and by lending redlining. That is, that minority neighborhoods
are abandoned by mainstream economies without the tools avail
able to the middle class to maintain themselves.
Improved information would promote compliance with and en
forcement of the Fair Housing Act. It would be helpful to have in
formation comparable to that which we have about banks.
HMDA, which requires reporting at the census tract level, as you
know, has proven very useful, indeed, to us. We rely heavily on
that information whicn allows us to identify the racial or ethnic
identity of a select area. Because of the wealth of other information
available by census tract, reporting at this level also permits us to
control for other legitimate, nonracial factors, so that we can fairly
evaluate whether race or risk is at the heart of a company’s deci
sion.
We also would find it useful to have information on the types of
policies issued by race and national origin in each census tract.
This information would allow us to evaluate, more readily, whether
minority neighborhoods are offered inferior insurance protection.
Information on claims paid and loss data would also be useful, par
ticularly because this will allow us to test one of the excuses insur
ance companies sometimes offer for choosing not to insure minority
homeowners in minority neighborhoods.
Improved reporting would not only aid our law enforcement ef
forts, equally importantly, it should also lead to improved perform
ance and voluntary compliance within the industry.
We have spent considerable time working with representatives of
the lending industry to obtain voluntary compliance with civil
rights laws and we have encouraged self-assessment and voluntary
corrective action.
This effort has been very, very successful, and many lenders are
now using HMDA data for their own benefit. Self-assessments are
being performed, business practices are being changed, and many
lenders seem to be doing a better job of serving minority neighbor
hoods. That doesn’t mean our work is done, but it means that we
are working in concert with the industry to address this problem.
To the best of our knowledge, most insurance companies do not
maintain information in a form today that readily lends itself to an
evaluation of their performance in minority neighborhoods.
We intend to devote considerable resources to uncovering and
remedying unlawful discrimination by property insurance compa
nies. I urge insurance companies to recognize the importance of
this issue to individuals, to neighborhoods, and to the Nation, and
I urge them to evaluate their own performance in compliance with
the law and to take voluntary corrective action.
Companies should be on notice that if they fail to address the
issue voluntarily, the Department of Justice will use its enforce
ment powers to compel a remedy. I therefore urge this Committee,
Mr. Chairman, to give us the tools to do this efficiently and fairly.
Thank you very much.
The C h a ir m a n . Thank you very much. Very helpful testimony.
17
Ms. Achtenberg, we’re delighted to have you back before the
Committee and we appreciate all the important work you’re doing
at HUD. We’d like to hear from you now.
OPENING STATEMENT OF ROBERTA ACHTENBERG, ASSISTANT
SECRETARY FOR FAIR HOUSING AND EQUAL OPPORTUNITY,
U.S. DEPARTMENT OF HOUSING AND URBAN DEVELOPMENT
Ms. A c h t e n b e r g . Thank you, Mr. Chairman, Senator D’Amato.
I’d be pleased to summarize the statement that I submitted for
the record, with your permission.
I’d like to thank you and the Committee for the opportunity to
discuss with all of you one of our Nation’s most pressing civil rights
and urban development issues.
As the Nation’s chief fair housing law enforcement agency, HUD
will vigorously enforce the Fair Housing Act as it applies to dis
criminatory insurance policies and practices, just as we do in the
areas of real estate sales, mortgage lending, and other practices re
lated to housing. But HUD’s interest in this issue extends beyond
its responsibility to fully enforce the Fair Housing Act.
Property insurance is essential for the revitalization of this Na
tion’s cities. Insurance is required to purchase or improve a home
or to start or expand a business. As the President’s National Advi
sory Panel on Insurance in Riot-Affected Areas concluded some 25
years ago, “Communities without insurance are communities with
out hope.”
As was succinctly stated by my colleague, the Attorney General,
the Fair Housing Act prohibits insurance redlining and other poli
cies and practices that deny insurance or make it unavailable on
the basis of race or any other protected status under the Act. The
Fair Housing Act also prohibits discrimination in the terms, condi
tions, costs, or other aspects of insurance coverage.
HUD and the Department of Justice have taken this position
since the issue first arose in 1978. In regulations implementing the
Fair Housing Amendments Act of 1988, as was referred to by the
Attorney General, HUD expressly prohibited refusing to provide
property or hazard insurance or providing such insurance dif
ferently because of race, color, religion, sex, handicap, familial sta
tus, or national origin.
In his recent Executive Order No. 12892 entitled, Leadership and
Coordination of Fair Housing and Federal Programs, Affirmatively
Furthering Fair Housing, President Clinton made explicit reference
to HUD’s jurisdiction when he called for the agency to promulgate
regulations describing the conduct prohibited by property insurers
under the Fair Housing Act.
The statement of Attorney General Patrick described in more de
tail the considerations the courts have given to this question. In
short, those courts considering the issue nave concluded that insur
ance' is covered by the Act. While the 4th Circuit held to the con
trary in the Mackey case, so referred to, in 1992, the 7th Circuit
in American Family, as quoted earlier, after HUD’s promulgation
of regulations asserting the coverage of insurance under the Act in
1988, found that the reasoning of the Mackey case no longer was
persuasive.
18
The court stated that events had bypassed Mackey noting that
HUD regulations were controlling on this issue because they were
based on HUD’s authority to act and they were according due def
erence to the regulatory exercise by the executive agency.
The key sections of the Fair Housing Act are section 3604(a),
which makes it unlawful to “otherwise make unavailable or deny
a dwelling,” and section 3604(b), prohibiting discrimination, “in the
provision of services or facilities in connection therewith.”
Because property insurance is required to secure a mortgage
loan, which generally is required to purchase a home, denying in
surance makes the home unavailable.
I will not reiterate the quote referred to by the Attorney General,
from the American Family case. That makes it quite clear that if
you are providing insurance on a discriminatory basis, then home
ownership will be so affected.
HUD’s authority in the area of discriminatory insurance policy
and practices is identical to its authority in other areas covered by
the Fair Housing Act. In addition to the investigation of com
plaints, the Secretary of HUD can begin investigations and file
complaints on his own initiative. Also, the Fair Housing Act re
quires the Department of Housing and Urban Development to seek
voluntary compliance, on the part of the industry, so that we can
move to prevent discrimination by encouraging the industry to
adopt policies and practices that bring about fairness in the provi
sion of insurance. We intend to do so.
We are authorized to fund private groups—we, meaning HUD—
to fund private group activity to educate the public and to assist
private enforcement efforts, and we intend to do so.
HUD is the only Federal agency charged with the responsibility
to promulgate regulations that would further define the substance
of Acts, that the Secretary will consider, that constitute reasonable
cause to believe that discrimination has occurred. And we intend
to do so.
The business of insurance, Mr, Chairman, is the business of dis
tinguishing among risks and grouping them in terms of the poten
tial for compensable losses that they pose during the life of a pol
icy. There is little doubt that the vast majority of insurers are at
tempting to make conscientious efforts to carefully consider the
perils that potential insurance risks pose when they market their
products.
However, we must recognize the possibility that unlawful dis
crimination can occur in the insurance industry. In some instances,
evidence of racial discrimination in the insurance industry can be
overt, as was referred to by Congresswoman Cardiss Collins and
which references were reiterated by other members of the panel.
More common, however, are various subtle forms of disparate
treatment. Paired testing has been used to uncover instances
where a caller from a predominantly minority area has been given
very different information than a caller from a white community
with identical financial and other socioeconomic characteristics, ex
cept for the racial composition of the areas of the callers involved.
The minority area resident often must call several times before
reaching an agent. The minority resident may be told that an in
spection will be necessary prior to offering a policy. The minority
19
resident may be offered a policy at a higher price for less coverage.
The minority resident may be referred to a F.A.I.R. plan. The mi
nority resident may never receive a quote in the mail, though a
quote was promised. The minority resident may just not be offered
any service whatsoever.
T h e caller from the w hite n eigh borh ood , on th e oth er h an d , m a y
b e offered a policy on the ph on e d u rin g the first call, m a y be of
fe re d a choice o f policies and p rem iu m s, and m a y b e ea gerly and
p olitely solicited as a client.
Or may not be. That is what we are here today to discuss, and
it is to those issues that the legislation introduced in the House by
Congresswoman Collins, the legislation introduced in the House by
Congressman Kennedy, and the legislation introduced in the Sen
ate by Senator Feingold, would begin to address.
To illustrate the outcome of such potentially discriminatory prac
tices, the Missouri insurance commissioner recently found that
residents of low-income minority areas of St. Louis paid 50 percent
more for similar coverage than did residents of low-income white
areas, even though the loss experience had been higher in the
white communities, a fact that was referred to by the Chairman.
Some industry underwriting practices that may be applied uni
formly and otherwise appear race-neutral, may have a disparate
impact on minority communities. This may occur when insurers,
for example, will not provide coverage or will refuse to provide full
replacement coverage on lower-valued or older homes. Some insur
ers may refuse to insure homes valued at less than $50,000, or
homes that were built before 1950.
In the United States today, 47 percent of black households, but
just 23 percent of white households, live in homes valued at less
than $50,000. Similarly, 40 percent of black households compared
to 29 percent of white households live in homes built before 1950.
The racial effect is clear—although these types of statistics do
not, in themselves, prove a violation of the Fair Housing Act.
Where there is a business necessity for such a practice and no less
discriminatory alternative is available, no violation would exist.
But when practices with such racial impacts are not legally or oth
erwise justified, a case-by-case, Fair Housing Act analysis is war
ranted.
There is evidence that these types of practices are pervasive in
some markets. The Texas Office of Public Insurance Counsel, for
example, conducted a review of the underwriting guidelines widely
used by Texas insurance companies. It found that approximately 90
percent of the market in that State is covered by insurers with re
strictive age and value criteria in their underwriting guidelines.
Secretary Cisneros is acting aggressively to combat insurance
discrimination. We have recently created a separate insurance unit
to address these issues at HUD. Its first task will be the develop
ment of regulations clarifying HUD’s current prohibition against
insurance discrimination.
It has been 25 years since passage of the Fair Housing Act and
HUD has yet to promulgate substantive regulations defining, in
depth, insurance practices that violate the Act and what HUD will
do to identify and redress those violations. It is high time that
HUD took the responsibility for doing so, and we intend to.
20
Several complex issues must be addressed in promulgating such
a regulation. I intend to do so in close collaboration with the insur
ance companies, their trade associations, State regulators, civil
rights groups, and community organizations of all kinds. I have al
ready begun informal discussions with representatives of these in
terests to learn more about their views and about issues that they
would like HUD to address through its rule-making power.
We will be conducting public hearings, around the countiy,
where we will invite industry representatives, insurance regu
lators, advocacy groups, and private citizens, alike, to testify on the
proposed rule’s content.
Based on the comments we receive, the evidence presented at
public hearings, the literature, and the written guidance we receive
from additional communications with the industry and others,
HUD will publish a proposed rule and, after careful consideration
of the comments received, a final rule.
We are considering what must be addressed in the regulation for
it to be effective as guidance to HUD investigators, State and local
civil rights agencies, and private fair housing groups, in order for
it to serve as a guidepost for preventive acts by the industry and
in order for the regulation to be a clear description of the rights
afforded protected classes under the Fair Housing Act.
To do so, the regulation will address specific practices that are
prohibited under the Fair Housing Act, describe the standards to
be utilized in determining whether violations of the Act occur, dis
cuss investigative techniques that will be utilized, discuss remedies
that will be sought where violations are found, and discuss the na
ture and extent of voluntary affirmative efforts that are appro
priate for the industry to undertake in order to eliminate discrimi
nation.
As in other areas of fair housing law enforcement, standards to
determine discrimination will be based on the principles of overt
discrimination, disparate treatment, and disparate impact. The in
vestigative techniques we will utilize will include those that have
grown from our fair housing investigative experience across the
board, but they will most definitely include statistical analysis of
disclosure data (should such disclosure data become available),
paired testing, and content analysis of underwriting manuals and
other documents pertaining to evaluation of risk and marketing
practices—all of the kinds of tactics that we currently utilize, as
does the Justice Department, in lending discrimination investiga
tions today.
A description of appropriate remedies will also be included in the
rule which will help guide self-evaluation and corrective actions by
the industry, even when we, HUD, are not otherwise involved in
the complaint investigation.
Remedies to be considered will be similar to the kinds of relief
that have already been obtained by the Department of Justice in
their lending discrimination settlements.
HUD’s primary enforcement tool has been the investigation and
conciliation of complaints. They include those brought to HUD by
individuals who believe their rights have been violated, and com
plaints brought by fair housing organizations. Additionally, the
21
Secretary, as I stated before, has the authority to initiate an inves
tigation to determine whether a complaint should be issued.
We are dismayed by the continuing efforts of some in the insur
ance industry to resist HUD’s investigation of insurance discrimi
nation cases. With the decision that was referred to by the Attor
ney General in the Nationwide case, we believe that this issue has
been settled.
Whether cases are investigated by HUD itself, by the Depart
ment of Justice, or by State and local agencies with laws substan
tially equivalent to the Federal Fair Housing Act, HUD is commit
ted to conducting full and fair investigations and we would call
upon the insurance industry to cooperate with our investigatory
processes rather than to resist them.
Assistant Attorney General Patrick and I intend to work to
gether, closely, to assure an effective and coordinated approach to
the investigations of claims of insurance discrimination.
As in the case of mortgage lending discrimination, the Secretary
of HUD and the Attorney General will join forces to conduct joint
investigations, where appropriate, combining our distinct powers
and authorities to more effectively address discriminatory behavior.
Through our fair housing initiatives program, we have supported
in the past, and will continue to support, private enforcement ef
forts in this area. We are expanding, significantly, our financial
commitment to these activities, for the coming year, to a level that
constitutes the largest single set-aside for any such activity in the
history of our program.
Now, a word on the disclosure bills that are currently before the
Committee.
As the Attorney General stated, a HMDA-like Federal disclosure
rule for insurance would be an invaluable tool for both enforcement
agencies, HUD, and Justice, in our enforcement efforts and for the
efforts of private enforcement entities, as well as State and local
enforcement entities.
The Administration is pleased that legislation to provide disclo
sure data for the insurance industry has been reported by two
House Committees. The Administration looks forward, as well, to
prompt House passage of legislation and urges favorable consider
ation of legislation by this Committee and the full Senate.
We fully support the principles of insurance disclosure, which in
clude information such as race, gender, and income of all appli
cants and policyholders, number of policies issued on a neighbor
hood basis, and an annual report analyzing this information by the
Federal Government.
Nationwide reporting, obviously, is preferable and reporting of
data that can be easily analyzed and accurately analyzed is pref
erable.
Let me say that HMDA has been extraordinarily helpful to the
enforcement agencies, as has been previously described. Similar in
formation for property insurers would be equally valuable informa
tion for our law enforcement efforts. Particularly when analyzed in
conjunction with United States Census Bureau data, through re
porting, we could identify neighborhoods and population groups
that appear to be underserved, given their income, value of prop
erty, and socioeconomic status, as well as their racial status.
22
While such information, alone, would not confirm or deny the ex
istence of discrimination, it would go a long way in pointing us in
the right direction.
Let me say, Mr. Chairman, Senator D’Amato, in closing, that, as
has been previously stated, but is worthy of reiteration, race has
no place in the insurance market and all of HUD are eager to work
with all of you and our colleagues at the Justice Department to
make sure that fair access to insurance is a reality in our urban
communities and throughout the Nation.
Thank you very much.
The C h a ir m a n . Thank you, Ms. Achtenberg.
Senator D’Amato, why don’t you go first and then Fll follow.
Senator D ’A m a t o . First of all, I want to compliment our col
league, Senator Feingold, for his work in this area.
Mr. Patrick, I want to compliment you. It seems to me that your
approach, as it relates to encouraging the industry to come forth
and to work together cooperatively in developing programs, stand
ards, et cetera, is really the way to go.
Mr. P a t r ic k . Thank you, Senator.
Senator D ’A m a t o . We can really do serious damage if we’re not
careful in the way we bring about the most well-intended rules and
regulations. By the same token, it’s absolutely unacceptable to
deny people the opportunity of insurance, to put unreasonable con
ditions, or to drive the cost up. This can very directly affect their
ability to own a home.
Mr. P a t r ic k . Right.
Senator D’A m a t o . That has absolutely got to be stopped. There
can be no ifs, ands, or buts.
To the Assistant Secretary, I remind you that during a very con
troversial period of time in the confirmation hearings, I supported
you. I voted for you.
I want to ask you something, as it relates to what I hear taking
place at HUD, and I’d like a candid answer.
Is your office attempting, or have you attempted, to develop rules
and regulations as it relates to real estate advertising, in publica
tions such as newspapers? Have we reached the point of, what I
call minutia, that we would actually begin to get into the language,
descriptive language, describing the proximity of a house, such as
within walking distance of transportation, railroad, and that there
may be certain words that a realtor would be precluded from print
ing because someone says that certain words or phrases do not
take into consideration those people who are ambulatory?
I raise this because a number of people have brought this to my
attention and to a number of my colleagues’ attention as well. I
have been told that rules were being circulated, or being explored,
in a proposal that would bar certain words and phrases from adver
tising.
Let me give you another example. I think your aware of my con
cern on this issue. I’ve asked my staff to contact HUD with respect
to this. And if it’s not the case, then I want to hear about it.
“Master bedroom.” I have heard that describing something as a
“master bedroom” may be considered inappropriate. If that’s the
case, and if we have people looking into this kind of thing, it is ri
diculous. We’re trivializing a very important area.
23
Ms. A c h t e n b e r g . Senator.
Senator D’A m a t o . Yes.
M s. A c h t e n b e r g . I w ou ld agree w ith y o u th a t i f HUD w e re a ct
in g in such a w ay, th a t w ou ld b e rid icu lou s. I w a n t to a ssu re y o u
th a t n o d ocu m ent th a t h as ever em a n ated fro m m y office h a s ever
m a d e any k in d o f suggestion rela ted to th e exam ple.
Senator D’A m a t o . So that if a newspaper were to turn down an
advertisement because they are afraid of being sued, as it relates
to the Fair Housing Act, you would be in a position to advise them
that they need not, that you would not take action as it relates to
the advertisement of a master bedroom or a walk-in closet or a
house or development being in close proximity to a church or a
temple, et cetera.
Ms. A c h t e n b e r g . HUD has never taken any such position and
we would not under my Administration.
Senator D’A m a t o . I want you to know, I think you’ve gone a long
way because this statement is important, so that it’s on the record
and so that this area of free speech, in terms of our commercial en
terprises, is permitted to go forth.
About 2 weeks ago I was informed that this is something that
some people have been confronted with. Apparently, some of the
fair housing groups, in conjunction with local newspapers, have de
cided, or the papers decided, that they wouldn’t take certain adver
tisements.
I think that this type of action trivializes the key elements that
we’re talking about today. This gives firepower to people who don’t
want to do anything.
I want to see that people have a right to buy a home, and I want
to see that they have all those things that may seem to be intangi
ble. How many people think about the fact that, unless you can get
that insurance, you’re denied that real opportunity.
I’m very glad, Ms. Achtenberg, that you’ve put that out for the
record.
Ms. A c h t e n b e r g . Thank you, Senator.
Sen ator D’A m a t o . I w a n ted to, b efore I le ft th is h earin g , b e able
to g et th a t out.
Mr. Patrick, good seeing you.
Mr. P a t r ic k . Thank you, Senator.
Senator D ’A m a t o . Thank you, Mr. Chairman.
The C h a ir m a n . Thank you, Senator D’Amato.
Just two questions because I want to go to our next panel. Let
me ask you both this. To what extent is the Federal Government
allowed to attack insurance discrimination without violating the
McCarran-Ferguson Act? And do you agree with recent court rul
ings like the NAACP versus the American Family Insurance Com
pany that McCarran-Ferguson does not apply to subsequently en
acted civil rights statutes such as the Fair Housing Act?
Mr. Patrick, why don’t you start?
Mr. P a t r ic k . I’ll take a stab at that first.
As you probably know, Mr. Chairman, no court has found the
McCarran-Ferguson Act a bar or barrier to the enforcement of the
Fair Housing Act, and that has been the position of the Depart
ment of Justice consistently.
24
We appreciate the issue that is presented, but, again, the weight
of legal authority is against the view that McCarran-Ferguson is
any bar or any obstacle to what we’re trying to accomplish under
the Fair Housing Act.
Ms. A c h t e n b e r g . Mr. Chairman, I would only point out that
that has consistently been our position since 1978, through four
Presidential Administrations, and will continue to be the position
of the Clinton Administration.
T h e C h a ir m a n . S o th a t p resen ts n o b arrier problem s in term s o f
yo u r g ettin g a t th ese issu es.
Ms. A c h t e n b e r g . Not from our point.
M r. P a t r ic k . N ot as w e see it.
The C h a ir m a n . All right. Very good. Now, let me ask you both
to summarize, if you would, what types of data could the Federal
Government collect that would be most useful in identifying and
combating this kind of insurance discrimination?
Mr. P a t r ic k . I went first the last time. Would you like to take
this?
Ms. A c h t e n b e r g . Mr. Chairman, we would prefer nationwide re
porting, which would be the most useful to us. I think the validity
of that stands on its own. We would prefer reporting that will as
sure that we can identify when insurance is denied, as well as
when it is offered, as was stated by my colleague.
It is important that the reporting be done in sufficiently small
geographic entities so that we can accurately assess whether or not
neighborhoods are being restricted for any reason, and so that we
might determine the reasons for those restrictions.
While we can aggregate data up, we can’t aggregate it down. So
smaller is better in this particular regard.
We also need data on acceptance, denials, renewals, and nonre
newals. This information would be exceedingly important in help
ing us determine whether or not people are being granted insur
ance, and helping us assess whether they are being granted or de
nied on the basis of similar terms or conditions.
This is the kind of information we are provided through the
HMDA data and I would expect that similar information would be
as useful to us in this area as it has been in the lending area.
Mr. P a t r ic k . Mr. Chairman, I concur in all of that and add what
I’m sure my colleague intended to add and would agree with, which
is policy types, because we have seen through some of the studies
that we’ve looked at that there are differences in quality of insur
ance offered that appear to be explained, or partly explained, by is
sues of race. I’d also say that we’d be looking for gender reporting
information.
Ms. A c h t e n b e r g . Race.
Mr. P a t r ic k . Race and gender as well.
The C h a ir m a n . This discussion, in the end, will come down, I
think, to two contesting points of view. On the one hand, there will
be people who will argue whether this involves too much Govern
ment interference, too much Government bureaucracy, too much
recordkeeping, too much costs associated with recordkeeping, and
so forth.
25
Is there something inherently distasteful or wrong about the
Government coming into private-sector activities with some kind of
a regulatory oversight ana regulatory regime?
Well hear voices stating those views.
On the other hand, what we’re really talking about is the Amer
ican birthright, the fact that in America, when a person is bom and
is an American citizen, there are certain basic rights that accrue
to that person that are as fundamental as air ana water. That is
the right to, all the guarantees under the Constitution, the Bill of
Rights.
This is really an American birthright issue, as I see it. That is,
does a citizen have the right, under the law, to the kind of equal
treatment that other people in the society receive that is being de
nied to them by the economic or the business system?
To the extent part of their birthright is being taken away, or
they’re being cheated out of it, then that’s when the Government,
in effect, has to step in to make sure that the practices are fair so
that somebody isn’t, in effect, turned into a second-class citizen by
virtue of an imperfection in the way the private sector is working.
This seems to me to be a classic case of that and it’s not some
thing that’s new on the radar screen. We passed laws in the past,
in the distant past, going back decades now, to try to get at this
issue of housing discrimination and to stop the practice of people
being cheated out of their birthright, either based on race, color of
skin, gender, or orientation, that we’ve said violates the basic fabric
of the law of this country. These are both God-given rights, as we
recognize them within our constitutional democracy, ana our laws,
our whole pattern of laws, designed to make sure that everybody
gets the same fair shake.
But what we find is this persistent pattern of people being cheat
ed out of their birthright. We find it in mortgage discrimination,
where somebody may go in. It may be a single mother. It may be
an African-American person. It may be a Hispanic person. And, in
a certain number of cases, they’re treated differently, and in an in
ferior way, and oftentimes cheated out of the opportunity to get
credit to buy a house that is given directly to the next person
through the door, or in line, whose status is a little bit different.
The same thing is happening in the insurance area. In some re
spects, it’s more subtle and it’s a little harder, perhaps, to see quite
as readily, although, if anybody makes any real effort to see it, it
certainly can be seen and identified, to deny people their birthright
based on business practices that cheat them, that cheat them out
of the full American franchise of citizenship.
It’s just outrageous. I mean, we send people to jail for relatively
inconsequential things by comparison. When you think about the
scale of the crime of cheating somebody out of their birthright,
where should that be on the scale of things that we apply sanctions
or punishment to?
That, to me, is a serious offense. That is a major offense, if you
come in and basically cheat somebody out of part of their citizen
ship and put them in an inferior status and, in a sense, lock them
there.
That, to me, is about as un-American as you can get. I read these
FBI reports that come in, that are done on nominees, as the Chair-
26
man of the Committee. That’s one of my responsibilities, to read
these confidential FBI documents. I’m always struck by the things
that the FBI goes out to find out about, whether somebody has any
relatives living in a Communist country, whether there’s some sug
gestive negative inference about that, or whether they might have
used marijuana in college, things of that kind.
There’s a great investigative effort made to determine these
things and identify them in an FBI report, but I’ve never seen in
an FBI report, yet, where it said, so and so was engaged in a busi
ness practice that cheated a large part of the rest of America out
of its birthright.
I’ve never seen that, and I’ve never seen that because they don’t
look at that. In other words, they have a blind eye on that issue.
We don’t measure that because it is not seen as something that’s
an important yardstick as to whether somebody does or doesn’t
reach the level of qualifying for Federal service.
I think what we have to do is reorient our thinking here, because
we’re not going to have a society that holds itself together, where
there are the bonds of common respect and community and affili
ation, if some of the people in this society are cheated out of their
birthright right from the beginning and then the system comes
along with an inequitable pattern of practices that locks in that
second-class citizenship and says, no, you’re not going to get a
mortgage or, no, you’re not going to get mortgage insurance.
The system can’t hold itself together that way. It’s not right. It’s
a violation of everything we say we believe, but it’s also unwork
able as a practical matter. You can’t ask people to give a full affili
ation to the country at the same time part of their citizenship is
being taken away from them. You can’t do that. No nation is going
to be able to hold itself together under that kind of an operating
pattern of conduct.
We have to get rid of that. Fortunately, we have an Administra
tion that is selecting people to serve who believe in correcting these
problems, as both of you do, and, certainly, as President Clinton
does, and is dedicated to the practice of getting it done.
I was so refreshed when the Comptroller of the Currency, Gene
Ludwig, was first named and came in here for his confirmation
hearing. I asked him about racial discrimination and discrimina
tion in lending through the banking system.
He said it was abhorrent to him and that he was going to rip it
out root and branch. That was his phraseology. He said it with feel
ing and conviction and I knew he meant it, and the practice, since,
has shown that.
I don’t know why it’s taken us nearly 300 years to get to the
point where we finally had a Comptroller of the Currency who
would say that to a part of the business system so we can make
sure that the American birthright and franchise is out there for ev
erybody and not just for some.
This is a very important hearing because this gets right to the
issue of whether or not we’re going to have full and equal citizen
ship for people and whether people can have homes.
We talk about the breakdown of the family unit. It’s not easy to
hold families together today with all the other pressures, problems,
27
crime, deprivation, lack of job opportunity, and a lot of other
things.
If a family can get far enough ahead of the game to move out
of an apartment, acquire a house, and work for the goal of main
taining a home and achieving home ownership and to raise a fam
ily within the greater security and well-being of a home that that
family can struggle to acquire, that’s a centerpiece in the whole
American dream. It’s the whole way we think about families devel
oping and having the chance to achieve the things that families
want.
If we somehow have a system that’s working so imperfectly that
we cheat a large part of the country out of the chance to get into
home ownership, at the same time we’re talking about family val
ues, it just doesn’t add up because, in effect, we’re handicapping,
hurting, and damaging the prospects of a large part of our country
from achieving the very things we all know we want.
I think America is changing for the better. We’re waking up to
these things, we’re dealing with these things, we’re forcing
changes, and we’re writing these laws systematically, one by one,
to get the information, to identify the practice, and to stop the
practice, all with the mind of making sure that everybody gets
their full American birthright.
I wish people got it automatically, as they should. We shouldn’t
have to write a lot of laws to make business do what it should do
without any laws. But the reason we need the laws, the practices,
and the data, is that too much of the business system is not willing
to see to it that everybody gets their full American franchise of citi
zenship so the Government has to come in to make sure that we’re
getting fair treatment.
The easiest way to get this done with no bureaucracy and no rec
ordkeeping is for every CEO in the country, that runs one of these
institutions, to say, we’re stopping those practices. We’re going to
offer things on an absolutely fair basis, and I don’t want anybody
turned away who comes in who’s creditworthy and who has a legiti
mate need. I don’t want them turned away based on race. I don’t
want them turned away based on gender. I don’t want them turned
away based on what clock they live on, this block versus three
blocks over in the other direction. I don’t want them turned away
based on their orientation or status.
I want people treated fairly and equitably. If every CEO in this
country would say that, and send that directive down the line, and
say that anybody that didn’t practice that was going to get fired,
and that people who did practice it were going to be rewarded and
would be the people who would advance in the company, we
wouldn’t have to have this hearing. We wouldn’t have to worry
about this problem. We wouldn’t have to worry about getting Gov
ernment in the act to figure out how to make sure it happens, after
300 years, because the private system would be doing it as a mat
ter of course, as a matter of decency and propriety, and because it’s
what America is supposed to be all about.
When I hear somebody say, well, we don’t want Government in
the act, when Government gets into the act for this purpose, to
make sure that every citizen gets their full citizenship, that’s when
28
the Government needs to be in the act. If the private sector will
do the job, then Government can get out of the act.
That would be my preference. But until such time as we see an
end to these practices, then Government has to be in the act be
cause Government is for the people, it’s of the people. It’s the only
real voice and power that the people have when some other force
is taking away part of their citizenship.
It’s the Government’s job to see that everybody gets their full
citizenship back, that it’s intact, it’s solid, and that it works.
So, if you happen to be an African-American, you can say to your
children, we’re on the same footing as everybody else in this coun
try, and you’re not going to have to try to run the race with a 50pound load on your back that’s put there in the form of discrimina
tion because you happen to have a different skin color than the
child that lives down tne street or next door.
We’re not going to have that. That’s what has to end in this
country. If we don’t talk about it, if we don’t focus it, and if we
don’t create a moral imperative about it, then we’re not going to
correct these problems.
I’d like to see more attention paid to it. I’d like to see a little
more investigative journalism, to go out, find these problems, and
put them on the air. Let the public see it, because the public will
not tolerate this. People are inherently fair and they don’t want
this denial of citizenship going on for a large part of our society.
I just know enough about the people of this country, representing
them here for 28 years, to know that most people don’t want that,
and that isn’t what they believe in. They believe in fairness, de
cency, and equity and that’s how they want the system to work.
If we can get it to work without the regulations, so much the bet
ter. But when it doesn’t work, then Government has a necessary
and proper role to stand in there beside the citizen to see to it that
they get their full citizenship.
That’s what we’re here about today. That’s why this is important.
This is not a garden-variety hearing. This is about America, about
who we are, what our future is going to be, and that’s why we’ve
got to get this done one way or the other.
Thank you very much for appearing here today.
Mr. PATRICK. Thank you, Mr. Chairman.
The C h a ir m a n . Let me excuse this panel and call the next panel
to the table.
Let me introduce our next panel as they’re coming forward and
being seated.
We have Mr. William Tisdale, who is the president of the Na
tional Fair Housing Alliance, which is an organization representing
60 private, nonprofit fair housing agencies located in 35 different
States across the country.
Mr. Robert Hunter is the commissioner of the Texas Department
of Insurance, and, previously, also held the position of Federal In
surance Administrator.
Ms. Lynn Schubert is assistant general counsel with the Amer
ican Insurance Association, a national organization representing
more than 270 insurance companies. She is testifying on behalf of
the Independent Insurance Agents of America and the Council of
Insurance Agents and Brokers.
29
Mr. Charles Kamasaki is here today representing the National
Council of La Raza. This organization represents 170 communitybased organizations which together serve 37 States and reach more
than 2 million Hispanic Americans annually.
We’re delighted to have you all. Mr. Tisdale, we’ll start with you
and we’ll make your full statement a part of the record. I would
appreciate, if you can, if you’d summarize as much as possible and
then we’ll go to questions.
OPENING STATEMENT OF WILLIAM R. TISDALE, PRESIDENT,
NATIONAL FAIR HOUSING ALLIANCE; ACCOMPANIED BY:
SHANNA L. SMITH AND CATHY CLOUD
Mr. T is d a l e . Thank you, Chairman Riegle. With me today are
Shanna Smith, who is the executive director of the National Fair
Housing Alliance, and Cathy Cloud, who is the program director
who coordinated NFHA’s homeowners’ insurance testing project.
They will be assisting me in presenting some of the testimony and
with the questions.
I’m William Tisdale, president of the National Fair Housing Alli
ance, NFHA, located in Washington, DC. I’m also the executive di
rector of the Metropolitan Milwaukee Fair Housing Council.
We applaud you, Senator Riegle, for convening this hearing to
deal with the discriminatory practices and policies of the home
owners’ insurance industry.
NFHA has been investigating these problems since 1991, and
will take this opportunity to reveal publicly, for the first time, some
of the evidence that we’ve accrued based on the conduct, discrimi
natory conduct, of two of the largest insurance companies in this
country—Nationwide and Allstate. These companies have engaged,
and continue to engage, in practices and policies that intentionally
deny, limit, and make unavailable homeowners’ insurance to people
living in minority neighborhoods throughout the United States.
Therefore, today, we are filing formal complaints, these com
plaints, with the U.S. Department of Housing and Urban Develop
ment against both Nationwide and Allstate to force them to elimi
nate their discriminatory policies which have had such an adverse
impact on minority and integrated neighborhoods throughout the
United States.
The testing in this project was funded, for the most part, through
the Fair Housing Initiatives program, a program created by Con
gress to improve fair housing enforcement efforts in this country.
The HUD-funded portion of the project was completed in early
1993. NFHA used its own resources to undertake the analysis and
to conduct additional testing in 1994, some as recently as just last
week.
The insurance industry has been aware of the problem of dis
crimination for decades now. Major changes were announced over
15 years ago. But the studies of the insurance discrimination from
that time through our own testing project, which was just com
pleted, as I indicated, demonstrate that the problem is still perva
sive and persistent. It has not gone away. We have found evidence
of discrimination in the very companies that were the focus of antidiscrimination activities over 20 years ago.
84-051 0 - 9 4 - 2
30
Within the private fair housing enforcement movement, testing
is always carefully constructed and controlled. This project was no
different. These tests compared the treatment between white
neighborhoods and African-American and Latino neighborhoods.
Nationwide and Allstate were selected specifically because of the
egregious nature of the complaints from actual homeowners. We
have been involved in the investigation of other companies as well.
One of the things we noted was that for all of the cities where
we conducted testing, insurance agents tended to be located outside
of minority neighborhoods. We have on display, here, maps of Mil
waukee and Atlanta. In Atlanta, for example, these two companies
have 180 offices, but only four are located in minority neighbor
hoods. All of those four border white neighborhoods. In Milwaukee,
there are no Allstate offices located in minority neighborhoods and
Nationwide operates in Milwaukee only through an 800 number.
The investigation, conducted by NFHA, identified the following
types of discriminatory practices and policies used by agents work
ing for Nationwide and Allstate.
First, was refusing to provide insurance because of the age of the
home. For example, in Louisville, an Allstate agent said, and I
quote:
I don’t like to insure anything over 30 years old. It is too hard, after all the re
modeling and things like that, to mess with. Fm trying to eliminate them from my
portfolio.
Another practice is refusing to insure properties because of the
market value of the home. A Nationwide agent in Milwaukee said:
We don’t insure for less than $55,000. Well, we can’t insure it. We can’t do it.
A third practice is requiring homeowners to provide the name
and telephone number of their mortgage lender, to gather informa
tion about the property before providing a quote, before a quote is
provided.
Just last week, a Nationwide agent asked a Latino tester, who
is your mortgage company? Do you have a number for them? The
reason, ma’am, is because it’s like the difference between an Escort
and a Cadillac. The mortgage company can tell us what the house
is like.
Other demands of minority applicants are requiring inspection of
the home prior to providing information on the cost or type of in
surance available, and requiring a credit check for applicants from
minority neighborhoods.
In addition to these incredible practices just described, the most
pervasive discrimination involved charging higher premiums for
comparable policies, charging higher rates for inferior policies, and
refusing to provide replacement cost coverage on the structure and
contents of homes in minority neighborhoods.
We have brought some displays which illustrate some of the dif
ferences in treatment that I’ve just described.
Based upon a conservative analysis of neighborhood-based tests,
NFHA can document the following for four cities in which testing
was conducted. In Louisville, African-American testers experiencea
discrimination more than 47 percent of the time. In Atlanta and
Milwaukee, African-American testers experienced discrimination
more than 60 percent of the time. And, in Chicago, Latino testers
experienced discrimination more than 95 percent of the time.
31
Insurance companies, in general, are engaging in practices and
policies that have the intent and effect of denying, limiting, or re
stricting homeowners’ insurance for people living in minority com
munities. What we have found is not unique to our investigation.
The practices and policies identified in our testing are widespread
throughout the industry. They have been identified over ana over
and over again in previous investigations.
Our recommendations to this Committee center around two is
sues. The first, is to increase funding for both education and en
forcement efforts. The second, is to create legislation which in
creases our access to information about insurance practices.
Therefore, we request that you continue to increase the enforce
ment budgets of both HUD and the Department of Justice. Specifi
cally, we want to encourage you to increase the allocations to the
Fair Housing Initiatives program, the program that funded this
work, and target a portion of community development block grants
for local fair housing enforcement activities.
Congress should enact disclosure legislation which will provide,
at a minimum, the following: Disclosure of underwriting guidelines.
Disclosure of loss data. Disclosure of the type of policies and the
cost of policies. Reporting of race, national origin, and gender of
policyholders. Reporting for all metropolitan statistical areas. And,
reporting the above information by census tract.
The insurance industry may argue that the cost of disclosure is
prohibitively expensive. Senator, I submit to you that nothing is
more expensive, nothing is more costly, thari the disinvestment in
our neighborhoods.
Senator Riegle, we would like to take just a moment, if we could,
to explain some of our displays.
The C h a ir m a n . Please do.
Mr, T is d a l e . Thank you. Shanna. Cathy.
Ms. S m it h . One of the constant criticisms is that houses that are
used in the testing are run down, vacant, and vandalized.
Senator, these two houses were used in this test of Nationwide
in Milwaukee, Wisconsin. This house was completely denied insur
ance coverage because they have a minimum insurance policy of
$55,000.
This house was quoted at less than $55,000, but the Nationwide
agent provided coverage for this particular house.
This house is in a white neighborhood. This house is in a neigh
borhood that is 80 percent African-American.
It says, “You’re in good hands with Allstate,” and we have a real
question about that.
It says, who gets inferior coverage for double the premium?
In this instance, the testers were told—this tester was calling on
this house in a predominantly white neighborhood and was pro
vided guaranteed replacement cost coverage at an annual premium
of only $155.
This home is in an African-American community. They were pro
vided only a market-rate policy for $330.
But if you look, what people talk about, these are not vacant and
vandalized homes. These are middle-income, moderate-income
neighborhoods. Good housing.
The C h a ir m a n . And well-maintained.
32
Ms. S m i t h . Yes.
The C h a ir m a n . If you look at this house, they planted flowers,
the lawn is in good condition, and the house is in good condition.
Obviously, it’s a house that’s well-tended and occupied by people
who care about the house.
Ms. S m i t h . That was critical to us in our testing, that we se
lected homes that people looking at them would say, why wouldn’t
you insure this property? Because, first, Senator, I would ask, could
you guess which home, here, would be in the minority neighbor
hood?
The C h a ir m a n . I have an idea that it may be the reverse of what
one would think. But you explain.
Ms. S m i t h . That’s correct.
[Laughter.]
In this situation, this white tester called an Allstate agent. Thir
ty minutes after leaving a message on the answering machine, the
Allstate agent returned the phone call. Two days later, the agent
provided the white tester with a quote that included guaranteed re
placement cost coverage.
An African-American woman called over a 5-day period leaving
three messages with the exact same Allstate agent with her name
and phone number. She also had an answering machine. This
agent never returned her call. These are lovely homes, and they
can’t, number one, get equal coverage. They can’t even get their
phone calls returned because they’re located in minority neighbor
hoods.
The C h a ir m a n . Thank you. Those are very powerful illustrations.
Does that complete your presentation, Mr. Tisdale?
Mr. T is d a l e . Yes.
The C h a ir m a n . Thank you very much for all this work and this
valuable analysis. This gets right at the issue and, of course, we
don’t have a way, in this country, of easily calculating what the
economic cost is of redlining and destroying neighborhoods. In a
sense, starving them to death. Not giving them the credit. Not giv
ing them the normal access to economic opportunity that’s avail
able other places.
As communities get into a downward spiral that goes on over the
years, and we end up with terrible problems, as we now have in
many cities, part of that has been caused and fostered by these
veiy discriminatoxy practices. So, in a sense, the cost of allowing
those practices to continue is a very high cost. It’s not just in the
denial of a person’s citizenship rights, in fundamental ways, but
there’s a huge societal cost. There’s a huge economic cost that
wrecks part of America, in effect, all in the name of good business
practice.
It’s a real subversion of the whole free enterprise system, if you
will, because it ends up doing so much costly damage. And, now,
we’re trying to figure out how to fix the damage.
So we com e a lon g w ith program s to try to go b ack in and re p a ir
u rban areas th a t a re in m u ch w orse shape, p a rtly b eca u se th ey’ve
been starved to death from the n orm a l flo w o f b u sin ess tra n s
action al efforts in m ortgages a n d in su ra n ce a n d so forth , th a t
should h a v e b een g o in g in over the years.
33
I think that’s another thing that the public needs to understand.
Where did today’s problems come from? So that when we fix these
defects in the system, over time, we ought to see these communities
start to rebuild and flourish in ways that they’ve not been able to
do.
Mr. Hunter, we’re pleased to have you and we’ll make your state
ment a part of the record. We’d like your comments now.
OPENING STATEMENT OF J. ROBERT HUNTER,
COMMISSIONER, TEXAS DEPARTMENT OF INSURANCE
Mr. H u n t e r . Thank you, Mr. Chairman. I am Bob Hunter, the
commissioner of insurance in Texas.
I’ve been in this room many times before because when I was
Federal Insurance Administrator, this is where both my appropria
tions and my oversight occurred. Even so, on a few times, I enjoyed
being in this room.
I was also here, several times, as president of the National Insur
ance Consumer Organization over the last 15 years. So, I do feel
at home. On this issue, I’ve been here before, too.
In Texas, we have maps, similar to this agent map. if you’d like
to see them, on State Farm in Houston and so on. It s not limited
to these two maps. There are other maps like this.
The C h a ir m a n . I understand.
Mr. H u n t e r . As Federal Insurance Administrator, I documented
insurance redlining on several occasions. During the 1970’s, we
even put evidence before this Committee that showed that you
were more likely in New York to get put into the F.A.I.R. plan, the
high-cost plan, if you were black than if you had a building code
violation.
We know that it was not risk-related based upon that data and
the data that has been flowing for 25 years, all the way since Presi
dent Johnson’s commission that was mentioned earlier.
The data still flows. The studies that we’ve seen from California,
the yellow lines replacing the red lines, the studies of ACORN, and
others. And, now, they don’t use lines at all. It’s income-related,
value of home that you’ve heard about, requiring supporting busi
ness, such as a homeowners’ policy, before they’ll sell you an auto
policy, and so on.
So if you don’t have money, you’re not going to get it, and that
has an obvious impact, not just in the cities, but in rural areas.
Redlining is not limited to poor people in the cities. We just had
a hearing in Big Spring, Texas, and we had witness after witness
coming forth, agents saying, we can’t find a company to write us.
The C h a ir m a n . I see. In rural areas.
Mr. H u n t e r . Even in rural areas, where houses go below these
magic numbers, like $60,000, or whatever the limits are. In rural
areas, that’s also true. I encourage you not to think of it as just
a problem of the inner cities, although that’s been a focus.
On March 31, 1994, I held a public hearing on redlining in Hous
ton. Among other things, and I’ll make the entire transcript avail
able if you would like that for your Committee, but I’ve attached
to my testimony the testimony of a gentleman from Habitat.
Now Habitat, in Houston, has 50 homes. This gentleman works
with the residents in trying to get them their financial needs, in
34
eluding insurance. They were only able to find one standard insur
ance company willing to write them—Farmers Insurance. No other
company would quote.
In the 3 years he’s been working with them, they’ve had one
claim on the 50 homes. A tree fell, during a storm, and damaged
the roof. That’s it. That’s good experience and you know, if you
know something about Habitat, you know that they’re careful in
terms of what they do.
The C h a ir m a n . Right.
Mr. H u n t e r . These are sweat-equity people, working hard to
make their community viable, and they can’t get insurance. I think
that’s very telling testimony. But it’s not just testimony that we
have. At the hearing, only two insurers spoke up. Trade organiza
tions came forward, but individual insurers are reluctant to come
forward in these things.
To their credit, Hartford Insurance Company came forward to
testify that they had undertaken a self-analysis, something that
you just called for, Mr. Chairman. It was painful. What they found
is what the witness testified.
As a result, they’re trying to change the way they’re doing busi
ness. They’re looking to get minority agents, place them in these
redlined places, train them, help them, and support them, to try to
make a difference.
That’s the testimony of Hartford. I don’t believe that many com
panies have done what Hartford has done. That is, the painful kind
of step-by-step review, what are we doing, sitting around a table
saying, gee, what we’ve been doing is wrong. That’s hard.
The C h a ir m a n . We may ask them to come as an example of
somebody who is trying to change and do something constructive
and positive.
Mr. HUNTER. Right. I think you should.
Additionally, I think you have to be careful how you define red
lining. Some try to define it, very narrowly, as, if you don’t draw
a line on a map, you haven’t done anything wrong.
I think you should define it as being unfair discrimination in
availability, price, benefits, service, or quality of insurance, to a
class of people, based upon factors that are not risk-related and are
outside the control of the customer. So that things like the age of
the house—if there’s no good evidence—you end up with redlining.
The C h a ir m a n . We put older houses on historic registries. An
older house isn’t necessarily—that’s a fake issue.
Mr. H u n t e r . Yes. So discrimination like race, gender, income,
value of house, age of house, those things need to be part of the
definition of redlining, in my view, for it to be a complete defini
tion. And how do you attack redlining? I think data is the essential
first step.
So, Senator Feingold and the people that came from the House
this morning are doing the right thing, in my view.
After decades of study, the insurers still say we don’t do it. Let’s
get the data on the table. Data, in small enough geographical units
to see what’s happening, by race, income, ana other demographics,
is essential to this long overdue determination.
We need to know. In homeowners, auto insurance, which is miss
ing from Senator Feingold’s bill, but which I think should be added,
35
and small business lines, at least, we need to know where compa
nies are writing, where they place their agents, what premiums
they charge, and what losses they incur, in my view, in order to
get the answers to these questions.
For a year, Texas has studied the problem. For the first time, we
have zip code data. We have looked at it and we see redlining docu
mented in our data. For example, and some of the data is attached
to my testimony, Mr. Chairman, we have shown that there is a di
rect correlation between lack of availability of auto insurance and
the racial and income make-up of the zip code.
In zip codes where auto insurance availability is two times worse
than the Statewide average, the minority population is two times
higher. And, where coverage is half as hard to get as the Statewide
average, the minority population is half the Statewide average.
We’ve seen a direct correlation. The data is attached to the testi
mony. We’d be happy to share the entire State zip code data with
you, if you would like.
Texas, starting only 1 year ago, is already a front runner in data
collection. The States need the momentum of congressional action
to get moving nationally on this important work. For instance, I’m
in the process of drafting a regulation to require a simple thing—
that underwriting guides be risk-related and based upon substan
tial evidence.
That sounds pretty easy and pretty obvious, but I anticipate a
huge fight on that issue and it will be viewed as a revolutionary
step. But I’m taking that step because of the study of the Texas
Office of Public Insurance Counsel, which was cited earlier, that
showed that 88 percent of the companies have age of home restric
tions in Texas, and 90 percent of the companies restrict writings
based on the value of the home. The typical value is $55,000 to
$60,000, sometimes higher. The median-value of homes in Texas is
$43,000. So, they’re writing off a lot of homes.
The C h a ir m a n . That’s interesting. The median-value is $43,000.
Mr. H u n t e r . Right.
The C h a ir m a n . S o half the houses in the State are at that level
or below that level.
Mr. H u n t e r . Right. Well more than half are below the $60,000.
The C h a ir m a n . Yes, absolutely.
Mr. H u n t e r . S o low-income areas of Texas are being written off,
and that’s redlining, in my view.
The C h a ir m a n . It’s second-class citizenship. That’s basically
what it boils down to.
Mr. H u n t e r . It’s exactly what you just said a few minutes ago,
Senator.
Last week, I imposed a record fine on Allstate Insurance Com
pany, $850,000, the most ever levied ag;ainst a company. They’ve
agreed to it as a consent order. It’s twice what my lawyers rec
ommended.
[Laughter.]
The C h a ir m a n . It looks like you need new lawyers, too.
[Laughter.]
Mr. H u n t e r . They wouldn’t write people with only one car. They
wouldn’t write people who were not in the market for other insur
36
ance, like homeowners. They wouldn’t write single people, and so
on. As a result, we had these kinds of practices.
At the same time, our staff, newly invigorated after this, has
brought actions against 59 other companies, disciplinary actions,
with similar problems. Those 59 are based upon our sample compa
nies of 75 that we reviewed. So 59 out of 75.
In your invitation to testify, you raised four questions which I
answered in my prepared statement. As an actuary, I believe you
should require census tract, not just five-digit zip, to enable re
search in homogeneous areas, such as you do in HMDA. I think
that’s the right approach.
I believe you should obtain age data, as well as race and gender,
because I think there’s discrimination by age.
I think you should get the claims data. If you don’t, when you
tell the company, look, you’re doing something bad here, they’ll say,
it’s a good business practice and we can prove it, but you won’t be
able to test that argument.
I encourage you to consider strongly S. 1917, Senator Feingold’s
bill. I would strengthen it. I would look at Congressman Kennedy’s
bill for ideas.
Specifically, I would collect auto insurance data, which the bill
does not. Many poor, in minority areas, don’t own homes. People
don’t own homes and they’re being discriminated against where
they have to buy auto insurance. Also, renter’s insurance should be
adaed, in my view.
I would like to make all the data open-record data. One of the
key problems I have in Texas is that, although I can now collect
underwriting guidelines, I cannot disclose them by company. So the
practices are not publicly available. And, I think, public exposure
of some of these tilings will change them, just the public exposure.
The C h a ir m a n . Can you hold a public meeting and describe the
practices, listing them as Company A, Company B, and Company
C?
Mr. H u n t e r . I can do that. In fact, that’s what we have done.
T h e C h a ir m a n . S o th e com pa n y th at w an ts to p u rsu e it can in
quire o f th e com pan y th a t th ey m igh t be ta lk in g to, w h ere th ey fall.
Mr. H u n t e r . Right. But, in Texas, the legislature, I think wisely,
has moved in the direction of using competition to control rates.
So I get out rate guides. People love the rate guides. They go out
by the tens of thousands, and we are encouraging people to shop.
But what happens is that, in areas like this, people come back
and say, well, I found a cheap company, but they won’t even quote
for me. And I can’t even tell them, don’t bother with that company,
based upon what I already know about their underwriting guide
lines, because I’m prohibited to release that.
The C h a ir m a n . Right.
Mr. H u n t e r . People get frustrated, give up, and go to the as
signed risk plan, the F.A.I.R. plan, or something.
Just speaking for myself, one last personal note. I think I need
your help. The momentum that these kinds of hearings and this
type of legislation being debated gives to the State is very, very im
portant.
Texas is ahead of the curve, in my view, of what’s going on in
the rest of the country. Only 22 States agreed to participate in the
37
National Association of Insurance Commissioners’ study of redlin
ing by supplying data, and that tells you something, in my view.
I do thank you for your leadership on this, and Senator Feingold
and the Ranking Republican.
The C h a ir m a n . Thank you for what you’re doing in Texas and
for coming today. It’s a very helpful example of national leadership
that you’re giving, and I much appreciate it.
Mr, H u n t e r . Yes, sir.
The C h a ir m a n . We’ll follow up at our end.
Ms. Schubert, we’d like to hear from you now.
OPENING STATEMENT OF LYNN M. SCHUBERT, ASSISTANT
GENERAL COUNSEL, AMERICAN INSURANCE ASSOCIATION
Ms. S c h u b e r t . Thank you, Mr. Chairman, for inviting us to tes
tify here today on the important topic of equal access to insurance
for all Americans. As you stated, ALA represents more than 270
companies writing property and casualty insurance in every State
and jurisdiction in the United States.
I think it’s probably important for me to state, on the record,
that the companies that have been discussed earlier today, on this
panel, are not members of the ALA, but the Hartford Company,
that Mr. Hunter did mention, is a member of ALA, and we have a
number of companies like that.
We’re here today, Mr. Chairman, as a group who is trying to do
something positive and trying to make a difference in this area.
The organizations, that I’m testifying on behalf of today, have
one overriding goal on the issue of insurance access and availabil
ity. That is, that urban residents and businesses, equally with all
other Americans, must be able to purchase attractive insurance
products at a price reasonably based on the risk. ALA, LIAA, and
the Council members are committed to working with legislators,
regulators, consumers, and brokers to ensure that this occurs.
To determine whether or not insurance is available in all areas,
we could support Federal data collection of insurance information
if it is designed to produce a fair, efficient, and effective study of
insurance availability and cost. H.R. 1188, pending in the House,
appears to be such a measure and AIA, IIAA, and the Council sup
port this bill.
Any data collection will show that some areas, especially in our
cities, have higher numbers of residents and small businesses with
out insurance than other areas. These discrepancies are related to
a whole host of socioeconomic circumstances faced by people who
work and live in the urban areas and which also increase the cost
of insurance. Outright racial or ethnic discrimination also may
occur. We hope that it is infrequent. We know that it’s a violation
of Federal and State law. AIA advocates stringent prosecution
wherever it is found.
The American birthright, that you described, to equal treatment
should be protected and it’s something that we certainly support.
Any discrepancies in availability of insurance, whether they’re
caused by discrimination or by economics, must be addressed.
It is cold comfort to the citizens for whom insurance is unavail
able to explain to them the reasons why the problem exists. We be
lieve that it is time to start attacking these problems head-on.
38
Mr. Chairman, the Committee has asked AIA to address three
specific questions in its testimony. We have done so, in detail, in
our written statement. This morning, I will attempt to summarize
for you our responses to those questions.
First, you asked for AIA’s views on whether insurers illegally dis
criminate. It’s impossible to say that there’s not one underwriter,
one agent, or one company out there, that is illegally discriminat
ing in the selling of insurance. We do not believe that it’s a sys
temic problem, but, rather, if it occurs, it occurs in isolated in
stances.
The most important thing to note in this discussion is that even
in the question itself, the description states that discrimination is
illegal, and that is a fact. It is illegal in every State. It’s illegal
through Federal law.
All States have laws against unfair discrimination in the insur
ance industry. The National Association of Insurance Commis
sioners Model Unfair Trade Practices Act prohibits arbitrary un
derwriting decisions, based upon geographic location alone, as un
fair discrimination. AIA strongly supports this provision. We sup
ported its adoption at the NAIC and we have urged its adoption by
State legislatures across the country.
If insurers are illegally discriminating, they should be pros
ecuted. States currently have the authority to do this and are doing
so in appropriate situations. As was stated earlier, insurance must
be based on the risk. It cannot be based on race.
The second request of the Committee was for AIA to discuss the
findings of recent property insurance studies. This is the bottomline question that really must be addressed, whether consumers
wishing to purchase insurance currently are able to do so.
A number of studies, from 1979 through 1993, show that a large
percentage, 98 to 99 percent, of homeowners have homeowners’ in
surance. More recently, in 1993, AIA commissioned a study, by an
independent agency, of the core of six American cities, and I’d just
like to point out some of the key highlights of this survey because
it did look at predominantly minority zip codes. It looked at areas
where we were concerned there may be a problem.
From those six cities, less than 2 percent of the homeowners sur
veyed did not carry any homeowners’ insurance. Ninety-three per
cent had comprehensive coverage and about 5 percent carried more
basic homeowner insurance policies covering fire damage only, or
fire and windstorm damage.
There were no significant differences among African-Americans
and whites in terms of the insurance coverage. Ninety-nine percent
of African-American homeowners had insurance with 92 percent of
those African-Americans having comprehensive homeowners’ insur
ance.
Ninety-eight percent of those homeowners identifying themselves
as white had some homeowners’ insurance, including 94 percent
with comprehensive policies. Nearly 9 in 10 urban homeowners
said it was very, or somewhat, easy to find homeowners’ insurance.
An even higher share, 93 percent, said it was convenient to contact
an insurance agent or insurance company. Very few respondents,
only 3 percent, said they were aware of anyone in their neighbor
39
hood who had experienced difficulty in obtaining homeowners’ in
surance.
There are other surveys and studies that show similar results.
Those results are discussed in detail in our written statement.
However, there are other studies that show other results and we’ve
heard discussions, today, of testing, that apparently may or may
not show different results when we finally get the details and we
can investigate that situation.
While we have a detailed analysis of the cause of the differences
between the studies and critiques of the different methodologies of
the studies, we are not here today to argue the numbers with you.
We agree that it is time to take a close look at this issue. And, in
stead, these differences are the very reason we’re willing to support
Federal data collection. We want to know what the information is
on the availability and cost of certain lines of insurance.
As Representative Kennedy has stated, those who have nothing
to hide, have nothing to fear. We welcome thoughtful investigation
and analysis of this issue.
This leads directly to the third question posed by the Commit
tee—What steps should the Federal Government take to identify
and combat illegal property insurance discrimination if such dis
crimination exists?
As mentioned earlier, a fair, economical, efficient data collection
scheme, on a Federal level, to determine if insurance is equally
available, based on the risk for all Americans, is something that
AIA, IIAA, and the Council support. The parameters of such a data
collection effort, however, are very critical.
Inordinate costs of collecting aata with limited value would not
assist the insurance consumer, but, rather, harm that consumer
with either higher prices or possibly less financially sound insurers.
Collection of data which reveals what insurance is being sold,
where it is being sold, who is selling it, and how much it costs the
consumer would answer the question of whether or not insurance
is equally available. This collection should be undertaken in a for
mat to provide the most information in the most useful form in the
most cost-efficient fashion.
H.R. 1188 appears to establish such a data collection scheme and
I urge this Committee to review that bill, as well as the bill pend
ing in the Senate. The information which will be provided by
H.R. 1188 would allow regulators to focus efforts on zip codes. It is
a collection by five-digit zip code. It would allow regulators to focus
their efforts on zip codes with significant disparities between the
number of homeowners and the number of homeowners’ insurance
policies, rather than being required to review detailed information
for huge numbers of census tracts where no problems exist whatso
ever.
We urge you to look at a data collection scheme that would allow
you to look broadly at all the zip codes where you have a concern
and then focus your efforts on where there are these disparities.
Spend the time of the regulators, then, getting into the more de
tailed information requiring insurers to produce more detailed in
formation, at a smaller level, at that point. Regulators, currently,
can do that and I urge that that is the way to move forward in try
ing to address this issue.
40
Mr. Patrick did state, very accurately, that several hundred thou
sand insurance applications are collected every year by insurers in
comparison to only thousands of mortgage applications.
HMDA is not necessarily appropriate for the insurance industry.
AIA recognizes there are a number of issues that may have an
impact on urban consumers and their ability to obtain insurance.
AIA members and others are taking steps to address these issues
and I would like to take just a minute, if I could this morning, to
address some of our efforts.
The CEO’s of many AIA member companies are doing just what
you, Mr. Chairman, suggested. They are making a commitment to
equal treatment. Additionally, it is our belief that partnerships be
tween the insurance industry and consumer organizations, civil
rights organizations, and housing organizations are the most pro
ductive way to move forward.
AIA would also like to work with legislators and regulators to de
velop products and marketing ideas which adequately serve all
residents and businesses. We want to discuss the establishment of
programs to facilitate greater access to insurers.
For example, we already are working with regulators, in a num
ber of States, on programs to bring urban-based independent
agents together with our standard companies. One State, which is
moving forward quickly on this issue, is Georgia. In Atlanta, AIA,
the Urban League, the IIAA, the insurance department, and State
legislators have established a task force to address the issue of in
surance in urban markets.
One of the first projects of the task force is the agent-insurer
partnership. This partnership has resulted in a number of appoint
ments already being made by standard companies to urban-based
and minority agents.
Now that this program has proven successful, we are working to
expand these efforts nationwide. We have specifically spoken with
Commissioner Hunter, and he has requested some detailed infor
mation on the Atlanta project. We’re looking forward to moving for
ward in Texas as well as a number of other States.
In addition, we also support the inclusion of homeowners’ insur
ance in existing F.A.I.R. plans to address the needs of homeowners
who are unable to find insurance in the voluntary market.
In California, we’ve joined forces with a broad coalition of con
sumer and civil rights groups, led by Consumers Union and the
Latino Issues Forum, to create an innovative, no-frills, no-fault
automobile insurance policy. This policy is a perfect example of a
creative, practical, and sound solution to a real problem of people
in need.
Independent actuaries have concluded that this policy could save
California consumers $1.8 billion in the first year alone. We are
working to inform legislators and regulators about the potential
benefits of this and similar new products.
Education----The C h a ir m a n . I’m going to ask you, if you can, to wrap up in
the next 30 seconds because I’m afraid we’re not going to have
enough time for Mr. Kamasaki and, then, I’ve got to go tend to an
other assignment, here, shortly.
41
Ms. S c h u b e r t , Yes, sir. I have just two points to make very, very
briefly.
The C h a ir m a n . Thank you.
Ms. S c h u b e r t , Education and reduction of the risk.
AIA is working with State regulators and legislators on the issue
of education, both for the insurance industry about opportunities in
urban markets, as well as consumers, to educate consumers on how
to shop for the best product.
And, second, most importantly, perhaps, we’re working with
consumer and community organizations like ACORN, to wont with
their Neighborhood Home and Safety Program, to reduce the risk
in neighborhoods so that we can make insurance more cost-efficient
and more available.
In conclusion, we just want to state that we’re committed to
working on this issue. We want to work with the Senate. We sup
port data collection and we look forward to working with you on
making it effective and efficient.
Thank you.
The C h a ir m a n . Thank you. I appreciate the constructive tone
that you present from your organization and some of the positive
steps that have been taken by tne Hartford and others.
Also, we may have some differences at the end of the day in
terms of what we collect and how we collect it, but I would suggest
that we try to work together to see if we can work that out.
Mr, Kamasaki, we’re pleased to have you.
OPENING STATEMENT OF CHARLES KAMASAKI,
NATIONAL COUNCIL OF LA RAZA
Mr. K a m a s a k i . Mr. Chairman, I’m happy to be here. Let me first
apologize on behalf of my president, Raúl Yzaguirre, who is not
able to be here. I apologize to you that you’re getting the second
string, and I will try to be brief in deference to your time con
straints. Let me try to make three points.
The first is, I believe, the evidence is sufficient to warrant imme
diate congressional action. Mr, Chairman, in my written statement
we go through, in quite exhaustive detail, the history of social
science research on discrimination, as it applies to Latinos, in vir
tually every field—employment, housing, the housing rental mar
ket, mortgage lending, and the emerging data that is now coming
out with respect to insurance.
The point I would like to make is that, at virtually every stage
of this process, there have been naysayers. They have argued—the
industry in each of these cases has argued—that the problem is not
serious, that these are isolated examples, and that they’re caused
by market factors not related to discrimination.
I would note that after disclosure or after there has been suffi
cient social science research on these issues, invariably, the evi
dence has demonstrated that, in fact, the problem is serious, that
it is widespread, that it goes beyond isolated examples, and that
alternative causes, like market factors, do not explain disparities
between minorities and nonminorities.
We believe such is the case with insurance redlining.
It is clear, and we agree, that there is not sufficient data to make
conclusive judgments about the scope and degree of insurance red-
42
lining with respect to either African-Americans or Latinos. We ac
knowledge the fact that the industry, at least, makes the argument
that some of these disparities may be attributable to factors other
than discrimination.
But, Mr. Chairman, given the history that we document in our
statement, and given the history on these issues, it seems to me
this is one of those cases where we think it’s a duck. It looks like
a duck. It walks like a duck. It quacks like a duck. And that, if
we needed no evidence other than the fact that the recent National
Fair Housing Alliance testing that we were pleased to join in and
which demonstrates a virtual statistical certainty of discrimination
for each Latino seeking insurance, we would argue the problem is
clearly sufficiently documented to warrant an immediate congres
sional response.
The second point I would make is to note how modest this re
sponse is. We are not asking for new regulations. We are not ask
ing for a new standard for judging discrimination. We’re not asking
for affirmative action or quotas in the number of insurance policies
sold. What we are asking for is data disclosure, to permit enforce
ment agencies to review data, to permit social scientists, research
ers, ana others to review the data, and, finally, to permit the kind
of self-analysis that you mentioned earlier today.
In fact, Mr. Chairman, if every CEO of every insurance company
were to issue the kind of order that you talked about earlier today,
that CEO would, in fact, need the very kind of data that is re
quired in this bill in order to assess the performance of his/her own
company and assess the performance of each and every employee
in that company.
It seems to me, given the scope of the problem, that the kind of
response that we and others are seeking embodied in Mr. Ken
nedy^ bill and Mr. Feingold’s bill, is really quite modest.
Finally, I would just note for the record that there have been
some press reports that have alleged that there is somehow some
disunity within the civil rights community on this issue. I just
wanted to introduce, and ask that it be made part of the record,
a letter sent to you yesterday, on behalf of a coalition of 24 civil
rights and consumer organizations, which demonstrates that those
press reports are not accurate.
Let me stop at this time and entertain questions.
The C h a ir m a n . Thank you. Thank you for very good remarks.
Without objection, we’ll make that letter a part of the record.
I want to thank all of you for your testimony today and for your
leadership in your respective capacities in this area. I think we’ve
really laid the problem out today, along with some very construc
tive ideas, legislative and other, as to how we go about dealing
with it.
But I want to thank each of you and I want you to know that
when we state intentions in this Committee, we have a very high
batting average of following through and doing what we say we’re
going to do. We’re going to push ahead on this because it’s very im
portant. I urge all of you to do so as well, and let’s collaborate as
we go and see if we can’t make a lot of progress here.
Thank you very much.
Ms. S c h u b e r t . Thank you, Mr. Chairman.
43
M r. T is d a l e . T h a n k you.
The Ch a ir m a n . The Committee stands in recess.
[Whereupon, at 12:22 p.m., the Committee was recessed.]
[Prepared statements, response to written questions, and addi
tional material supplied for the record follow:]
44
PREPARED STATEMENT OF SENATOR RUSSELL D. FEINGOLD
Thank you, Mr. Chairman, for holding this hearing on the problem of discrimina
tion in the determination of who has access to affordable high quality homeowners’
insurance in America. I would also like to thank you for requesting the recent GAO
study on data needed to examine issues pertaining to the availability, affordability,
and accessibility of property insurance, and for your past leadership and continued
efforts through community banking and reinvestment legislation to expand financial
services to individuals and communities that have been bypassed in the past by the
private market. Your leadership on these and other issues will be sorely missed here
m the Senate and by the entire Nation, and I feel fortunate to have had the oppor
tunity and privilege of serving with you in this body—even for the short perioa of
time it has been.
During this hearing the Senate Banking Committee will gather testimony from
a diverse group of individuals and organizations on a subject that strikes at the core
of the ability of many Americans to participate fully in our society by being able
to enjoy that which has come to be known as the “American dream”—home owner
ship.
Three decades of research, studies, and reports have reaffirmed the extent of the
iroblem of insurance redlining that prompted the 1968 National Advisory Panel on
nsurance’s description that “Communities without insurance are communities with
out hope” as well as the inadequacy of State and Federal responses to address it.
These studies, as well as testimony and evidence gathered before both the House
Banking and Energy and Commerce Committees, thanks to the efforts of both Rep
resentatives Joseph Kennedy and Cardiss Collins, have indicated that hundreds of
thousands, if not millions, of individuals and entire neighborhoods continue to be
denied or provided inferior insurance coverage and that insurance redlining prac
tices are currently widespread throughout the United States.
It is not only disturbing that discrimination continues to exist today, but it trou
bles me even more so that the fine city of Milwaukee, Wisconsin, has received na
tional attention regarding this problem. In fact, a CNN television report even stated
that Milwaukee is becoming not only famous for beer, but for insurance discrimina
tion as well.
Today’s disturbing results concerning the occurrence of insurance discrimination,
to be released by the National Fair Housing Alliance and later discussed here by
its President, William Tisdale, a fellow Wisconsinite who shares my concerns over
the problem at a national level as well as in Milwaukee, will add to this perception
and come on the heels of an Urban Institute report which showed that Milwaukee
was the most economically segregated city among the 100 largest U.S. metropolitan
areas.
Some will argue that the disparities in the access to and availability of insurance
are solely based on principles of economics and statistically based risk assess
ments—and not on principles of prejudice. Sadly enough, it appears that this is not
always the case. Take the words of an insurance company district sales manager
from Milwaukee, who was recorded as telling his agents:
“.
I think you write too many blacks. .
You gotta sell good, solid premium
paying white people. . . . They own their homes, the white works.
Very hon
estly, black people will buy anything that looks good right now.
But when
it comes to pay for it next time.
You’re not going to get your money out of
them.
”
This “quit writing all those blacks .
prejudicial policy was not only commu
nicated to agents verbally, but was placed m writing as well. And it has been re
ported that tne manager even showed one agent how to accomplish this goal by stat
ing that:
“If a black wants insurance, you don’t have to say, just tell them, because based
on this kind of policy, the company will only allow me to accept an annual pre
mium. Do it that way.”
Activity of this type that has prompted such allegations of discrimination in the
insurance industry cannot and must not be tolerated anywhere in our society.
To address this problem, I introduced S. 1917, “The Anti-Redlining in Insurance
Disclosure Act of 1994,” which would require insurance companies to disclose infor
mation regarding where they write property insurance, patterned after the reporting
requirements already required of banks and thrifts unaer the Home Mortgage Dis
closure Act. This information would allow members of the public, regulators, the in
surance industry, and Congress the means to identify the extent of tne problem and
would assist affected individuals and Federal and State agencies efforts at enforcing
our Nation’s anti-discrimination laws.
f
45
There are several components which S. 1917 includes which are critical for any
meaningful measure designed to address the problem of insurance redlining.
First, it is important tnat any data collection and reporting requirements on in
surance costs and policies be done at the most detailed level which is reasonably
feasible. For example, it is preferable to require reporting by census tract rather
than zip code since it allows for more detailed detection ana analysis, as census
tract populations are much smaller and homogeneous than many urban zip codes
which often contain neighborhoods that have a diverse range of economic, racial,
and housing stock characteristics, thereby resulting in a “masking^ effect for pur
poses of statistical analysis. The census tract reporting standard that is required of
the banking industry under the Home Mortgage Disclosure Act should be applied
to the insurance industry as well. S. 1917 takes this approach.
Second, the collection of data on insurance losses and claims should also be in
cluded in any insurance disclosure initiative. Such data would be essential for any
proper analysis necessary to resolve disputes that arise involving claims that dis
parities in the price of insurance between different neighborhoods or groups of peo
ple are solely based on loss experience and the associated risk involved rather tnan
prejudice.
Finally, since the collection and disclosure of such data will provide affected indi
viduals and Federal and State regulators valuable information necessary to enforce
our Nation’s anti-discrimination laws, it should be made available to the greatest
number of communities and individuals as possible. S. 1917 would require that data
be collected in 150 Metropolitan Statistical Areas.
I am also interested in exploring whether or not the insurance industry should
be required to meet the same requirements that are imposed upon the banking in
dustry under the Community Reinvestment Act. To that end, S. 1917 calls for a
study which would review the feasibility of creating reinvestment requirements for
insurers similar to those already applied, to depository institutions.
In conclusion, I would like to again thank Senator Riegle for holding this hearing
and look forward to working with the distinguished Chairman and other Members
of this Committee toward addressing the problem of insurance redlining.
PREPARED STATEMENT OF SENATOR ALFONSE M. D’AMATO
Mr. Chairman, I would like to join you in welcoming our distinguished panel of
witnesses, particularly our colleagues from the House, Representatives Collins and
Representative Kennedy. The presence of so many distinguished witnesses says a
great deal about the importance of the subject matter of today’s hearing.
Today the Committee is focusing its attention on insurance redlining-whether,
and to what extent, it occurs. In this context, we are concerned with discrimination
against residents of economically distressed areas in the writing of homeowners’ in
surance based on factors unrelated to sound actuarial practices.
Mr. Chairman, this Committee, under your leadership, has invested a great deal
of energy in addressing the plight of economically deprived individuals and commu
nities. The Committee has worked hard to ensure that every American is treated
fairly in seeking financial services.
Mr. Chairman, just last session, you and I sponsored a bill designed to eliminate
the abusive mortgage lending practices of unscrupulous “loan sharks” whose highinterest mortgage scams are airected at the elderly and low-income homeowners.
With respect to the insurance industry and homeowners’ insurance, the guiding
principle must be sound and objective underwriting practices. My father is an insur
ance professional. He always took great pride in helping his customers’ plan for
their future financial security. Now, the risk of insuring any two homes is never the
same—that is why rates should vary from case to case. But my father has always
believed that every potential customer must be judged by one standard—by the risk.
Any other standard just won’t do.
This conviction is shared by the Committee. Discrimination of any kind is wrong
and it is unacceptable. For this reason, the Committee has invested time and energy
to ensure fair access to financial services for one simple reason.
At the same time, we must remember that insurance companies are in business,
and an integral part of that business is assessing and managing risk. Insurers must
be able to assess risk in a businesslike fashion. That means that they must be able
to charge rates in accordance with the risks involved.
Mr. Chairman, the Committee must review the available evidence and ascertain
whether, and to what extent, the insurance industry engages in discriminatory prac
tices in the writing of homeowners’ insurance.
46
PREPARED STATEMENT OF DEVAL L. PATRICK
A s s is t a n t A t t o r n e y G e n e r a l
U.S. D e p a r t m e n t
of
fo r
C iv il R ig h t s
J u s t ic e
Mr. Chairman and Members of the Committee, I appreciate the opportunity to ap
pear before this Committee to discuss what the Attorney General and I believe to
be an increasingly significant issue in civil rights enforcement—discrimination on
the basis of race or national origin in the provision of homeowners’ insurance. I as
sure you that I will use the full extent of my authority as Assistant Attorney Gen
eral for the Civil Rights Division to eliminate all forms of race and national origin
discrimination that hinder the ability to obtain and enjoy housing, including dis
crimination in the property insurance industry.
One lesson we have learned from more than 25 years of fair housing enforcement
is that the actions of many players can have an impact on the ability to obtain and
enjoy housing. The Civil Rights Division has often challenged direct providers of
housing by taking action against unlawful discrimination m the sale and rental
markets. In recent years, we have targeted unlawful discrimination in the mortgage
lending industry and have learned the devastating impact that such discrimination
can have on individuals and communities. Although we know less about the extent
of discrimination in the homeowners’ insurance industry, we do know that ending
such discrimination is critical to fair housing enforcement. As the United States
Court of Appeals for the Seventh Circuit succinctly noted, “no insurance, no loan;
no loan, no house . .
NAACP v. American Family Mutual Insurance, 978 F.2d
287, 297 (7th Cir. 1992), cert, denied, 61 U.S.L.W. 3771 (U.S. May 17, 1993).
Our Department recently completed an investigation involving discrimination in
the provision of homeowners’ insurance, and we have also investigated claims of in
surance discrimination originally presented to the Department of Housing and
Urban Development [HUDjT Although the results of the investigations are not yet
public, the long and detailed investigations have helped educate us regarding how
such discrimination occurs. Our work has also clarified for us the type of factual
information that is necessary to detect unlawful discrimination in this industry. We
do not purport to have all the answers to the difficult questions but I am pleased
to share our experiences with you.
Discrimination in Insurance is Prohibited by the
Fair Housing Act
Historically, the efforts of our Department to address the issue of homeowners’
insurance discrimination have been slowed by claims that such discrimination is not
sufficiently related to housing to evoke coverage under the Fair Housing Act. The
coverage issue has been the focus of several lawsuits and initially resulted in incon
sistent Federal court decisions. The confusion caused several insurance companies
that were subjects of Federal investigations to refuse to cooperate, even in the face
of HUD subpoenas. We believe that tnis issue can now be laid to rest such that we
can proceed to resolve the merits of discrimination claims.
Our Department and HUD have consistently argued that discrimination by insur
ers of housing violates the Fair Housing Act. The Department of Justice presented
that argument in 1978 as amicus in the first case to address the issue, Dunn v.
Midwestern Indemnity Co., 472 F, Supp. 1106 (S.D. Ohio 1979). In that case, the
court held that “the concerns of insurance redlining are within the especial province
of the Fair Housing Act.” Id. at 1112. The Court of Appeals for the Fourth Circuit,
however, reached a contrary holding in Mackey v. Nationwide Insurance Companies,
724 F.2d 419 (4th Cir. 1984), and this is the decision upon which investigative tar
gets relied in resisting our enforcement efforts. This decision, however, did not
change the position of the enforcement agencies. On January 23, 1989, HUD pub
lished regulations implementing the Fair Housing Act stating that “[r]efusing to
provide , . . property or hazard insurance for dwellings or providing such
in
surance differently because of race, color, religion, sex, handicap, familial status, or
national origin” constitutes a violation of the Act. 24 C.F.R. § 100.70(dX4).
But the road to substantive enforcement has been difficult. The Civil Rights Divi
sion received allegations in 1988 that American Family Insurance, a Wisconsin in
surer, had directed its agents not to sell property insurance in areas where AfricanAmericans constituted a majority of the population. An investigation was initiated
and the company initially said it would cooperate. When a formal request for infor
mation was presented to the company, however, it declined to provide any informa
tion, citing its position that the Department of Justice lacked authority under the
Fair Housing Act to investigate claims of racial discrimination in homeowners’ in
surance. The local NAACP chapter subsequently filed a lawsuit raising claims under
the Fair Housing Act, 42 U.S.C. §§ 1981 and 1982, and the company raised the same
47
defense to the Fair Housing Act claim. We supported the private plaintiffs’ Fair
Housing Act claim as amicus curiae,, but the district court dismissed tnat claim. The
private lawsuit was permitted to proceed since other causes of action remained; but
since the remaining statutes do not provide enforcement authority to our Depart
ment our investigation was thwarted. We participated as amicus curiae in the Court
of Appeals for the Seventh Circuit and tnat Court, on October 20, 1992, reversed
the district court decision and held that the allegations did state a claim under the
Fair Housing Act. NAACP v. American Family Mutual Insurance, 978 F.2d 287 (7th
Cir. 1992), cert, denied, 61 U.S.L.W. 3771 (U.S. May 17, 1993). That decision al
lowed our investigation to resume after a delay of almost 4 years. The private case
has yet to be resolved.
Nationwide Insurance Company also claimed the Fair Housing Act did not reach
racial discrimination in homeowners’ insurance when it refused to comply with a
HUD request for information necessary to investigate a complaint of discrimination
in the provision of homeowners’ insurance. The company sued the Secretary of HUD
over the issue and in February 1994, after more than 2 years of litigation, an order
was entered upholding our position that the Fair Housing Act prohibits racial dis
crimination in the provision of homeowners’ insurance. Nationwide Insurance Co. v.
Cisneros, Civil Action No. C-3-92-52, (S.D. Ohio, Feb. 24, 1994). The company has
appealed the district court’s decision. The Department oi Justice urged this same
position as amicus curiae in private litigation before the District Court for the
Northern District of Indiana in United Farm Bureau Insurance Co. v. Metropolitan
Human Relations Commyn., C.A. No. F89-252 (N.D. Ind.) and that Court agreed
with us, following the Seventh Circuit precedent in American Family. We also urged
the District Court for the Northern District of Ohio to allow a Fair Housing Act
claim that a private mortgage insurance company was discriminating on the basis
of national origin. Briceno v. United Guaranty Residential Insurance Co., Case No.
3:89CV7325.
In view of this litigation, the HUD regulations are particularly significant, since
they are entitled to substantial deference by the courts. And we do not believe that
there is an existing conflict between the circuits, since the Fourth Circuit’s decision
in McLckey was rendered prior to the issuance of the regulations. The Seventh Cir
cuit concluded that the regulations were valid and controlling, and we believe that
reasoning should be followed by other courts that are presented with the issue.
Enforcement Efforts
As a result of our work on discrimination by property insurance companies, our
initial lesson is that such investigations can be exceedingly complex. The investiga
tions are sometimes compared to lending discrimination investigations, but several
differences make insurance discrimination investigations even more complex.
The Home Mortgage Disclosure Act (HMDA) provides us useful information about
the lending practices of depository financial institutions. We know the number of
minorities who apply for mortgage financing, the number accepted, and the number
rejected. This information is very helpful in selecting targets for investigation. We
do not have comparable information about the performance of property insurance
companies. Also, while a depository institution may receive several thousand loan
applications in a given year, a property insurance company may write or renew sev
eral hundred thousand policies in the same time period. Unlike lenders, property
insurance companies, to the best of our knowledge, do not maintain records regaraing the rejected applications nor do they record the race of applicants or policy
holders.
For these reasons, it is difficult to assess claims that a particular company is dis
criminating, for example, by refusing to insure properties in African-American or
Latino neighborhoods. Discrimination can occur without an application or a record
if an agent declines to return a phone call, fails to keep an appointment, or simply
tells a prospective customer that the company will not insure the property because
of its condition or location. Given the state of available records, this type of discrimi
nation might best be detected by traditional fair housing testing. We are anxious
to review the results of testing of property insurance companies which we under
stand are soon to be released by the National Fair Housing Alliance so that we can
obtain a better understanding of discrimination that might be occurring at this pre
application stage.
We also believe that possible discrimination at the underwriting stage must be
addressed. Our own investigations have revealed that policyholders in minority
neighborhoods might receive insurance only after being subjected to inspections or
being required to make repairs that are not required in other neighborhoods. In ad
dition, properties in some minority neighborhoods are appraised for a lower amount
than the replacement cost, but we do not believe that such facts should preclude
48
a property owner, who is willing to purchase the necessary insurance, from protect
ing his or her property and guaranteeing that it can be replaced in the event of a
catastrophe. However, we have found that dwellings in minority neighborhoods
often receive inferior insurance coverage that only allows for the repair, rather than
the replacement, of a dwelling in the event of a property claim, or imposes low dol
lar limits that prevent its replacement. We also believe that some companies may
charge a higher premium per dollar of insurance coverage for the inferior repair cost
policies than they do for the more desirable replacement cost policies.
Finally, insurance companies can inflict substantial damage on neighborhoods by
refusing because of the racial or national origin identity of the neighborhood to mar
ket their products. It is often revealing to examine the number of insurance offices
located in minority neighborhoods as compared to the number located in identifiably
white areas. The marketing of insurance products relies heavily on a community ap
proach with agents located in close proximity to the areas they intend to serve. If
a company desires to redline an area as off-limits for company business, it can do
so without drawing a red line on a map—it can simply clecline to open offices in
the area.
We are attempting to address these issues in our investigations. We have ex
pended considerable resources to correlate the addresses from company records with
the census tract in which the property is located, a process known as “geocoding.”
This enables us to evaluate the racial impact of questionable policies. We are exam
ining differences in pricing and policies offered. We are looking closely at marketing
efforts and comparing market shares in white and minority neighborhoods. Of
course, insurance companies remain free to make decisions based on risk, but there
is a substantial difference between risk discrimination and race discrimination.
As should be obvious from my description of the work of our Department, im
proved information sources would promote better compliance with, and enforcement
of, civil rights laws. It would be nelpful to have information comparable to that
which we have about banks, such as the number of applications received by race
and national origin and the corresponding number of acceptances and rejections.
HMDA, which requires reporting at the census tract level, has proved very useful.
We rely heavily on that information, which allows us to identify the racial or ethnic
identity of a select area. If the information is reported in such a way that it includes
diverse neighborhoods, it still may be necessary to perform the arduous and expen
sive task of geocoding policy activity to determine the impact of company practices
on minorities. We also would find it useful to have information on the types of poli
cies issued, by race and national origin, in each neighborhood. This information
would allow us to evaluate readily whether minority neighborhoods are offered infe
rior insurance protection. Information on claims paid and loss data would also help
us. The Administration is pleased that legislation to combat redlining in insurance
has been reported by two House Committees. The Administration looks forward to
prompt House passage of the anti-redlining legislation and urges favorable consider
ation of similar legislation by this Committee and the full Senate this year.
Improved reporting would certainly aid our law enforcement efforts, but equally
importantly it should also lead to improved performance and voluntary compliance
within the insurance industry. We have spent considerable time working with rep
resentatives of the lending industry to obtain voluntary compliance with civil rights
laws, and have encouraged self-assessment and voluntary corrective action. We con
tinue to sue those who fail to heed our message, but we credit voluntary corrective
action, as indicated by the limited relief that we requested from the Shawmut Mort
gage Company when it took corrective action before we began an investigation.
Many lenders are now using HMDA data for their own benefit. Self-assessments are
being performed, business practices are being changed, and many lenders seem to
be doing a better job in serving minority neighborhoods. To the best of our knowl
edge, most insurance companies do not maintain information in a form that readily
permits an evaluation of their performance in minority neighborhoods.
In closing, I emphasize that we intend to devote considerable resources to the
issue of unlawful discrimination by property insurance companies, and we expect to
be able to better articulate the problems and potential solutions through our litiga
tion program. We agree that a significant problem exists in the industry that is
comparable to the problems discovered in the lending industry. We hope that insur
ance companies, which to this point have fought their inclusion under the Fair
Housing Act, will recognize the importance of the issue. We urge them to evaluate
their own performance and compliance with the law and to take voluntary corrective
action. Such voluntary corrective action will be a very positive factor as we exercise
litigation judgment. Companies should also be on notice that, if they fail to address
the issue voluntarily, the Department of Justice will use its enforcement powers to
49
compel a remedy and the costs likely will greatly exceed the costs of voluntarily
compliance.
I appreciate the opportunity to present our views on this important subject and
look forward to working with the Committee.
PREPARED STATEMENT OF ROBERTA ACHTENBERG
A s s is t a n t S e c r e t a r y
U .S . D e p a r t m e n t
fo r
of
F a ir H o u s in g
H o u s in g
and
and
E q u a l O p p o r t u n it y
U rba n D e v e lo p m en t
Mr. Chairman, Members of the Committee: Thank you for the opportunity to dis
cuss with you one of our Nation’s most pressingcivil rights and housing and urban
redevelopment issues. It is a top priority for HUD to assure the American people
the full protection of the Fair Housing Act (the Act). To do so, we are charged with
providing the same protection from cfiscrimination in the provision of property in
surance that we provide in all other residential real estate related transactions in
cluding real estate practices and mortgage lending.
HUD’s interest in this issue extends beyond its legal responsibility to fully enforce
the Fair Housing Act. Property insurance is essential for the revitalization of this
Nation’s cities. Insurance is required to purchase or improve a home or to start or
expand a business. As the President’s National Advisory Panel on Insurance in RiotAffected Areas concluded 25 years ago, “Communities without insurance are com
munities without hope.” (The President’s National Advisory Panel on Insurance in
Riot-Affected Areas (1968) Meeting the Insurance Crisis of Our Cities, Washington,
DC: The President’s National Advisory Panel on Insurance in Riot Affected Areas.)
Let me take a few minutes to discuss with you our authority under the Fair Hous
ing Act and why discriminatory insurance practices, when tney exist, can be such
a serious problem for urban communities and for the people within them who are
protected by the Act. Then I would like to describe how Secretary Cisneros plans
to move to identify insurance redlining and discrimination and use the full weight
of our enforcement authority to remedy and prevent it. Finally, I will suggest what
can be done to more effectively support the efforts of our agency, community groups,
and the insurance industry to eliminate discrimination and assure fair access to
property insurance.
Insurance and the Fair Housing Act
Discrimination on the basis of race and national origin in the provision of prop
erty insurance is prohibited by the Fair Housing Act oi 1968. HUD and the Depart
ment of Justice have consistently taken this position since the issue first arose in
1978.
Several Administrations, beginning with a HUD General Counsel opinion in 1978,
have concluded that the Fair Housing Act prohibits: (1) insurance redlining and
other policies and practices that deny insurance or make it unavailable on the basis
of race or any other protected status, and (2) discrimination in the terms, conditions,
costs, or other aspects of insurance coverage. Again, in regulations implementing the
Fair Housing Amendments Act of 1988, HUD determined that the Act prohibits “re
fusing to provide . . property or hazard insurance , . or providing such
in
surance differently because of race, color, religion, sex, handicap, familial status, or
national origin*’ (24 C.F.R. Section 100.70(aX4}). And in his recent Executive Order
12892—Leadership and Coordination of Fair Housing in Federal Programs: Affirma
tively Furthering Fair Housing—President Clinton made explicit reference to the
coverage of property insurance discrimination under the Fair Housing Act when he
called Tor HUD to promulgate regulations describing the nature and scope of cov
erage and the conduct prohibited by property insurers under the Fair Housing Act.
The statement offered today by Assistant Attorney General for Civil Rights Pat
rick describes in more detail the consideration the courts have given this question
in the face of opposition to our attempts to investigate complaints we have received.
The Department of Justice, HUD, and those courts considering the issue have con
cluded that insurance is covered by the Act. Dunn v. Midwestern Indemnity Mid
American Fire & Casualty Co., 472 F. Supp. 1106 (S.D. Ohio 1979), McDiarmid v.
Economy Fire & Casualty Co., 604 F. Supp. 105 (S.D. Ohio 1984), NAACP v. Amer
ican Family Mutual Insurance Co., 978 F.2d 287 (7th Cir. 1992), cert, denied, 113
S. Ct. 2335 (1993), Nationwide Mutual Insurance Co. v. Cisneros, No. C3-92-52
(S.D. Ohio Feb. 24, 1994). But see Mackey v. Nationwide Insurance Co., 724 F.2d
419 (4th Cir. 1984). In addition, we do not believe there is a conflict between the
circuits. While the Fourth Circuit held to the contrary, that decision was rendered
prior to HUD’s regulation stating that insurance is covered by the Act. The Seventh
50
Circuit in the 1992 American Family case, in finding that insurance is covered,
found the reasoning of that 1984 decision unpersuasive. The court stated “events
have bypassed Mack e y and found the regulations to be controlling, based upon
HUD’s statutory authority to issue them and the weight such regulations are ac
corded.
The key sections of the Fair Housing Act are 3604(a) which makes it unlawful
to “otherwise make unavailable or deny, a dwelling” and 3604(b) prohibiting dis
crimination “in the provision of services or facilities in connection therewith. Be
cause property insurance is reauired to secure a mortgage loan, which generally is
required to purchase a home, denying insurance makes that home unavailable. As
the Court explained in the American Family case, “no insurance, no loan; no loan,
no house.” I should point out that the Supreme Court had the opportunity to modify
this ruling and declined to do so.
HUD’s authority with respect to discriminatory insurance policy and practices is
identical to other areas subject to the Fair Housing Act. In addition to the investiga
tion of complaints, the Secretary is authorized to initiate investigations and the Sec
retary can issue complaints on his own initiative where warranted. Furthermore,
the Fair Housing Act requires the Department to go beyond enforcement to seek vol
untary compliance on the part of the industry so we can move to prevent discrimina
tion by adopting policies and practices that Dring about fairness in the provision of
insurance. We are authorized to fund private group activity to educate the public
and to assist private enforcement efforts. And, HUD is the only Federal agency
charged with the responsibility to promulgate regulations that define the sub
stantive acts that the Secretary will consider to constitute reasonable cause.
Before I describe to you how we are moving to carry out those responsibilities,
it is important to consider the kinds of policies and practices that are the targets
of our charge.
Risk, Race, and Insurance
The business of insurance is the business of distinguishing among risks and
grouping them in terms of the potential for compensable losses they pose during the
life of a policy. ‘"Fair discrimination”—the concept of treating similar risks simi
larly—is the bedrock principle of insurance underwriting. There is little doubt that
the vast majority of insurers are making conscientious efforts to carefully consider
the perils that potential insurance risks pose when they market their products.
However, we must recognize the possibility that unlawful discrimination can occur
in the insurance market.
Sometimes such discrimination takes an overt and explicit form. More often it is
more subtle and more widespread. Sometimes even policies and practices that are
neutral on their face have a disproportionate racial impact. Some of these may vio
late the law where they cannot meet the established test of business necessity and
the showing that there is no less discriminatory alternative. Below are some exam
ples of analyses by State Insurance Commissioners, consumer advocates, and aca
demic researchers depicting situations which may operate to discriminate and deny
fair access to property insurance to those protected by the Act. In considering the
documented examples that follow, it is imjx)rtant to stress that the finding of a vio
lation occurs on a case by case basis—after full investigation and opportunity for
an administrative hearing or a trial in Federal court.
In some instances, evidence of racial discrimination in the insurance industry can
be overt. A district manager for a Wisconsin insurance company wrote, “Quit writ
ing all those blacks!!” on an agent’s list of life insurance clients. The same district
manager was tape-recorded advising an agent under his supervision, “You write too
many blacks. You gotta start writing good, solid premium-paying white people.” The
insurance company, American Family, is the defendant in a lawsuit filed in Federal
court under the Fair Housing Act by the NAACP, the ACLU, and other plaintiffs
including a member of the City of Milwaukee’s Common Council.
Another documented example of redlining occurred when several agents with the
California Insurance Group provided sworn affidavits to the California Insurance
Commissioner that their company provided them with maps of San Francisco with
the low-income and minority communities outlined in yellow. These maps were ac
companied by instructions not to produce any commercial insurance in these areas.
The California Insurance Commissioner charged the company with 252 violations of
the State’s insurance code. The company subsequently reached a settlement with
the Commissioner in which it agreed to pay a fine of $400,000 and undertake sev
eral reforms to increase its market share in underserved areas.
Discrimination also can occur in marketing or advertising by insurance providers.
It can occur in the pre-application process or in the processing of an application
through the use of discriminatory underwriting guidelines. And not all cases involve
51
denial of insurance. Where insurance is provided discrimination still can occur in
the disparate treatment of applicants based upon their race or the racial character
istics oi their neighborhood.
Paired testing nas been used to uncover instances where a caller from a predomi
nantly minority area will be given very different information than a caller from a
white community with identical financial and other socioeconomic characteristics,
except for the race of the residents. The minority area resident often must call sev
eral times before reaching an agent, may be told an inspection will be necessary
prior to offering a policy, is offered a policy at a higher price for less coverage, gets
referred to a FAIR Plan, never receives a quote in the mail, or is just not offered
any service. The caller from the white neighborhood, on the other hand, will be of
fered a policy on the phone during the first call, is offered a choice of policies and
premiums, and is eagerly and politely solicited as a client.
To illustrate the outcome of such practices, the Missouri Insurance Commissioner
recently found that insurers charged policyholders in low-income minority areas
higher premiums than they did policyholders in low-income white areas for comarable policies even though the losses were higher in the white communities. Resients in the minority areas were paying one and one half times the premium in the
white areas ($7.30 per thousand dollars of coverage compared to $4.65) for inferior
“limited policies,” wnile the loss ratios were 72 percent in the white areas, but just
57 percent in the minority areas.
In addition to discriminatory treatment, seemingly neutral policies can have an
adverse racial impact and may violate the Act. For example, sometimes insurers
have minimum value or maximum age requirements for property that they will in
sure. Homes valued at less than $50,000 or built before 1950 often do not qualify
for insurance, or only qualify for limited policies like basic fire or market value poli
cies rather than full replacement cost policies. These practices have a clear adverse
impact on racial minorities because among owner-occupants in single family dwell
ings, Black households are more than twice as likely as white households (47 per
cent of black households but just 23 percent of white households) to reside in homes
that are valued at less than $50,000. Similarly, 40 percent of black households but
only 29 percent of white households live in homes that were built prior to 1950.
And these may not be isolated policies. According to a 1994 review of underwrit
ing guidelines of major insurers in Texas carried out by the Texas Office of Public
Insurance Counsel, approximately 90 percent of the market in that State is covered
by insurers that have underwriting restrictions associated with the age and value
of homes. (Office of Public Insurance Counsel (1994) “A Review of Homeowners In
surance Underwriting Guidelines Used in Texas,” Austin: Office of Public Insurance
Counsel.)
I want to stress that, as is the case with mortgage lending and all other areas
covered by the Act, a disproportionate adverse impact alone does not constitute a
violation of the Fair Housing Act. Where there is a business necessity and no less
discriminatory alternative, no violation exists.
Current HUD Initiatives
HUD has much to learn about the nature of the insurance industry’s policies and
practices. We intend to move deliberately, but speedily to apply our expert knowl
edge of the Fair Housing Act and the forms discrimination takes to this high prior
ity area. We will utilize all of the measures we have open to us to assure tnat there
is fair, non-discriminatory access to property insurance.
I would like to briefly describe some of these measures to you.
§
I n s u r a n c e U n it
Meeting the commitment both Secretary Cisneros and I made to The President
in the Secretary’s Performance Agreement, last month the Department created and
began staffing a separate unit charged with examining fair housing related insur
ance matters.
One of the first tasks of that unit is to develop regulations implementing the Fair
Housing Act’s prohibition against discriminatory insurancepractices. It has been 25
years since the passage of the Fair Housing Act and HUD has yet to promulgate
regulations that define the lending and insurance policies and practices that violate
the Act. It is high time that we do so.
I n s u r a n c e R e g u la tio n
Several complex issues must be addressed in promulgating the regulation. We in
tend to work in close collaboration with insurance companies, trade associations,
State regulators, civil rights groups, and community organizations. We have already
begun informal discussions with representatives of these interests to learn more
about their views and about issues they would like HUD to address through its rule
52
making. These will continue in the form of group meetings, individual detailed dis
cussions, structured focus groups and consultations on specific issues with insurance
companies, advocacy groups and trade associations.
In addition, I will nold a series of public meetings around the country for indus
try, advocacy groups, and private citizens alike to testify on the rule’s content.
Based on the comments we receive, the evidence presented at the public hearings,
and the written guidance we receive from additional communications with the in
dustry and others, we will publish a proposed rule. Following careful consideration
of the response we receive to that proposal, we will issue a final regulation.
We are considering what must be addressed in the regulation for it to: (1) be effec
tive as guidance to HUD investigators, State and local civil rights agencies, and pri
vate fair housing groups; (2) serve as a guidepost for preventive acts by the indus
try; and (3) be a clear description of the rights afforded protected classes. To do so,
the regulation will address specific practices that are prohibited under the Fair
Housing Act, describe the standards to be utilized in determining whether violations
of the Act occur, discuss investigative techniques that will be utuized, remedies that
will be sought where violations are found, and voluntary affirmative efforts that are
appropriate to eliminate discrimination.
The standards to determine discrimination in this area as in all other covered
areas—will be based on the principles of overt discrimination, disparate treatment,
and disparate impact.
The investigative techniques we consider will grow from our experience, but they
might include statistical analysis of disclosure data, paired-testing, content analysis
of underwriting manuals and other documents pertaining to evaluation of risk and
marketing practices—all employed in lending discrimination investigations today.
A description of appropriate remedies will help guide self-evaluation and correc
tive action by the industry—even where we are not involved through a complaint
investigation. Remedies we will consider would be those appropriate to the insur
ance industry and may include many similar to those in lending discrimination set
tlements secured by the Department of Justice.
Let me emphasize that no conclusions have been drawn, except the fact that
many complex issues must be addressed. We have many questions that we will ex
plore in as open and comprehensive a manner as possible. All relevant parties—in
surance companies and agents, trade associations, regulators, civil rights groups,
community organizations, consumer advocates, and others—will be involved in this
process.
FHIP G r a n t s , T e s t in g , and P r iv a t e E n f o r c e m e n t
For many years HUD has worked in partnership with local public and private fair
housing enforcement organizations. Primarily through our Fair Housing Initiatives
Program (FHIP), we have provided financial support for private enforcement initia
tives that carry out the promise of the Fair Housing Act and will continue doing
so in the future. In fact, the complaints filed with HUD today by the National Fair
Housing Alliance are the product of a FHIP grant.
This fiscal year I have set aside a portion of the FHIP appropriation to fund pri
vate enforcement initiatives that will address discrimination in property insurance.
This set-aside is the single largest amount devoted solely to this enforcement area
in the history of the program.
C o m p l a in t I n v e s t ig a t io n s
HUD’s primary enforcement tool has been the investigation and conciliation of
complaints brought to the agency by individuals who believe their rights have been
violated, fair housing organizations, and others. The Secretary also has the author
ity to initiate a complaint when there is reason to believe that violation has oc
curred or is about to occur and can begin an investigation to determine whether
such a complaint should be issued.
We are dismayed by the continuing efforts of some in the insurance industry to
resist our efforts to investigate insurance discrimination cases. With the decision in
the Nationwide Insurance Co. v. Cisneros, we believe this issue has been settled.
Whether cases are investigated by the Department itself or by State or local agen
cies with laws which are substantially equivalent to the Fair Housing Act, we are
committed to conducting full, fair investigations. We call on the insurance industry
to cooperate with our investigatory processes.
There has not been a significant number of complaints filed involving insurancediscrimination. A total of six insurance discrimination cases are now pending, five
with the Department and one with a State or local agency which has a law that
is substantially equivalent to the Fair Housing Act. Several of these are expected
to result in enforcement action.
53
We do not believe that the number of cases reflects a lack of discrimination; rath
er they are a result of a lack of knowledge regarding the Act’s coverage of such dis
crimination and the difficulty applicants ana homeowners have in knowing when
discrimination has occurred. That is why we commend you, Mr. Chairman, and the
Members of your Committee for these hearings today regarding this important mat
ter. We expect to see a significant increase in complaints filed with the issuance of
further regulations to clarify the application of the Act: the adoption of disclosure
of data requirements through legislation of the kind Congress is considering; in
creased attention to this serious problem and the Department’s enforcement pro
gram by the media; and enhanced financial support for education, outreach and pri
vate enforcement through HUD’s FHIP program. This has been our experience in
the lending area where, over the past 4 years as similar measures were taken, the
number oflending complaints per year has increased 4 times and the proportion of
lending complaints filed each year has doubled.
A HUD/DOJ J o in t I n v e s t ig a t io n T a s k F o r c e
Assistant Attorney General Patrick and I are working together closely to a assure
an effective and coordinated approach on our investigations and our mutual inter
ests in addressing insurance discrimination. As in the case of mortgage lending, the
Secretary of HUD and the Attorney General will join forces to conduct joint inves
tigations where appropriate, combining their distinct powers and authorities to
more effectively address discriminatory behavior.
The Need for Data Disclosure
Our activities have been assisted significantly by the valuable informational hear
ings that Congress has held during the past 2 years on this issue. Your hearing
today no doubt will reveal additional important information.
One issue that Congress has been discussing is a national disclosure bill for insur
ance. A HMDA-like disclosure law for insurance companies, containing race and
gender information on a neighborhood basis, would be an immensely constructive
tool for our agency in its enforcement efforts and for the organizing and local en
forcement efforts of private and public fair housing organizations.
The Administration is pleasea that legislation to prevent redlining in insurance
has been reported by two House Committees. The Administration looks forward to
prompt House passage of anti-redlining legislation and urges favorable consider
ation of anti-redlining legislation by this Committee and the full Senate this year.
As the GAO noted m its recent report, “Property Insurance: Data Needed to Ex
amine Availability, Affordability, ana Accessibility Issues,” (February 1994) impor
tant informational gaps persist. The report stated that information on the number
of properties insured, types and costs of policies, loss data, marketing activity, and
agent location is needea to address availability, affordability, and accessibility is
sues.
We fully support the Administration’s principles for insurance disclosure which
the Office of Management and Budget provided in September. These principles call
for the disclosure of the following:
• race, gender, and income of all applicants and policyholders—this is data similar
to that required by HMDA and is likely to yield the same benefits;
• number of policies on a neighborhood basis; and
• an annual report analyzing this information by the Federal Government as is
done for HMDA data.
In addition, differentiation by policy type and disclosure of loss experience; the re
quirement that insurers provide rejected applicants with the reason for the adverse
decision (HMDA provides for optional reporting of this information); and the report
ing of investment activity aimed at low- and moderate-income communities will add
information to assist our enforcement efforts, better utilize limited resources, enable
industry self-evaluation, and better assure that justice is done when a complaint is
investigated.
The lessons of HMDA are instructive. In the area of lending today we can identify
variations in mortgage loan activity by neighborhood, race, and gender as well as
by income and other demographic factors. Individual application approval and rejec
tion rates can be examined. While HMDA data alone cannot conclusively determine
whether or not a lender has violated the law, it is most useful for targeting inves
tigations and, in conjunction with other information, helping us determine whether
or not a given institution has violated the law.
Similar information for property insurers would be equally valuable information
for law enforcement in this area. Particularly when analyzed in conjunction with
Census Bureau data, we could identify neighborhoods and population groups that
appear to be underserved given their income, value of their property, and other so
54
cioeconomic information. Again, such information alone would not confirm or deny
the existence of discrimination. But it would be useful for targeting investigations,
determining where to file Secretary initiated complaints, and supplying critical in
formation as part of those investigations.
Such disclosure would also provide vital information for local enforcement efforts.
Community organizations which have negotiated $30 billion in CRA agreements in
over 100 communities with lenders, according to the National Community Reinvest
ment Coalition, often rely on HMDA data as the basis for their challenges and their
agreements. Access to HMDA-like information for property insurers may well yield
similar benefits. Given the value of this information to local community organiza
tions and fair housing groups, we recommend that, like HMDA, the data be made
available for eveiy metropolitan area in the United States. The mere fact of disclo
sure will lead at least some property insurers to more carefully consider their un
derwriting and marketing practices. Sunshine is often the best antidote.
HMDA has been an invaluable tool in the area of mortgage lending. It is impera
tive we develop a similar tool for insurance.
Risk, Race, and Insurance, Revisited
Race has long been a central factor in the operation of the Nation’s housing mar
kets. The Fair Housing Act was passed and has been amended for the purpose of
changing this reality. Just as we intend to vigorously enforce the law in the areas
of real estate sales and rental practices and mortgage lending, we intend to do so
in the area of property insurance.
Race has no place in the insurance market. We are eager to work with you to
make this a reality in our urban communities and throughout the Nation.
This concludes my statement Mr. Chairman. I would be pleased at this time to
answer any questions you or the Members of the Committee may have.
PREPARED STATEMENT OF WILLIAM R. TISDALE
P r e s id e n t , N a t io n a l F a ir H o u s in g A l l ia n c e
A c c o m p a n ie d
by:
S hanna L. S m it h
and
C a t h y C lo u d
Members of the Committee, my name is William R. Tisdale and I am President
of the National Fair Housing Alliance located in Washington, DC. I am also the Ex
ecutive Director of the Metropolitan Milwaukee Fair Housing Council. With me this
morning are Shanna Smith, the Executive Director of the Alliance, and Cathy
Cloud, who is the Program Director and who coordinated the homeowners’ insurance
testing project. NFHA is a private, nonprofit corporation representing some 60 pri
vate, nonprofit fair housing agencies located in 35 States, the entire organized pri
vate fair housing movement. In addition there are approximately 40 supporting
members representing State and local civil rights agencies and other organizations
and individuals who support the principles of fair housing.
The National Fair Housing Alliance was founded in 1988. The mission of the Alli
ance is to promote the achievement of ‘the policy of the United States to provide,
within constitutional limitations, for fair housing throughout the United States.”
The Alliance believes that by vigorous, positive, and focused action we can work to
gether to achieve fair housing through outreach, education, litigation, conciliation
and research into the nature, extent, and effects of housing discrimination.
The members of the Alliance are dedicated to working to develop and implement
strategies to reduce, and eventually eliminate, racially and ethnically segregated
housing patterns and to make all housing accessible regardless of race, color, reli
gion, sex, familial status, disability, or national origin.
This hearing examines *Discrimination in the Homeowners9 Insurance In
d u s t r y But before we present the evidence NFHA has developed about this issue,
the National Fair Housing Alliance wants to recognize the leadership of the Senate
Banking, Housing, and Urban Affairs Committee for its long history of examining
discriminatory practices in housing and especially its work on the issues involving
mortgage lending. This Committee did not simply examine the lending institutions,
but has scrutinized as well the practices and policies of the Federal regulatory agen
cies, the private mortgage insurance industry, and the secondary mortgage market.
Because of your diligence and dedication to ensuring fair and equal treatment and
access to credit, more and more people living in minority, integrated and low- and
moderate-income neighborhoods throughout America have a real opportunity to pur
chase housing. Your work will result in the preservation of housmg and improved
quality of life for the millions of families living in these communities.
55
We applaud you, Senator Riegle, for convening this hearing to deal with the discriminatopr practices and policies of the homeowners’ insurance industry. NFHA
has been investigating these problems since 1991, and we will take this opportunity
to reveal publicly for the first time some of the evidence of discriminatory conduct
on the part of two of the country’s largest insurance companies—Nationwide and
Allstate. These companies, we believe, have engaged and continue to engage in prac
tices and policies that intentionally deny, limited, and make unavailable home
owners’ insurance to people living in minority neighborhoods throughout the United
States.
The U.S. Department of Housing and Urban Development provided much of the
financial support for this investigation. Former HUD Secretary Jack Kemp made
fair housing one of the Department’s priorities, and supported fair housing finan
cially through the Fair Housing Initiatives Program (FHIP). NFHA was awarded a
$500,000 FHIP grant, in part, to conduct testing of the homeowners’ insurance in
dustry. In addition, hundreds of hours were donated by NFHA staff and affiliates
as well as staff and affiliates of the National Council of La Raza. Their efforts to
conduct a responsible investigation of the insurance industry have produced compel
ling documentation of discrimination in American cities.
While test reports provided evidence of discriminatory practices from the begin
ning, NFHA did not file complaints during the Kemp administration because of the
serious backlog of complaints at HUD’s Office of General Counsel. At that time, we
also lacked confidence in HUD’s ability to conduct an investigation of this mag
nitude against such a large industiy.
NFHA has now decided to file administrative complaints with HUD because there
has been a substantial improvement in HUD’s ability to handle systemic complaints.
Under the direction of Secretary Hemy Cisneros, HUD’s Office of Fair Housmg and
Equal Opportunity is implementing comprehensive procedures and putting in place
competent, trained staff. NFHA has great expectations that HUD will effectively in
vestigate claims of the magnitude presented in this testing project.
Equally important is Secretary Cisneros’ support of the Federal appellate courts’
interpretation that the Fair Housing Act covers not only disparate treatment, but
disparate impact—not only intentional acts of discrimination, but policies and prac
tices that have the effect of denying, limiting, or otherwise making unavailable
housing, financing, and insurance.
In addition, Attorney General Janet Reno has pledged to use the full extent of
the Fair Housing Act to pursue claims of discrimination. With HUD and Justice
both finally speaking with one strong voice about the law, NFHA will utilize the
system Congress implemented with the Fair Housing Amendments Act of 1988 to
address our charges of discrimination.
Introduction
I. NFHA’s Homeowners’ Insurance Testing Project
In 1990, the National Fair Housing Alliance concluded that, if left unchallenged,
homeowners’ insurance discrimination, like racial steering practices and mortgage
lending discrimination, would have the same effect: the disinvestment, deteriora
tion, demolition, and demise of neighborhoods. In July, 1990, NFHA submitted a
proposal to HUD, in response to a Fair Housing Initiatives Program (FHIP) applica
tion, to conduct testing of homeowners’ insurance companies in selected cities across
the United States. The application was funded, and in October, 1991, NFHA began
work on this project. The HUD funded-portion of the project was completed in early
1993. NFHA used its own resources to undertake the analysis and to conduct addi
tional testing in 1994, some as recently as last week.
Congress passed the Federal Fair Housing Act to combat the individual and sys
temic practices and policies that discriminate against America’s minority and inte
grated neighborhoods. Let’s be perfectly clear: We are going to discuss violations of
Federal law—laws that you enacted to protect residents of our communities from
unlawful discriminatory practices. Congress also passed and authorized funds for
the Fair Housing Initiatives Program (FHIP) to provide direct funding to private,
nonprofit fair housing agencies to conduct testing, investigation, conciliation, and
litigation against entities violating the Fair Housing Act. These congressionally au
thorized and appropriated funds have made it possible for NFHA to conduct this in
vestigation, the results of which we will describe today.
We should be clear that discriminatory practices and policies that deny, limit or
otherwise make unavailable homeowners insurance because of the race, national or
igin, sex, color, religion, familial status, or disability of the individual or the racial
or ethnic make-up of the neighborhood where the property is located VIOLATE the
Federal Fair Housing Act. The issue of insurance discrimination is not new. Com
56
munity organizations, fair housing agencies, and HUD have been looking at this
problem since the 1970’s. In fact, HUD published a handbook in 1979 that stated:
While mortgage redlining has most severely affected lower income and minority
families, the impact of insurance redlining extends to more than just the poor.
Homeowners and investor owners of multi-family dwellings in urban neighbor
hoods are painfully familiar with the practices of non-renewal of policies on prop
erties which they have owned for years. Because of the growing impact insurance
is having on those areas of cities targeted for revitalization efforts, the urgency
of examining the problem more closely is obvious.
“Insurance Redlining: A Guide For Action,” U.S. Department of Housing and
Urban Development, 1979.
Today there has been a renewed focus on lending discrimination, and many people
point out that the movement to combat lending discrimination dates back to the late
1960’s. What is not so well remembered is that the anti-redlining movement began
as a response to insurance redlining before it moved on to lending discrimination.
Public focus on the problems that we are reviewing here today, the types of stud
ies done recently by ACORN, the use of testing, and the move for Federal legislation
in this area, began in the late 1960*8 and early 1970’s as people in minority neighborhoods and people in integrated neighborhoods found insurance companies cancel
ing their policies. After years of organizing around this issue, the National People’s
Action, the largest community-based coalition working on the anti-redlining cam
paign, declared 1979 the year of insurance.
In March of 1979, there was a national conference in Chicago. At this conference,
the President of Allstate announced a new commitment by that company to innercity neighborhoods and a new type of policy and a special program to build the ca
pacity to rebuild urban communities. Within a year, Aetna, Travelers, State Farm,
and several other companies had signed pledges not to discriminate. Aetna even
published an advertisement in many national magazines depicting themselves as a
man eating a dinner of crow to symbolize their admission of past practices.
Community groups engaged in various forms of testing. Indeed, the testing section
of the 1981 guide to fighting lending discrimination developed by the National Peo
ple’s Action quotes extensively from test reports from Nationwide where people in
minority communities were discouraged from obtaining insurance.
The insurance industry has been aware of the problem of discrimination for dec
ades now. Major changes were announced over 15 years ago. But the studies of in
surance discrimination from that time through our own testing project that was
completed just last week demonstrate that the problem is pervasive and persistent.
It has not gone away. We have found what we believe to be evidence of cnscrimination in the very companies that were the focus of anti-discrimination activities over
20 years ago.
n . NFHA’s Testing Process
NFHA designed and implemented a nationwide program of testing for insurance
discrimination. This nationwide approach ensured consistency in the cities in which
testing was conducted. Each site conducted between 30 and 40 matched pair tele
phone tests.
The testing conducted in this program consisted of matched pairs of testers calling
the same insurance office and routinely speaking with the same agent. Neighbor
hood tests were based on the racial/ethnic composition of the neighborhood (minority
tester with property in predominantly minority neighborhood; white tester with
property in white neighborhood). NFHA matched the characteristics of the houses
for which insurance was being sought.. In Chicago, all minority testers were His
panic; in the other cities, minority testers were African-American.
The tester houses were well-maintained homes located in moderate/middle income
neighborhoods. Every effort was made to match neighborhoods based on value of
housing (this was difficult to achieve because the historically dual housing market
has devaluated properties in minority neighborhoods). Most houses were built before
1950. Houses were always matched on type of construction (brick/frame/stucco) and
generally were matched on age and square footage.
IQ. Target Company and Agent Selection
The selection of Nationwide and Allstate was based on bona fide insurance dis
crimination complaints. Sites were requested to provide information about the num
ber and geographical distribution of captive agents for each company. These compa
nies were selected for two reasons:
(1) the egregious nature of existing complaints; and
(2) the preponderance of captive agents in the participant cities.
57
The use of captive agents was important to ensure consistency, because captive
agents generally write policies for only one insurance company. A random selection
process was utilized for identification of specific agents/offices for testing. In smaller
cities, all captive agents were put into a pool and selected at random. In larger
cities, a smaller geographic subset of agents was put into the pool, and agents were
selected at random from that subset.
IV. Testing Results
A. The investigation conducted by NFHA identified the following types of dis
criminatory practices and policies used by agents working for Nationwide and All
state:
1. In minority neighborhoods, refusing to provide insurance because of the age of
the home.
“I don’t like to insure anything over 30 years old.
. It is too hard after all
the remodeling and things like that to mess with. I am trying to eliminate them
from my portfolio.”—ALLSTATE agent for a home in Louisville, Kentucky.
2. In minority neighborhoods, refusing to insure properties because of the market
value of the homes.
“We don't insure for less than $55,000. Well, we can't insure it. Can't do it."—
N a t io n w id e agent for a home in Milwaukee, Wisconsin. This company provided
a replacement cost coverage policy for a similar property located in a white neigh
borhood.
3. In minority neighborhoods, recruiring homeowners to provide the name and
telephone number of tneir mortgage tender to gather information about the property
before providing a quote.
“Who's your mortgage company? Do you have a number for them? .
The rea
son ma'am is because it's like the difference between an Escort and a Cadillac.
[The mortgage company] can tell us what the house is like."—N a t io n w id e agent
for a property in Chicago, Illinois.
4. In minority neighborhoods, requiring inspection of the home prior to providing
information on the cost or type of insurance available.
5. Requiring a credit check for applicants from minority neighborhoods;
6. Charging higher premiums for less coverage for properties in minority neigh
borhoods as compared to similar homes (square footage, age, construction type) in
white neighborhoods;
7. Refusing to provide replacement cost coverage on the structure and contents
on homes in minority neighborhoods;
8. Refusing to return telephone calls of minorities seeking insurance.
B. Based upon a conservative analysis of neighborhood-based test reports, NFHA
can document the following for four cities in which testing was conducted:
1. Louisville: African-American testers experienced discrimination more than 47
percent of the time.
2. Atlanta: African-American testers experienced discrimination more than 60 per
cent of the time.
3. Milwaukee: African-American testers experienced discrimination more than 60
percent of the time.
4. Chicago: Latino testers experienced discrimination more than 95 percent of the
time.
V. Insurance Companies, In General, are Engaging in Practices
and Policies that have the Intent and Effect of Denying, Limiting,
or Restricting Homeowners9Insurance for People Living in
Minority Communities
While Nationwide and Allstate were revealed to engage in specific discriminatory
practices in our investigation, other companies are known to engage in discrimina
tory practices as well. I will now discuss the wide range of discriminatory practices
common throughout the country.
A . M in im u m I n s u r a n c e A m o u n t s
What is the difference between a mortgage lender and insurance company insti
tuting a minimum policy? Nothing. Both practices can have a disparate impact on
people living in minority and integrated neighborhoods. In Ohio, an insurance comany failed to renew a homeowners’ policy on a property that was built through the
[abitat for Humanity program because the company would not insure properties
valued at less than $65,000. HUD and the Federal regulatory agencies recently is
sued a policy statement indicating that there are circumstances in which minimum
g
58
loan policies would constitute discrimination. Minimum insurance policies have the
same discriminatory effect.
B . M a x im u m A g e R e q u ir e m e n t s
Lenders cannot use the age of housing alone to deny a mortgage loan; yet, insur
ance companies are refusing to write policies for homes built as late as 1960 regard
less of the condition of the homes. Maximum age standards strike hardest at our
Nation’s cities, where the housing stock is generally older than in the suburbs. If
the lender’s underwriter and appraiser and the secondary mortgage market believe
that the property is a good investment, why should an insurance company be able
to deny coverage based solely upon the age of the property?
C . C r e d it C h e c k R e q u ir e m e n t s
fo r
S o m e A p p l ic a n t s
Instances have been reported in which minority applicants applying for insurance
in minority neighborhoods were told they would have to have a credit report run
before the agent could provide a quote. This requirement was posed by the agent
without any indication that the testers credit was a problem. All the agent knew
was the address and value of the property, yet the same agent did not tell the white
tester calling about property located in a white neighborhood that a credit check
must be run prior to receiving a quote.
Sound familiar? Mortgage lenders used the credit issue to discourage minority
loan applicants when evidence now clearly demonstrates that 80 percent of all loan
applicants have items on their credit reports that require explanation.
However, when the insurance application is for a minority homebuyer, we know
the mortgage lender and mortgage insurer have already reviewed the credit report
and found tne applicant to be a good credit risk. Should the insurance company be
able to use the same credit report to reject an applicant? If I pay my bills on time
and meet my credit obligations, why should the amount of credit I have determine
whether I receive a homeowners* policy? Why am I a “moral hazard” to the insur
ance company, but a “good credit risk’’ to the lender?
These differences in treatment are similar to the practices used by mortgage lend
ers to discourage minorities from applying for credit, practices now determined to
be discriminatory.
D. I n s p e c t io n R e q u ir e m e n t s P r io r
to
P r o v id in g
a
Qu o t e
fo r
In su ra n ce
It sounds like good business to inspect a property before you insure it. Lenders
conduct appraisals to evaluate a property, but lenders appraise every property. In
surance agents have told applicants from minority neighborhoods that inspections
are required, but the same agents rarely if ever tell applicants from white neighbor
hoods that an inspection is required and virtually never tell them an inspection
must be done before a quote can even be given!
E . Is Y o u r C u r r e n t P o l i c y B e in g C a n c e le d o r N o n -R en ew ed ?
Many agents immediately assume that applicants from integrated and minority
neighborhoods are calling because their policies have been canceled. It is frequently
one of the first questions asked of these callers. If your insurance policy has been
canceled, of course, it is highly unlikely that you wifi be able to get other coverage,
so it is a way of immediately disqualifying the applicant. Callers from white neigh
borhoods, however, are rarely asked about cancellations during their initial contact
with an agent.
Are policies in minority and integrated neighborhoods canceled more often? No re
cent studies are available, but old studies and much anecdotal evidence indicate
that policies in those neighborhoods are canceled more readily for reasons which do
not apply to white neighborhoods. NFHA members, for example, report homeowners
being canceled after 20 years of coverage when they made their first claim.
F . R e p l a c e m e n t Co s t C o v er a g e
versu s
M a r k et Valu e Coverag e
Insurance companies will state that they are writing policies in minority neigh
borhoods. This may be true, but Congress must ask the same questions it asked the
lenders: are these inferior policies, do they provide adequate insurance coverage, are
they priced based upon real or perceived risks, are homeowners offered the oppor
tunity to purchase replacement cost coverage or are they relegated to the Fair Plan
or market value policies?
Insurance companies will claim that homeowners living in lower priced homes in
urban areas, where the houses are older and cost more to rebuild than to sell, create
a “moral hazard” if the company provides replacement cost coverage for their homes.
These companies allege that, if they provide replacement cost coverage, the home
owners have an incentive to burn down their homes to collect the insurance because
the replacement cost substantially exceeds the market value. Can you honestly
59
imagine thousands of homeowners in the neighborhoods pictured in these photos
burning down their homes? Is there proof that minorities and whites living in
homes puilt before 1960 are more likely to burn down their property? Is there evi
dence linking the availability of replacement cost coverage for moderately priced
housing with an increased likelihood that the homeowner will bum down his/her
home? NO!
The Federal Insurance Administration (FIA) challenged this presumption in a
1978 report. An FIA investigation of Detroit found that “[t]here was no evidence
that any policy-by-policy analysis was made to determine whether the low-market
value of a property, in relation to replacement value, had increased the probability
that the owner might resort to arson.” (See Federal Insurance Administration, U.S.
Department of Housing and Urban Development, Insurance Crisis in Urban Amer
ican (1978), at 5.)
Yet, the “moral hazard” reason is raised repeatedly in defense of denying home
owners replacement cost coverage for homes located in minority and integrated
neighborhoods throughout the United States.
VI. The Fair Housing Act Prohibits Insurance Discrimination
There should be no doubt that the Fair Housing Act covers homeowners’ insur
ance under both sections 3604 and 3605. Five out of six Federal court decisions have
held that homeowners’ insurance discrimination is within the purview of the Fair
Housing Act. The most recent decision involves one of the companies NFHA tested:
Nationwide Mutual Insurance Company. Nationwide sued HUD to prevent HUD
from conducting an investigation of allegations of discrimination in Ohio. In Septem
ber, 1993, the court in Dayton, Ohio ruled for HUD. The Toledo Fair Housing Cen
ter, a NFHA member, has also sued Nationwide in State court alleging discrimina
tion against African-Americans living in Toledo’s minority neighborhoods. In addi
tion, the Seventh Circuit Court of Appeals ruled that insurance discrimination is
prohibited under the Fair Housing Act in NAACP v. American Family Mutual Insur
ance Company Co., 978 F. 2nd 287 (1993). The only dissenting decision comes the
fourth circuit in 1984.
F ederal F air Housing Act: Sections 3604 and 3605
Section 3604 makes it unlawful:
(a) To refuse to sell or rent after the making of a bona fide offer, or to refuse
to negotiate for the sale or rental of, or otherwise make unavailable or deny
a dwelling to any person because of race, color, religion, sex, familial status,
(handicap) or national origin.
(b) To discriminate against any person in the terms, conditions or privileges of
sale or rental of a dwelling or in the provision o f services or facilities in con
nection therewith, because of the race, color, religion, sex, familial status, (hand
icap) or national origin.
Section 3605 states:
(a) In General—It shall be unlawful for any person or other entity whose busi
ness includes engaging in residential real estate-related transactions to discrimi
nate against any person in making available such transaction, or in the terms or
conditions of such a transaction, because of race, color, religion, sex, handicap, fa
milial status, or national origin.
(b) Definition—As used in this section, the term “residential real-estate related
transaction” means any of the following:
(A) for purchasing, constructing, improving, repairing, or maintaining a dwell
ing; or
(B) secured by residential real estate.
The phrase “otherwise make unavailable” has been broadly construed to various
practices including practices that result in segregated housing patterns. While the
Act does not explicitly mention mortgage lendmg or insurance discrimination, Fed
eral courts have held that this section prohibits these types of discrimination. (See
e.g., Laufman v. Oakley Building & Loan, (Ohio 1976); Dunn v. Midwestern Indem
nity (Ohio 1979), McDiarmid v. Economy Fire and Casualty Co., (Ohio 1984).
In the American Family decision, the Seventh Circuit Court of Appeals stated:
“The phrase *in provision of services,’ ‘in connection’ with the sale or rental of a
dwelling has been broadly construed to encompass discriminatory practices of insur
ers. Insurance is thus viewed as a ‘service’ that is supplied *in connection with’ the
sale or rental of a house.”
NFHA also contends that section 3605 prohibits insurance discrimination, because
the purpose of insurance is to provide financial assistance to the homebuyer/homeowner “for purchasing, constructing, improving, repairing, or maintaining a dwell-
,
60
mg.” For example, most homeowners rely on insurance to repair a dwelling that
has been damaged by wind, storm, fire, vandalism, or other insurable events; home
builders must have insurance in order to construct a dwelling; homebuyers are
required to have insurance in order to finance the purchase o f their dwelling;
and in order to improve a dwelling through a home improvement loan, the home
owner must have adequate insurance. [See attached Amicus Curiae Brief United
Farm Bureau Mutual Insurance Company v. Metropolitan Human Relations Com
mission, Case No. 93-1739, U.S. Court of Appeals, Seventh Circuit. August 1993.]
VII. Why Are So Few Complaints Filed?
A. W ho R e c o g n iz e s D is c r im in a t io n T oday ?
HUD conducted studies of housing discrimination in the rental and sales markets
in 1989 and found that when African-American and Latino testers inquired about
housing, they experienced discrimination more than 50 percent of the time. HUD
estimated that more than 2 million instances of housingaiscrimination occur annu
ally. How many complaints are filed annually with HUD, State, local, and private
fair housing organizations? Fewer than 20,000. Why? Because discrimination is sub
tle and sophisticated and, without testing, it is difficult to detect. Sometimes it is
simply easier to find other housing than to have to face discrimination.
The Federal regulators have had access to mortgage loan applications for decades,
but only recently have they begun to report evidence of discrimination to the De
partment of Justice. Why? There are numerous reasons including a previous lack
of presidential leadership on the issues of fair housing; failure of Federal regulators
to even acknowledge that discrimination existed; bank examiners without training
about what constitutes a violation of the Fair Housing Act; little credibility in Fed
eral and State civil rights agencies to investigate complaints. .
So even though a handful of neighborhood groups and fair housing agencies have
known and challenged discriminatory insurance practices, there has been little or
no support for Federal or State government to folly investigate fair housing act vio
lations.
Who reads an insurance policy? In informal surveys conducted by NFHA with au
diences that have included real estate agents, mortgage loans officers, neighborhood
residents, and civil rights advocates, it is rare indeed to find more than two people
who have read their homeowners* insurance policy. If an agent states that “it’s the
company policy or practice and this is the best coverage I can provide,” who really
challenges it? You may question the agent, but who complains? Most people simply
call another company to try to get better coverage or simply take the agent at his/
her word. We are reminded of loan officers who told people that the only loan they
could make in minority and integrated neighborhoods required a 20 percent down
payment, Government financing or only a shorter loan term. Most people have no
way of determining if they are victims of insurance discrimination.
B . W hat
has t h e
Go v e r n m e n t D o n e
to
E n fo r c e
th e
L aw s ?
The level of enforcement activity by State and Federal agencies against discrimi
nation in insurance has actually declined over the years. In the late 1970’s and
early 1980’s, several State agencies investigated complaints, did studies, and fined
companies found to be engaging in discriminatory practices. HUD took an active
role, and even issued a handbook on fighting discrimination: “Insurance Redlining:
A Guide for Action.”
But today, the major efforts to combat insurance discrimination remain even more
in the hands of private fair housing groups and community organizations than was
the case over a decade ago. The Government has generally failed to take an active
role in enforcement. Now, with Congress paying more attention to this form of dis
crimination and with new leadership at HUD and the Department of Justice, we
see a heightened interest in the enforcement of the fair lending laws against insur
ance discrimination.
C. T h e R o l e
of
I n fo r m a t io n
As with lending discrimination, one key to increased enforcement lies in providing
the public with information about where companies do and do not issue policies.
Neither enforcement agencies nor members of tne industry can effectively review in
surance availability problems until there is a good source of data defining where
policies of different types are and are not made. Just as the Home Mortgage Disclo
sure Act data has lead to both an increased level of enforcement by many parties
and to self-review and reform by some industry leaders, so insurance disclosure
holds the potential to be a catalyst for enforcement and reform in insurance dis
crimination. At its best, such data define cases of potential discrimination for testing
61
or direct litigation. At a minimum, such data provide a focus for discussion about
what constitutes a legitimate business practice.
D. E d u c a t io n
and
Ou t r e a c h
While HUD and private fair housing groups have engaged in extensive media
campaigns to teach individuals how to recognize and report discrimination in rental
and sales practices, these same efforts have not been duplicated in the areas of lend
ing and insurance. Because discrimination is so subtle and sophisticated, people
need to be taught how to identify suspicious actions.
V lil. Recommendations
The National Fair Housing Alliance makes the following recommendations to ad
dress discrimination in the homeowners* insurance industry:
A . D ir e c t iv e fr o m C o n g r e s s t h a t F u n d s in F e d e r a l P r o g r a m s b e
A p p r o p r ia t e d f o r E d u c a t io n , E n f o r c e m e n t , and R e s e a r c h C o n c e r n in g
H o m eo w n er s* I n su ra n c e
1. Fair Housing Initiatives Program (FHIP) Funds: Congress can begin by increas
ing the 1995-96 FHIP allocation in enforcement from $9 million to $15 million or,
at the least, reallocating $3 million from education to enforcement. It makes little
sense to increase the knowledge of the public about housing, lending, and insurance
discrimination without providing adequate funds for enforcement efforts. There
must be adequate funds in enforcement to cover the costs of systemic investigations,
analysis, ana expert witness costs.
2. CDBG Funds: While more than 800 cities in the United States receive CDBG
funds, fewer than 30 cities provide funding for enforcement of the Federal Fair
Housing Act. The CDBG program requires each city to “affirmatively further fair
housing.” Those cities that do not provide funding for fair housing may not, in fact,
be fulfilling this mandate. Congress should support HUD*s recommendations to es
tablish a separate funding line item for fair housing education and enforcement ac
tivities. In addition, HUD must establish standards that define affirmatively fur
thering fair housing** which should include comprehensive education, enforcement,
and research components. At a minimum, these should address the application of
the Fair Housing Act in rental, sales, lending, and insurance issues for all protected
classes, regardless of income.
3. Congress should allocate additional funds for HUD*s Fair Housing and Equal
Opportunity Division to hire and train staff specializing in systemic type investiga
tions.
B . L e g is l a t io n : C o n g r e s s S h o u l d E n a c t
a
D is c l o s u r e B il l
1. Insurance discrimination is a civil rights issue. HUD has primary authority
for enforcing the Fair Housing Act; therefore, HUD should receive the disclosure
data, have resources to analyze and investigate patterns of disinvestment and make
the data available to the public. The Senate House, Banking, and Urban Affairs
Committee should monitor the fair housing elements of the insurance industry in
the same affirmative manner that it has monitored the mortgage lending industry.
2. Congress should enact disclosure legislation which will provides, at a minimum,
the following:
a. Disclosure of Underwriting Guidelines
b. Disclosure of Loss Data
c. Disclosure of type of policy, cost of policy
d. Reporting of the race, national origin and gender of policyholders
e. Reporting for all Metropolitan Statistical Areas
f. Reporting the above information by Census Tract
C. W hy
are t h ese
E lem en ts N ec essa ry?
1. Access to Underwriting Guidelines
We must have access to underwriting guidelines. Several years ago, mortgage
lenders claimed their underwriting guidelines were “trade secrets” and they claimed
they would lose their competitive edge if forced to reveal the guidelines. The insur
ance industry is making the same claims now. What has happened since the lenders
made their underwriting standards public? Better underwriting policies and prac
tices are being put into place. Antiquated and discriminatory guidelines were identi
fied and removed. Souna lending in urban areas is underway in many cities. Legiti
mate underwriting guidelines can be defended. No one is insisting that insurance
companies write policies for people who burn houses, inflate claims, file false claims,
or commit other illegal acts. No one is asking an insurance company to insure a
property that is a fire or safety hazard. What tne Fair Housing Act requires is poli
8 4 -0 5 1 0 - 9 4 - 3
62
cies that do not have a discriminatory effect or impact on a person based upon race,
color, religion, sex, familial status, disability, or national origin—or against a neigh
borhood because of the race or national origin of the residents.
2. Loss Data
The industry must present information about the number, type, and amount of
claims filed. Without this information, Congress and the public will not know if
higher premiums charged in minority, integrated, older, or lower income neighbor
hoods are based upon higher risks or whether these high premiums are being used
to subsidize other neighborhoods, as some studies reveal.
3. Disclosure of Type and Cost of Policies
Just as it is important for us to know if conventional loans are available in all
neighborhoods, we need to know what types of policies are being written in minority
neighborhoods. Certainly insurers will come forward with numbers showing they are
writing policies in some of the same neighborhoods where we have documented dis
crimination, but do these policies include their top of the line packages or are they
minimum insurance at maximum price? Remember when lenders made loans in mi
nority neighborhoods, but the terms and conditions were more restrictive, not based
on risk, but based on race. The insurance industry must provide documentation that
their business decisions are based on risk and not race. This provision will give gov
ernmental agencies, fair housing advocates, and neighborhood groups that informa
tion necessary to determine if fair treatment is reality.
4. Reporting Race, National Origin, and Gender
This information is critical and easy to obtain. Just as mortgage lenders record
the information or have the loan applicant complete the section on race, insurance
companies can include this information on their application. If they are uncomfort
able asking the applicant over the telephone, they can send a form for the applicant
to complete and return in one of the companies regular mailings to the policyholder.
People are not offended by the question when they understand that the information
is being used to guarantee equal treatment. The Fair Housing Act was not passed
to tell minorities, women, disabled persons that they have rights, it was passed to
tell the individuals and companies who own and manage housing, lending and in
surance that these people and the neighborhoods where they live have rights under
the law.
5. Reporting Information By Census Tract
Currently, insurance companies keep information by zip code. Zip codes are large
geographic areas that encompass many minority and non-minority neighborhoods.
Many zip codes are so large that they actually take in very high-income white
neighborhoods, moderate-income integrated and minority neighborhoods as well as
very low-income communities. An insurance company could report that it is writing
20 percent of the policies in the zip code, but that 20 percent could be confined to
the high-income white neighborhood. Census tracts usually have 5,000 people within
their boundaries and provide demographic data that is essential to determining the
characteristics of neighborhoods such as race, income, and age of housing. Census
tract reporting is required of mortgage lenders, and Congress gave them 1 year to
convert from zip code to census tract after passing the Home Mortgage Disclosure
Act. It is certainly much easier and less expensive now to convert. However, I have
no doubt that insurance companies will argue that the expense is enormous and will
be passed on to the consumer in higher premiums. Congress listened to this same
argument from the lenders, and the lenders who went out of business did so because
of their failure to provide safe and sound loans, not because of a burdensome ex
pense of reporting loan information by census tract.
6. Reporting for ALL Metropolitan Statistical Areas
Insurance discrimination is a violation of Federal law. This is a civil rights issue
and Congress has the responsibility to assist HUD, Justice, and the public in insur
ing fair enforcement of the law. Tlie Federal Fair Housing Act provides protection
based upon race, color, religion, sex, familial status, disability, or national origin.
It also protects people who live in minority and integrated neighborhoods. Clearly
the MSA in the United States include people and neighborhoods represented in the
protected classes. How can we justify protecting some, but not all, of the residents
in the country? Reporting must be inclusive.
Conclusion
It is important that Congress take swift and comprehensive action to address dis
crimination in the homeowners* insurance industry. For more than 20 years, the
63
mortgage lending industry claimed that denial of loans in minority neighborhoods
was based upon sound lending practices. We are confident that insurance companies
will claim they are insuring risk, not race. But we believe the evidence disclosed
to you today is simply the tip of the insurance discrimination iceberg. Congress
must stand firm with this powerful and wealthy industry. America’s neighborhoods
are counting on you to provide HUD and the public with the tools necessary to iden
tify and eliminate discrimination in all forms.
PREPARED STATEMENT OF J. ROBERT HUNTER
C o m m is s io n e r , T e x a s D e p a r t m e n t
of
I n su ra n c e
Insurance Redlining
Mr. Chairman and Members of the Committee, I am pleased to appear before the
Banking Committee once again. When I was Federal Insurance Commissioner, I
spent many long and occasionally enjoyable hours here before Chairman Proxmire.
More than 25 years have passed since President Lyndon B. Johnson’s Commission
on Insurance Availability in Urban America, formed in the aftermath of Los Angeles
rioting, told us that “communities without insurance are communities without
hope. As Federal Insurance Commissioner, I ran the riot reinsurance and FAIR
Plan programs which sprang from that commission’s studies. During the 1970’s, we
at the Federal Insurance Administration performed several studies documenting the
fact that some insurers were avoiding certain neighborhoods. One of our reports
showed that in New York City, residents were more likely to be denied homeowners’
insurance if they were black than if they had building code violations.
Today, we still find insurance companies making underwriting decisions based on
all kinds of factors that have nothing to do with a statistically measured or measur
able probability of risk. One of these factors, unfortunately, is your location on a
city map that probably does not have any red boundary lines drawn on it but it
might as well because the results are the same. I commend this Committee and the
Members of Congress who have introduced redlining legislation for their recogni
tion—and rejection—of this antiquated form of underwriting and their determina
tion to do something about it.
Definitions
In your deliberations on these bills, you will hear and see the word “redlining”
over and again. I think it behooves each of us to be clear about the meaning of this
emotionally laden word when we use it.
My definition of redlining is simple. By redlining I mean unfair discrimination in
the availability, price, benefits, or quality of insurance for a class of consumers
based on factors outside the control of the consumer.
Redlining is not only geographic; it includes unfair discrimination based on race,
gender, age, income level, value of home, age of home, or other characteristics that
the consumer cannot change.
By this definition, redlining occurs when a class of individuals is denied insur
ance, charged a higher price, provided fewer benefits or given inferior service for a
reason unrelated to their risk or for a reason that is contrary to public policy.
Data indicate that unfair discrimination against several classes of individuals,
particularly minorities and low-income citizens, is practiced all too commonly in the
insurance industry, denying many consumers the ability to purchase cars or homes
or maintain small businesses. The Federal Fair Housing Act, as interpreted by the
Supreme Court, is an example of the Federal interest in ensuring that redlining
does not prevent classes of consumers from purchasing a home. Denying insurance
to citizens and small businesses in economically underdeveloped areas contributes
to the web of inadequate economic opportunity and social decay. There was a tele
vision investigative report a number of years ago called “The Poor Pay More.” We
see the same result in insurance. The channeling of low-income people into highrisk, high-rate insurance companies is redlining.
I want to make it clear that I am coming at this problem not only as a State offi
cial with a consumer protection mandate and a personal history as an insurance
consumer advocate. I also approach redlining as an actuary with experience in risk
analysis and insurance loss projection. Underwriting standards that take a mono
lithic approach to neighborhoods, home value and age of home just don’t make
sense. They treat well maintained, structurally sound, and burglar-resistant homes
the same as fire traps.
64
When I conducted a public heading on redlining in Houston on March 31, I re
ceived a clear-cut example of this approach to underwriting. Habitat for Humanity,
an organization associated with former President Jimmy Carter, has built 50 homes
for—and, of particular importance, with—low-income families in Houston. This is
sweat equity, and the new owners have not only worked on the construction but
learned a great deal about home maintenance. These homes and their owners are
quality risks. This group of homeowners in Houston have made only one claim—
for a tree that fell during a storm. Nonetheless, only one major insurance company,
Texas Farmers, was willing to insure these homes. A transcript of the Habitat for
Humanity representative’s testimony is shown as At ta c h m en t A.
The approach taken by other companies to the question of insuring these Habitat
for Humanity homes was not genuine underwriting. Rather, it was the application
of certain preconceptions about how people across the tracks live their lives and
tend their homes. So when I urge action, including voluntary action by insurance
companies, to end redlining, I’m not talking social engineering but I am talking the
use of sound business practices that separate profitable from unprofitable insurance
business one risk at a time.
Data Collection
How do we begin to attack redlining? The essential first step is to identify all
areas that are being unfairly discriminated against in the availability, price, bene
fits, or quality of insurance. To solve the problem of redlining, all areas where con
sumers are being unfairly discriminated against must be identified.
This requires reliable data broken into small enough pieces to see what is happen
ing to specific neighborhoods and other specific groups of consumers. I would like
to see data that tells what is happening in rural areas as well as in cities, and I
would like to have urban data all the way down to the census tract level.
Certain lines of inurance—homeowners, automobile, and small business commer
cial policies—are most susceptible to redlining, and those are the lines where the
need for good, reliable statistical data are most crucial.
The kinds of statistics we need include data about service (including locations of
insurance agents), coverages sold in an area, premium volume, prices charged, and
losses (including loss ratios).
Texas has been trying for more than a year to assess the degree to which redlin
ing occurs in our State. We have used data calls, a consulting actuary’s study and
the more anecdotal route of public hearings. This is an ongoing effort to come up
with the sometimes elusive truth about whether the industry does things its spokes
persons deny it does. It is a pity that Texas has taken so long to identify the prob
lem. It is doubly tragic that 1 must report that Texas is at the cutting edge of data
analysis on this issue.
The paragraphs that follow will summarize some of our preliminary findings in
Texas.
Evidence of Redlining—the Texas Experience
One place we looked for evidence of redlining was the placement of drivers in the
Texas Automobile Insurance Plan (TAIP), which operates as our State’s assigned
risk plan for drivers who have been rejected for coverage by insurance companies
in the voluntary market. It is noteworthy that the TAD5 offers only liability cov
erages, not comprehensive coverage that pays when cars are stolen or vandalized.
Even if one assumes—as I do not—that low-income or high-minority neighborhoods
are by definition high-crime neighborhoods, this should not be a reason for sending
drivers to the TAIP.
Studies of TAIP assignments in 1993 show that consumers who live in zip codes
with predominantly low-income and minority populations are disproportionately in
sured through the TAIP compared with those from zip codes with higher-than-average income and higher-than-average Anglo populations. Rural consumers also are
disproportionately represented in the TAD3.
TATP assignments are one indicator of auto insurance availability in particular
neighborhoods. Statistics gathered through this year’s NAIC call to insurers
writing auto insurance m Texas also showed a direct correlation between
lack o f availability and the ethnic and racial minority percentages o f the
population in the zip code. The data presented in ATTACHMENT B graphically por
tray the existence of auto insurance redlining. The data show that the higher the
minority population in a zip code, the worse the auto insurance availability. In addi
tion, lower availability of auto insurance correlates with lower median household in
come in a zip code.
65
• In zip codes where auto insurance availability is two times worse than the State
average, the minority population percentage in the zip code also is twice the State
average; and
• In zip codes where auto insurance is written in non-standard companies half as
much as the Statewide average, the minority population percentage is also half
the Statewide average.
The data also show a strong correlation between low median household income
and low availability of insurance. I can supply a copy of these data in full zip code
detail to this Committee if you desire it.
The Texas Department of Insurance is doing a similar analysis of homeowners’
insurance by zip code, but we do not have any preliminary results at this time.
I cannot overemphasize the importance of gathering thorough and reliable statis
tics on the insurance marketplace. Like the NAIC, Texas’ efforts in this direction
are just beginning. Texas is a pioneer in gathering insurance data independently of
industry-controlled statistical organizations, and we intend to have a continuing
flow of zip coded data on automobile, homeowners, and commercial insurance. We
have just scratched the surface in our ability to detect redlining and other unfairly
discriminatory insurance practices, and this data flow will put us in a much better
position to protect the insurance-buying public in the future. To have any hope of
a national picture in the foreseeable future, we must have congressional action. In
surance is, I fear, just too powerful a special interest in most States.
Underwriting Guidelines
We are looking not only at evidence of redlining, We are reviewing companies’ un
derwriting guidelines to determine the causes. Several common underwriting guide
lines adversely affect the availability of homeowners’ insurance in minority and lowincome neighborhoods as well as in some rural and inter-urban communities. Texas’
Office of Public Insurance Counsel (OPIC) reviewed the homeowners underwriting
guidelines filed with our Department by insurance companies this year and issued
a report (ATTACHMENT C) showing that:
• 88 percent of the companies have age-of-home restrictions.
• More than 90 percent of the companies have minimum coverage amounts, often
above Texas’ median housing value of $42,500. Many won’t write homes valued
under $60,000 or higher!
• 60 percent of the companies have location restrictions. While not as overt as a
red line map, these restrictions prohibit coverage for certain consumers in very
vague terms without objective standards (e.g. “unprotected areas”).
All of these underwriting guidelines have an adverse impact on the availability
of homeowners’ insurance for consumers in older or lower income inner-city and
rural neighborhoods.
Regulatory action against unfair underwriting practices is a new phenomenon but
I predict you will see much more of it as the trend toward election or appointment
o f consumer-oriented insurance commissioners continues throughout the Nation. In
Texas, the Commissioner received clear-cut legislative authority only last year to re
quest and receive companies’ underwriting guidelines and to use them in enforce
ment actions (but we are not free to disclose them to the public, which inhibits out
ability to make competition in price fully effective—we need an informed consumer
for that). Our Department just initiated disciplinary actions against 59 companies,
including some of the largest auto writers, for alleged unfairly discriminatory under
writing guidelines, which either exclude people or force them into high-rate compa
nies merely because they are single, have only one car, have a drivers license from
the “wrong’’ country, won’t buy another kind of insurance policy from the same com
pany or had been rejected or canceled by a different company.
On May 2, 1994, I imposed a Texas record fine of $850,000 on Allstate for apply
ing similar guidelines. In this case, we actually received videotapes showing agent
after agent turning down an applicant because he was single, had only one car and
was not in the market for any insurance but an auto policy.
Although the practices involved in these disciplinary cases do not involve exclud
ing specific neighborhoods, they do have disproportionate impact on racial and eth
nic minority populations and low-income consumers. That is why the focus on red
lining should not be limited to discrimination based on geographic location alone.
Besides the disciplinary actions mentioned above, I am considering further rulemaking, enforcement actions and proposed legislation to further combat unfair un
derwriting practices and redlining. One of the ideas Fm considering is requiring that
an underwriting guide be demonstrated as risk related by statistics in order to use
it in Texas.
8 4 -0 5 1 0 - 9 4 - 4
66
In your letter of invitation to testify at this hearing, your staff asked me to ad
dress several specific issues related to various data collection bills now pending in
Congress. These are (1) reporting by census tract versus reporting by zip code, (2)
the collection of race and gender data, (3) reporting on claims information, and (4)
including an automatic sunset provision with the reporting requirements. I will dis
cuss each of these in order.
Reporting by Census Tract Versus Reporting by Zip Code
Although current reporting of neighborhood-related insurance data is by zip code
in the vast majority of States, it is not sufficient for your purposes. Congress would
come closer to getting the information it needs on redlining n it requires reporting
by census tract. Demographic information from the Bureau of the Census is far
more complete and accurate for census tracts than it is for zip codes. In addition,
census tracts tend to be smaller and more demographically homogeneous than zip
codes, thereby giving a clearer, better focused picture of the treatment of minority
and low-income neighborhoods than we get from looking at zip code data. Further
more, zip codes exist to expedite mail handling and are subject to change whenever
necessary to further that objective. Census tract boundaries are less likely to shift.
I believe the larger insurance companies that would object to reporting could easily
adapt to census-tract reporting because the necessary software is readily available.
In fact, section 14(b) of S. 1917, the legislation now under review by this Committee,
requires the Secretary to provide insurance companies with this software. If insur
ers claim that census tract is too difficult, you might consider requiring reporting
by 9-digit zip code, which can, as I understand it, be used to construct census tracts.
All insurers must have this information in order to obtain mail cost savings.
Collecting Race and Gender Data
I support the collection of race and gender data because it will further the objec
tive oi discouraging unfair discrimination, including those forms of discrimination
that have the intent and/or effect of redlining. This kind of data collection is essen
tial if Congress is to determine whether insurance companies are, in fact, discrimi
nating based on race or gender. It will also provide stronger factual support for leg
islative, judicial, and regulatory actions to protect the rights of racial ana ethnic mi
nority populations to buy insurance at fair and affordable rates. Finally, and pos
sibly oi greatest importance, such data might increase the availability of insurance
to minorities by opening the eyes of many insurance companies to the effect of their
underwriting guidelines on these populations. Such companies could be expected to
act to increase their minority busmess to avoid discrimination suits, regulatory ac
tion, and further legislation.
At my March 31 redlining hearing in Houston, an ITT Hartford representative
testified the company had gone through a painful process in which it came to realize
its underwriting practices were causmg unintended problems for some minorities.
The company is taking initial steps to turn this around, including a program to de
velop more minority agents to work in underserved areas. Thus, by forcing compa
nies to examine the effects of their underwriting guidelines on minorities and lowincome consumers, we hopefully will begin the process of their own self-examination
and reform.
Reporting of Claims Information
Claims information is vital if Congress, regulators, and others are to accurately
assess the extent of redlining and otner forms of unfair discrimination in the sale
and pricing of insurance and take appropriate corrective action. What if you collect
data from an insurer and see what appears to be unfair discrimination based on
where it wrote its policies? What if it tnen says it is making a sound business judg
ment based on the claims? Where are we then?
Claims data are highly relevant information, and we seek it in our own data calls
because assuring fair rates is only half the regulatory battle. If a homeowner in a
predominantly minority area pays the same rate as one in an upscale corner of the
same city but receives scaled-down benefits if he or she has a fire or other loss, that
is redlining as surely as rejection for coverage or assignment to a high-risk com
pany. This, in fact, might very well be happening in some of our cities. Public ad
justers have told our Department that some insurance companies have different
payment standards in minority areas than in other areas. Trial lawyers and some
adjusters also have alleged that insurers pay smaller damages for minority claim
ants. But, as an insurance regulator, I need solid statistical support for rulemaking
and enforcement actions against such behavior, and I believe Congress also needs
more than hunches and anecdotal evidence. You need these data!
Finally, enlightened insurance regulators and companies are coming to under
stand that they have an obligation to attack the root cause of rising insurance rates
67
by encouraging better construction, stronger defenses against crime and other lossprevention measures. The Texas Department of Insurance recently created its first
Safety Unit to lead the way in this effort. Claims data, by neighborhood, can help
regulators and the industry target areas where loss prevention efforts can do the
most good.
Automatic Sunset Provision
There should be no automatic sunset provision. The purposes of the bill set out
in section 2 will continue to exist in the future. If the bill makes sense now, it will
make sense 10 years from now, and Congress would be remiss to pretend that red
lining is a disease that can be cured in a short period of time with no danger of
remission.
Review of S. 1917
Finally, the Committee staff asked for my review of Senator Feingold’s bill,
S. 1917. Following are some problem areas and my recommendations for cnange:
1. The bill does not include the collection of data on personal auto insurance. Auto
insurance is as important, if not more important, than homeowners* insurance in
minority and low-income neighborhoods where home ownership is more of a dream
than a reality. Our data show significant redlining in auto insurance. The bill
should include collection of data on auto insurance.
2. In addition to region, race, gender, age of home, and location of home, data on
the age of the insured should be collected for homeowners and auto.
3. Section 12(aX2) should be changed to make data available to the public as soon
as possible. I suggest that the agency be allowed time to review the data but that
a preliminary report should be issued 30 days after the data is due from insurers
and a final report issued 60 days after the data is due from insurers.
4. Section 12(c) makes losses by individual insurers by zip code or census tract
a closed record. To combat unfair discrimination in claims payments, it is crucial
that this information be open to the public.
5. Renters* insurance is excluded from the definition of residential property insur
ance in section 13(c)—designated lines of insurance. Because many low-income and
minority consumers are unable to purchase homes but need insurance for their per
sonal property, data should be collected on renters insurance.
I should point out that S. 1917 is stronger than the House Commerce Committee
version of the bill in that body, but far weaker than the Banking Committee version.
It represents a good compromise, in my view, if the items I listed above are amend
ed onto S. 1917.
Conclusion
Based on available data, it is obvious that redlining is a problem in many urban
communities and rural areas. Other forms of discrimination, such as underwriting
uidelines that deny coverage or assign consumers to high-risk, high-rate companies
ecause they are single or own only one car or don*t buy other kinds of insurance
from a company often have the same effect on minority and low-income people as
geographic redlining. More data is necessary if policymakers and insurance regu
lators are to fully understand the scope of this problem and take the actions nec
essary to protect the insurance-buying public. Congress is to be commended for tak
ing on this difficult but extremely important issue. I urge you to strengthen and ap
prove S. 1917.
f
PREPARED STATEMENT OF LYNN M. SCHUBERT
A s s i s t a n t G e n e r a l C o u n s e l, A m e r ic a n I n s u r a n c e A s s o c ia t io n
I. Introduction
The American Insurance Association is a national trade organization representing
more than 270 companies writing property and casualty insurance in every State
and jurisdiction of the United States. AIA members write 36 percent of all commer
cial property and casualty insurance in the United States. They also write a signifi
cant amount of personal, homeowners, and automobile insurance. AIA member com
panies employ more than 145,000 people and pay $2.2 billion in State taxes and fees
(including payroll taxes) to State governments each year.
We appreciate the opportunity to be here today to discuss the important topic of
access to insurance for urban residents and businesses. I have been authorized to
present this testimony not only on behalf of the AIA, but also on behalf of the Inde
pendent Insurance Agents of America (IIAA) and the Council of Insurance Agents
68
and Brokers (The Council). IIAA is a trade association representing nearly 300,000
independent insurance agents. The Council (formerly the National Association of
Casualty and Surety Agents, NACSA) represents the Nation’s 300 largest commer
cial insurance agencies and brokerages who write over $70 billion in premiums an
nually.
These organizations have one over-riding goal on the issue of insurance access
and availability Chairman Riegle: Urban residents and businesses, as all other
Americans, must be able to purchase attractive insurance products at a price rea
sonably based on the risk. AIA, IIAA, and The Council members are committed to
working with legislators, regulators, consumers, and brokers to ensure that this oc
curs.
It is a fact that certain areas—especially in our cities—have higher numbers of
residents and small businesses without insurance than other areas. These discrep
ancies are related to a whole host of socio-economic circumstances faced by people
who work and live in the urban areas, and which also increase the cost of insurance.
Outright racial or ethnic discrimination may also occur. We hope that it is infre
quent. We know that it is a violation of Federal and State law. We advocate strin
gent prosecution wherever it is found.
To determine whether or not insurance is available in all areas, we could support
Federal data collection of insurance information—if it is designed to produce a fair,
efficient, and effective study of insurance availability and cost. H.R. 1188, pending
in the House, is such a measure, and AIA, IIAA, and The Council support this bill.
The details of the bill and the reasons for our support are addressed later in this
testimony.
If there are availability problems, however, whether they are caused by economics
or by discrimination, they must be addressed. It is cold comfort to the citizens for
whom insurance is unavailable to explain to them the reasons why the problem ex
ists. We believe it is time to start attacking these problems head-on.
We encourage and will participate in thorough and thoughtful dialogue on the
subject of urban insurance coverage with the goal of developing workable solutions
to the problems. Specifically, we would like to work with legislators and regulators
to develop products and marketing ideas which adequately serve urban residents
and businesses. We want to discuss the establishment of market assistance plans
and other programs to facilitate greater access to insurers. For example, we already
are working with regulators in a number of States on programs to bring urbanbased independent agents together with companies.
One State which is moving forward on this issue is Georgia. In Atlanta, AIA, in
conjunction with the Urban League, the Independent Insurance Agents of America,
the Insurance Department and State regulators, has established a task force to ad
dress the issue of insurance in urban markets. One of the first projects of the task
force is the Agent/Insurer Partnership. The task force has put together a directory
of profile forms completed by minority agents and has distributed it to standard in
surance companies. These companies have reviewed the profiles, and are in the
process of making agency appointments from the directory. While the process is not
complete, a number of appointments already have been made. Now that this pro
gram has proven successful, we are working to expand these efforts nationwide.
In addition, we also support the inclusion of homeowners’ insurance in existing
Fair Access to Insurance Requirement Plans (FAIR) to address the needs of home
owners who are unable to find insurance in the voluntary market.
In California we publicly have supported a proposal which would provide for
Statewide rating of basic first party insurance in the context of a cost effective no
fault system. We believe this system should be considered in other States where
auto insurance costs are too hign, especially for low-income urban citizens. We will
continue to explore other options along these lines for other lines of insurance.
Allegations have been made that insurance is unavailable in urban areas due to
insurance redlining. Further, this Committee has asked us to address the question
of whether insurers illegally discriminate in providing property insurance. Before
any further discussion of the positive efforts of the industry in the areas of insur
ance availability and cost let me address the specific question posed by the Commit
tee. To answer this question we need to staxt with a discussion of the concept of
redlining.
II. Definition of Redlining
Redlining is an illegal offensive practice that cannot be tolerated. AIA opposes
redlining. The term has long referred to an attempt to discriminate on the basis of
race or ethnic origin by not doing business within certain “redlined” neighborhoods.
The practice is illegal, reprehensible, inexcusable, and must not be tolerated. Viola
tors should be punished.
69
Defining the term, however, in order to identify the specific activities that con
stitute the practice of redlining, has proven a difficult task. Redlining is an emotion
ally charged term that connotes different things to different people and, as such, has
defied definition in a universally accepted manner. In recent years insurance regu
lators and industry representatives have spent many hours debating the issue and
hammering out language describing and prohibiting certain practices that constitute
redlining. This language, embodied in the National Association of Insurance Com
missioners (NAIC) Model Unfair Trade Practices Act, defines and prohibits the fol
lowing as unfair discrimination:
Making or permitting any unfair discrimination between individuals or risks of
the same class and of essentially the same hazard by refusing to insure, refusing
to renew, canceling, or limiting the amount of insurance coverage on a property
or casualty risk solely because of the geographic location of the risk, unless such
action is the result of the application of sound underwriting and actuarial prin
ciples related to actual or reasonably anticipated loss experience.
We strongly endorse this provision. We have supported its adoption by the NAIC
and its passage in State legislatures across the country. The language clearly pro
scribes arbitrary underwriting decisions based upon geographic location alone. It
recognizes, however, that sound underwriting and actuarial principles cannot be ig
nored by the industry—insurers must make rational underwriting decisions that
will preserve their solvency and protect consumers.
AIA has a clear record of opposition to the arbitrary reliance by insurers upon
physical location alone when rendering underwriting decisions. Throughout our his
tory we have consistently stated our belief that insurance should be readily avail
able, subject to fair and sound underwriting principles. The first step to making in
surance available is the development of products tnat are attractive and affordable
for urban residents and businesses.
m . The First Step Toward a Solution
We must focus our attention immediately on a fundamental issue, i.e., the rapid
development and approval of attractive and affordable insurance products tailored
to meet the needs of urban consumers.
In California, we have joined forces with a broad coalition of consumer and civil
rights groups led by Consumers Union and the Latino Issues Forum to create an
innovative, no-frills no-fault automobile insurance policy. This policy is a perfect ex
ample of a creative, practical, and sound solution to a real problem o f people in
need.
The policy is designed to be offered Statewide at one low price. It guarantees all
injured accident victims at least $15,000 in medical and wage loss benefits for a uni
form Statewide price, and removes nuisance and minor injury litigation for non-economic losses. Independent actuaries have concluded that the basic coverage could
be sold, on an actuarially sound Statewide basis, for $220, saving California con
sumers $1.8 billion in the first year alone.
Unfortunately, this product is not yet available to California consumers. During
recent sessions of the California legislature, the bill that would pave the way for
provision of this product in the marketplace has been bottled up.
The National Conference of Insurance Legislators recently has approved a model
automobile insurance law providing $15,000 of basic personal compensation cov
erage and property damage liability coverage, along with limits on lawsuits and
health care costs. AIA estimates the standard price for the minimum mandated cov
erage would be $117 in Vermont, $146 in Missouri, and $199 in Texas.
We are working to inform legislators and regulators about the potential benefits
of this and similar new products. AIA encourages all State legislators and insurance
regulators to maintain a receptive attitude toward innovative and experimental in
surance products. These kinds of products will serve as the key that will guarantee
an open door to insurance for all consumers.
IV. Is Insurance Available?
In addition to the question of illegal discrimination, the Committee also has re
quested us to address the findings of recent property insurance studies. This is the
bottom line question that must be addressed: Whether those urban consumers wish
ing to purchase insurance currently are able to do so.
A number of studies from 1979 through 1993 show that a large percentage of
homeowners have homeowners’ insurance. In 1979, the AIA sponsored a nationwide
study on availability of homeowners’ insurance. That study showed that 98 percent
of homeowners had homeowners’ insurance. Eighty-eight percent of those surveyed
had comprehensive coverage covering fire, theft, vandalism, and liability. Fifteen
percent had more limited coverage.
70
In 1980, R.L. Associates, a private research firm in Princeton, NJ, conducted a
similar survey for the All-Industry-Research Advisory Council (now Insurance Re
search Council). This survey focused on the urban core of America, surveying urban
core neighborhoods in six of the largest American cities: Atlanta, Chicago, Cleve
land, Los Angeles, Philadelphia, and New York (borough of Brooklyn). The results
of the 1980 survey were consistent with those of 1979. In fact, the total percentage
of homeowners with homeowners’ insurance increased from 98 percent to 99 per
cent. The percentage with comprehensive versus more limited coverages also in
creased, from 88 percent to 90 percent.
In 1993, a similar study was commissioned by AIA and performed bv R.L. Associ
ates. The survey covered the experiences of homeowners with regard to obtaining
various forms of property insurance and their attitude toward that coverage. The
results of this study conclusively demonstrate that property insurance is widely
available in each of the cities surveyed, and that such coverage is “very easy” or
“somewhat easy” to obtain for the vast majority of policyholders. During the past
5 years few respondents have been turned down for any type of homeowners’ cov
erage, and an even smaller percentage have had their coverage canceled or non-renewed for any reason.
When combined with previous empirical research, this survey indicates that the
percentage of urban homeowners with property insurance is comparable to that in
suburban and rural jurisdictions. Moreover, to the extent that some homeowners are
without coverage, this decision is likely to be based on personal choice or economic
constraints, rather than difficulty in obtaining insurance.
Following are key highlights from the survey of 1,502 urban homeowners in Chi
cago, Los Angeles, Atlanta, Brooklyn, Cleveland, and Philadelphia:
• Less than 2 percent of the homeowners surveyed in the six cities did not carry
any homeowners’ insurance. Ninety-three percent had comprehensive coverage
covering fire, theft, storm damage, vandalism, and liability. About 5 percent car
ried more basic home insurance policies covering fire damage only or fire and
windstorm damage.
• The percentage oT homeowners without any coverage ranged from less than 1 per
cent in Chicago and Atlanta to 3 percent in Los Angeles.
• There were no significant differences among African-Americans and whites in
terms of insurance coverage. Ninety-nine percent of African-American home
owners carried either comprehensive (92 percent) homeowners’ coverage or more
basic policies (7 percent). Ninety-eight percent of those identifying themselves as
white nad some homeowners’ coverage, including 94 percent with comprehensive
policies and 4 percent with basic coverage.
• Nearly nine in ten (87 percent) of urban homeowners said it was very or some
what easy to find homeowners’ insurance. An even higher share (93 percent) said
it was convenient to contact an insurance agent or insurance company.
• Urban homeowners use a variety of methods to shop for and purchase home
owners’ insurance including telephoning an agent, having an agent come to their
homes, traveling to an agent’s office, purchasing through a mortgage or real estate
office, and purchasing home insurance by mail. A significant number of home
owners in each of the six cities have used one of these methods to purchase home
insurance.
• Only 3 percent of urban homeowners had experienced cancellation or non-renewal
of home insurance coverage during the past 5 years. Reasons for cancellation or
non-renewal cited by this small group of homeowners included non-payment or
late payment of premiums, loss experience (multiple theft or fire claims), insur
ance companies that ceased writing home insurance, and physical problems with
the property.
• Very few respondents (3 percent) said that they were aware of anyone in their
neighborhood who had experienced difficulty in obtaining homeowners’ insurance.
Other surveys and studies show similar results. The Roper Organization, Inc., a
nationally known public opinion survey firm, included a question on homeowners’
insurance as part of a representative sample of 1,976 Americans for the 1992 Public
Attitude Monitor published by the Insurance Research Council. Conducted during
June 1992, the survey found that 94 percent of those surveyed owning homes haa
homeowners’ insurance. Not surprisingly, the Roper survey showed some variations
in the share of homeowners with insurance by region, income, and community type.
For example, the survey indicated that homeowners living in central cities of metroolitan areas with populations of greater than 250,000 actually were more likely to
ave homeowners’ insurance (96 percent) than the population as a whole. Central
city residents owning homes, including those living in very large cities with more
than a million people, were more likely to have homeowners’ insurance than persons
g
71
living in suburbs (95 percent) and those living in rural areas and small towns (91
percent).
The 1992 Roper findings are nearly identical to those from another independent
national survey conducted by Cambridge Reports, Inc. in 1989 on home ownership
and home insurance rates. The Cambridge survey found that 95 percent of home
owners nationally carried homeowners* insurance.
Other surveys also document that small business insurance consumers are largely
able to obtain desired insurance coverages and that voluntary market availability
improved steadily during the 1980’s. The purchase of various insurance coverages
by small businesses increased during the 1980’s. Urban small businesses were in
some cases more likely to have some coverages than their suburban, small town or
rural counterparts. Nationally representative surveys of small businesses (Small
Business Attitude Monitor 1991, Business Attitude Monitor, 1988) conducted by the
Insurance Research Council showed that the share of small businesses purchasing
key coverages such as property and liability rose from 1988 to very hign levels by
1991. This evidence of general availability and affordability tracks well with the
independent data on the use of FAIR plans in commercial lines.
This evidence of general availability and affordability tracks well with the inde
pendent data on the use of FAIR plans in commercial lines.
Participation of the FAIR plan in the marketplace is decreasing. Commercial lines
premium volume written through the FAIR Plans declined steadily in most States
from 1986 through 1991 as a percentage of total commercialpremium volume. For
example, commercial written premiums in the Illinois FAIR Plan declined from
about 1 quarter of 1 percent (0.24 percent) to less than Vioth of 1 percent (.088 per
cent) of total commercial premium in the voluntary market. In addition to Illinois,
AIA analyzed the District of Columbia and eight large urban States: California,
Georgia, Massachusetts, Michigan, New Jersey, New York, Ohio, and Pennsylvania,
and found that each experienced significant drops in FAIR Plan commercial lines
premium volume in relation to the State commercial voluntary market. The decline
in FAIR Plan premium volume in relation to the commercial voluntary market gen
erally indicates that businesses were having an easier time finding insurance
through the voluntaiy market.
Personal lines FAIR Plan premium volume also decreased in Illinois and other
States relative to the voluntary market from 1986 through 1991. For example, FAIR
Plan premium volume in Illinois dropped from iust over Vz of 1 percent (0.52 per
cent) to about ¥3 of 1 percent (.036 percent) of total personal lines (homeowners)
premiums from 1986 to 1991. In New York, FAIR Plan premium dropped from
Vioths of 1 percent of the voluntary market to under 9ioths of 1 percent. Overall,
in eight out of the nine States that we analyzed, premium volume in the FAIR plan
declined in relationship to the voluntary market between 1986 and 1991, and in the
remaining State where FAIR plan market penetration did not decline, the increase
was less tnan Cloths of 1 percent.
The number of applications received by the FAIR Plans for commercial and per
sonal lines combined also generally decreased from 1986 to 1991. Our analysis of
applications and policies issued in the ten jurisdictions examined indicates that over
that period, the number of applications received annually declined by an average
of 25.5 percent and the number of policies and binders issued by each of the plans
annually decreased by an average of 23.3 percent. These are all signs of improving
roperty insurance availability m the voluntary markets of these States and the
listrict of Columbia. The bottom line conclusion is that the percentage of risks writ
ten by the FAIR Plan versus the voluntary market is decreasing steadily.
Small businesses located in major cities were just as likely as their counterparts
in suburbs, non-metropolitan cities and towns, and rural areas to have liability in
surance, more likely to have business interruption insurance, and somewhat less
likely to have property insurance. In this context, urban applies to small businesses
located within the city limits of a large city. As for specifically inner-city businesses,
the only representative study of which we are aware on these businesses was pub
lished m 1982 by All Industry Research Advisory Council (now the Insurance Re
search Council). This study, Availability and Use of Business Insurance by Small
Urban Businesses, covers inner-city small businesses in Chicago, Atlanta, Boston,
Brooklyn, Cleveland, Detroit, Los Angeles, and Philadelphia. Tne survey indicated
that at that time 92 percent of the inner-city firms had some type of property-liability coverage and 86 percent had the property coverages of fire, wind, and vandalism.
Small businesses in inner-city Los Angeles were as likely as businesses in the
overall sample to have at least one or more of the property-liability coverages.
Although it appears from the data that insurance is available and affordable to
the vast majority of residents and businesses in inner-city areas, we recognize that
this still might leave some insurable risks without coverage for one reason or an
P
72
other. We must continue our efforts to make insurance available for all insurable
risks.
V. Potential Federal Action
The third question asked by the Committee is what steps the Federal Government
should take to identify and combat illegal property insurance discrimination, if such
discrimination exits. As mentioned earner, a fair, economical, efficient data collec
tion scheme on the Federal level to determine if insurance is equally available,
based on the risk, for all Americans, is something AIA, IIAA, and The Council could
support. The parameters of such a data collection effort, however, are critical. Inor
dinate costs of collecting data with limited value would not assist the insurance
consumer, but rather, harm that consumer with either higher prices or possibly less
financially sound insurers.
Collection of data which reveals what insurance is being sold, where it is being
sold, who is selling it and how much it costs the consumer, would answer the ques
tion of whether or not insurance is available. This collection should be undertaken
in a format to provide the most information, in the most useful form, in the most
cost-efficient fashion.
The most that is needed today is a data collection and study effort, designed to
collect a limited amount of data from insurers on limited lines, in a limited number
of cities, for a limited period of time. The data should answer the four issues ad
dressed above, who, what, where and how much. It should require reporting in a
format that readily is obtainable and is cost effective—five-digit zip code. Various
studies could be undertaken for more complex issues such as commercial insurance,
agent appointments and terminations, insurance applicants, and the effectiveness of
the data collection.
Data collection should be limited in time and scope to answer the question is in
surance available equally in all neighborhoods at a cost commensurate with the
risk. The question of whether or not insurers are illegally discriminating is a ques
tion of regulation for insurance regulators. Federal data collection can assist regu
lators in making this determination, but data collection alone cannot answer that
question or enforce discrimination laws.
The lines of insurance raising concern are homeowners, dwelling fire and allied
lines and private passenger automobile. Data collection should be limited to those
lines. To determine the current situation, the duration of the study should be lim
ited. Anywhere from 1 to 5 years of data collection would show the status of insur
ance availability. The unit of measurement also needs to be reasonable. Data on the
existence and details of policies is kept by most insurers on a computer system
which includes at most only the five-digit zip code of the property insured. Insurers
have gone to great expense in recent years to install these systems so that they can
report five-digit zip code information to State regulators.
fe e full aaaress of the property insured often only is in a manually kept record,
and sometimes, if the property is part of an umbrella policy, not even there. Vir
tually no insurer has a statistical reporting computer system which collects or could
store nine-digit zip code information. The capital which would be required to convert
existing computer systems to report on any basis other than five-digit zip code is
extraordinary, and the value of data on any other basis is limited.
Last year, Mr. Chairman, you requested the U.S. General Accounting Office to in
vestigate the necessity of Federal reporting to determine availability, affordability,
and accessibility of property insurance. The GAO undertook an extensive investiga
tion of this issue, starting with a review of existing literature, proposed bills, and
State requirements. GAO staff discussed the issue with trade associations—includ
ing a number of meetings with AIA staff—consumer groups, statistical agents, re
search organizations ana the NAIC, and ultimately developed their own analyses
and conclusions.
The result of this study, the GAO Report to the Chairman, Committee on Bank
ing, Housing, and Urban Affairs, “Property Insurance, Data Needed to Examine
Availability, Affordability, and Accessibility Issues” was released on February 9,
1994. In analyzing what data should be reported, the report begins with the state
ment that collection of zip code data is important. While the report does state that
census tract reporting could be more useful, it goes on to say that this would require
restrictions due to the volume of data which would be generated. “The value of this
reporting must be weighed carefully against the additional burden it places on com
panies to comply.”
According to the report, the data that can be reported readily by companies (zip
code data tor policy information) would be useful to examine availability and afford
ability issues, “but will not be sufficient to determine conclusively whether unfair
discrimination exists or why.” However, the data would be a marked improvement
73
over what is available today and “could serve to point regulators more effectively
in directions for further probing.”
At this time insurance is regulated by the States. It is up to State insurance regu
lators, among other things, to investigate insurer practices and punish those insur
ers who illegally discriminate. Analysis of data reported by zip code clearly would
allow regulators and others to determine if there is a disparity between the number
of owner-occupied homes within a zip code and the number of homeowners’ policies.
Any large disparity then would indicate that the appropriate regulator needs to ask
insurers for more detailed information within that particular zip code. This would
limit the volume of information collected by the Federal Government to a manage
able load, but would provide the information needed to investigate areas of concern.
Additionally, the number of areas to be studied under any Federal data collection
should be related to the question raised. The allegations giving rise to this issue at
the Federal level address insurance availability in urban areas. To determine the
solution, data needs to be collected for those areas where there is alleged to be a
problem. Twenty-five of the largest urban areas include approximately 43 percent
of the United States population. This number clearly is sufficient to address the
question.
Any Federal data collection should be drafted in a fashion to provide quick, effi
cient, and cost effective data collection. H.R. 1188, introduced by Congresswoman
Cardiss Collins of Illinois, is a bill which would gather the information necessary
to determine if insurance is available in all neighborhoods equally, where the insur
ance is being provided, who is providing the insurance, what kind of insurance is
provided, and what it costs, on a five-digit zip code basis. This information could
be reported within 1 year, rather than the up to 3 years it could take for insurers
to begin providing census tract data, and at a reasonable cost to the consumer, rath
er than the exorbitant cost to consumers of insurers implementing an entirely new
reporting system for the Federal Government, while keeping in place their reporting
systems for State regulators.
The information which would be provided by H.R. 1188 would allow regulators to
focus efforts on zip codes with significant disparities between number of home
owners and number of homeowners’ policies, rather than being required to review
detailed information for huge numbers of census tracts with no problems whatso
ever.
Time and effort could be spent on addressing real problems of our urban cores
rather than over-reporting and reviewing of unnecessary data.
AIA, HAA, and llie Council urge the Committee to consider a Federal data collec
tion effort which will address the issue of insurance availability and cost in a ration
al, effective fashion.
VI. Affirmative Efforts by the Industry
AIA recognizes that there are a number of issues that may have an impact on
urban consumers and their ability to obtain insurance. The term redlining often is
used loosely as a catch-all, short-hand term to identify a wide variety of issues af
fecting the urban areas of our Nation. AIA members and others are taking steps
to address these issues. It is our belief that partnerships between the insurance in
dustry and consumer organizations, civil rights organizations and housing organiza
tions are the most productive way to move ahead.
A. L o c a tio n
of
Ag en ts
w it h
C o m pa n y A p p o in t m e n t s
One of the most controversial issues is whether agents with standard company
appointments can be found in urban areas. Some urban consumers contend that in
surance companies selectively place agents in locations that discourage applications
for policies from certain geographic areas. The majority of AIA memoer companies
market their products through the independent agency system. These agents are
truly independent entrepreneurs, usually representing more than one insurer. These
agents are not company employees placed in a particular office by the insurance
company. The companies they represent cannot and do not control the location of
the agent’s place of business.
However, AIA, IIAA, and The Council are committed to the idea of increasing the
number of minority and urban based agents with standard company appointments.
The Atlanta Agent/Insurer Partnership discussed earlier is one example of what the
industry is doing to move forward on this issue. Another example is insurers’ par
ticipation in agent association programs such as INVEST, a program which assists
students with financial difficulties to start a career in the insurance industry.
AIA and the IIAA also are working closely together on other efforts to increase
the number of minority agents with standard company appointment in programs
such as grouping of agents and special brokers programs.
74
A number of these types of programs are in the development stages around the
country.
B . Ot h e r M a r k e t in g S y s t e m s
It is important to note, however, that about 60 percent of personal lines insurance
is not sold through the independent agency system. It is sold by captive agents, em
ployees, or by direct marketing such as mail and telephone. Increasing the number
of independent agents serving urban areas alone will not be enough to resolve ac
cess problems. Other new marketing techniques will be required to better serve
urban areas. AIA members currently are participating in discussions at the NAIC
meetings on this subject, and would be delighted to assist regulators in developing
such techniques within the limits of State and Federal antitrust laws.
C. T h e P r o d u c t
Residents in urban areas must have ready access to insurance products that are
attractive to them as well as affordable. As discussed above, the first requirement
is the product. Products can be offered which provide coverages and limits which
are desirable, at a price that is affordable. AIA consistently has supported the au
thorization by law and the approval by regulators of such products. However, each
policy must be underwritten carefully, looking at underlying risk factors. To make
such products available to greater numbers of consumers, risk factors must be de
creased. Specifically, I am referring to the high risk of fire, high crime rates, exces
sive increases in automobile repair costs, and health costs.
Some view refusals to issue replacement cost coverage in homeowners’ policies for
risks that evidence a substantial disparity between replacement cost and market
value as redlining. Again, we disagree. AIA acknowledges that many insurers will
not provide this coverage when there is a wide disparity between the replacement
cost and the market value of a structure. In most instances, in order to qualify for
replacement cost coverage, an insured must buy enough coverage to represent at
least 80 percent of the cost of replacing the structure. This requirement may present
a dilemma for insureds who own homes built before the 1950’s. For many of these
older homes the cost to replace the building with the exact type of materials used
in the original construction far exceeds what the owner might spend to replace the
dwelling using modern materials and simpler construction techniques. Replacement
cost coverage is not well-suited for these kinds of properties, because insureds may
be unwilling or unable to buy—and insurers may be unwilling to sell—insurance in
an amount well beyond the price for which the dwelling could be sold. Insurers,
however, have responded to the needs of property owners faced with this disparity.
For example, a lower cost variable percentage replacement loss settlement endorse
ment has been developed.
This endorsement permits an insured to choose to insure at a lower percentage
of insurance to replacement value (i.e., usually 50 percent, 60 percent, or 70 per
cent), rather than the 80 percent usually required for replacement cost coverage.
Thus, if the policyholder decides to insure for 50 percent of the replacement cost of
the property, the premium would be lower than coverage for 70 percent of the re
placement cost. However, if the property is totally destroyed, the proceeds to the
policyholder would be 50 percent of the replacement cost of the property.
Additionally, the industry has developed repair cost policies that delete the re
placement cost provision and provide that a damaged dwelling will be repaired or
replaced with commonly used building materials instead of materials of like kind
and quality. It is important to recognize that when faced with a dilemma such as
this, the insurance industry has responded with products suited for a particular
purpose. Without the regulatory approval of these products, the industiy cannot ad
dress the problems of the consumer. These are further examples of the industiy re
sponding to consumer needs with specialized products.
It has been asserted that these types of products are inferior and do not provide
the same coverage as is available to properties without this large difference between
maiket value and replacement costs. However, the cost of insurance is based in part
on the expected cost to pay a claim under the policy. The price of providing replace
ment cost policies for these types of properties is so high that prior to the develop
ment of alternative products, consumers asserted that property insurance was un
available because it was unaffordable. To address this concern, insurers developed
new products that would provide a measure of coverage and protection, but would
not cost so much as to be unavailable. The price of insurance must reflect the risk.
Companies will continue to attempt to address the needs of consumers by creating
legitimate products that can be purchased for a reasonable price.
75
D . M a r k e t in g
and
S e r v ic e
Marketing and servicing products in communities where English often is a second
language presents special aifficulties. AIA companies have added Spanish speaking
customer service representatives to personal lines service centers and toll tree hot
lines, print posters and brochures in both English and Spanish, and have programs
to target small and disadvantaged contractors. These efforts make marketing and
servicing products in the urban areas more effective.
E . E d u c a t io n
Education is critical to the availability and accessibility of insurance. Education
must be conducted both for consumers and for insurers. Currently AIA members are
working with individual State regulators as well as the NAIC Insurance Availability
and Affordability Task Force Subcommittee on Education to develop programs to in
crease the knowledge of average and low-income consumers about insurance, and
to increase the knowledge of underwriters and agents about urban neighborhoods
and opportunities. We believe this concerted effort will assist in increasing the avail
ability and accessibility of insurance products which are truly priced based on the
risk.
F. P r o d u c t P r ic in g
Some consumers allege that insurers intentionally overprice a product to prevent
sales in urban areas. In many States, pricing of insurance is determined through
a regulatory approval process. Rates are not intentionally set higher for urban risks
to prevent people from purchasing the insurance. Individual company rates should
be and are based on loss experience and are subject to the review of the State insur
ance regulator. In many cases, urban insurance rates are the same as or lower than
those applicable to risks in suburban areas. In other instances policies for risks in
urban areas will be more expensive than similar risks in suburban areas. This rate
disparity is due to the increased costs of certain policies, such as increased incidents
ana severity of claims in certain areas, or increased distance from a fire station.
Automobile insurance is a good example of this cause and effect. The cost of auto
mobile insurance reflects the costs of goods and services that are paid by automobile
insurance premiums, including litigation, health care, and auto repair. The fre
quency of bodily injury liability claims countrywide has increased 19.0 percent from
1987 through the third quarter of 1992, according to NAH/ISO Fast Track data. The
increase in the loss cost for bodily injury liability during that same period was 67.1
percent. Meanwhile, the frequency of property damage liability claims and collision
claims in countrywide has decreased 11.0 percent and 16.0 percent, respectively,
with the loss costs increasing 13.0 percent and 1.4 percent. This helps demonstrate
that liability claims and the resulting medical costs are a major source of high and
rising auto insurance costs.
Of course, these costs are not uniform between States, or within areas of certain
States. For example, in Illinois, the average loss cost for bodily injury liability is
$109, compared to the countrywide average loss cost of $119. In the District of Co
lumbia, the average loss cost for bodily injury liability is $191, compared to the
countrywide average loss cost of $119. Virginia has an average loss cost for bodily
injury liability of $101, and Maryland $144. Some other average loss costs as of the
third quarter of 1992 for these areas are: property damage liability: Illinois $65,
D.C. $79, Virginia $47, Maryland $62, countrywide $56; collision: Illinois $117, D.C.
$150, Virginia $76, Maryland $104, countrywide $103. These costs are reflected in
rates charged in these States.
There also are significant differences in how costs are distributed within States.
For example, according to a 1990 report, the bodily injury liability claim frequency
per 100 insured cars in Chicago was 3.07 versus the Illinois Statewide average of
1.81. This means that bodily injury liability claims were filed 69.6 percent more
often in Chicago than for the State, as a whole. Also, there were 52 bodily injury
liability claims for every 100 property damage liability claims in Chicago compared
to 34 for the State, as a whole, feus, in Chicago, an injury claim was 55 percent
more likely to be filed for each property damage claim, than the average for the
State of Illinois.
We understand the role that insurance costs play in limiting access to insurance.
We know that rates based on loss experience may be more than some consumers
are able to pay. In fact, if insurance is not affordable, it essentially is not available.
However, we believe there are rational solutions to problems relating to the cost and
the value of auto insurance. One such solution is the development of innovative,
low-cost products such as our no-frills, no-fault automobile insurance policy pro
posed in California. With cooperation and creativity we can meet the needs of all
consumers, including those who are low income, by offering them useful and afford
76
able insurance products. If these products are available, and consumers are inter
ested in purchasing them, insurers are sure to increase their marketing efforts to
be the company making those sales.
G. R i s k R e d u c t io n
To lower cost and increase insurance availability, the risk in some neighborhoods
must be decreased. ALA. and member companies are working with community orga
nizations such as the Association of Communities for Reform Now (ACORN) on pro
grams such as the ACORN Neighborhood Home and Safety Program to assist com
munities in lowering their risks, and then providing insurance for those participat
ing residents.
VII. Other Issues
In recent discussions of redlining, many of the above issues have been raised. Ad
ditionally, issues of insurance investment in urban areas, affirmative action within
insurance companies and contributions to minority or urban-oriented organizations
have been added to the discussion. I would like to address what we are doing in
connection with these broader social and economic issues. We are proud of our suc
cesses, and would like to mention them. However, we recognize that more has to
be done by everyone, including insurers, to help revitalize our urban areas.
A. A f f ir m a t iv e A c t io n
w it h in
I n s u r a n c e C o m p a n ie s
AIA member companies fund and participate in the recruitment and advancement
of minorities through a variety of programs, both external and internal. The exter
nal organizations supported by our members include: The Urban League, Black Ex
ecutive Exchange Program, and SER (Jobs for Progress, Inc.), minority summer in
ternship and scholarship programs like INROADS, inner-city youth job training ini
tiatives like INVEST and the STAG Program. Internal initiatives include targeted
college recruitment, company-wide managerial diversity training, the creation of
specific, regular opportunities for minority employees to meet and network with top
management, minority career development programs, coaching and mentoring pro
grams. Several companies have made an explicit commitment to promote minorities
to the highest professional and managerial positions. Some advance the process by
auditing the employee mix to ensure representation of the labor pool at large.
In addition to employee opportunities for minorities, many companies also target
minority companies as vendors. For example, one AIA member’s targeted minority
vendors program played a part in 12 percent of the company’s 1991 purchases being
made through minority or women vendors.
B . J o bs
in
U r ba n Ar e a s
Probably the most important need of urban centers today is jobs for urban resi
dents. Many AIA member companies maintain significant facilities in urban areas
across America. Thousands of workers are employed by the insurance industry in
cities such as Chicago, Atlanta, Baltimore, Boston, Cleveland, Dallas, Detroit, Los
Angeles, Miami, New York City, Philadelphia, and Pittsburgh.
C. I n s u r a n c e C o m p a n y I n v e s t m e n t
in
U r ba n A r e a s
Related to the issue of jobs is the issue of investment. Insurance industry invest
ment in urban areas is significant. Excluding all other types of insurance invest
ments in urban areas and only considering the purchase of municipal bonds, the in
surance industry has invested billions of dollars a year into urban areas. For exam
ple, for the most recent year for which we have data, 1989, insurers held or sold
a total of $1.8 billion in various Chicago municipal bonds. For Los Angeles city
alone, insurers held or sold $573.4 million in these bonds. For Los Angeles county,
insurers sold or held $501.2 million in bonds. The total of all bonds sold or held by
insurers as of the end of 1989 for both Los Angeles city and county was approxi
mately $1.1 billion. In addition to these investment dollars, companies also contrib
ute financially to many regional and national organizations dedicated to the ad
vancement of minority interests and the revitalization of American cities. A list of
some of these organizations is attached as Appendix A.
These contributions assist not only residents in urban areas, but all of America,
by giving urban residents a chance for housing, education, cultural activities, career
training, and a host of other opportunities which would not be available without the
efforts of corporate America. The insurance industry is proud of its participation in
these programs, and intends to look for additional ways to assist in the revitaliza
tion of our urban areas.
77
V lll. Conclusion
This sensitivity to the benefits of working with minorities and women as employ
ees and managers, as a sales force, as vendors, as customers, and as citizens, is well
documented in the insurance industry. We are committed to work with the Con
gress, State legislators, insurance regulators and consumers to address both broad
social and economic and insurance specific problems.
In summary, we oppose redlining. We support the NAIC Unfair Trade Practices
Model Act. We support H.R. 1188. Further, we look forward to working with the
Senate as you address these important issues.
PREPARED STATEMENT OF RAUL YZAGUIRRE
P r e s id e n t
of th e
N a t io n a l C o u n c il
of
L a R aza
I. Introduction
Chairman Riegle and distinguished Members of the Senate Banking, Housing,
and Urban Affairs Committee, my name is Raul Yzaguirre, and I am President of
the National Council of La Raza (NCLR). NCLR is the Nation’s largest constituencybased Hispanic organization, representing 170 affiliated community-based organiza
tions which together serve 37 States, Puerto Rico, and the District of Columbia, and
reach more than two million Hispanics annually. We are pleased that the Commit
tee is conducting this hearing on insurance discrimination, and we appreciate the
opportunity to present our views on this important economic development and civil
rights issue.
Since its inception, NCLR has maintained a major institutional focus on civil
rights and the principle of eoual access to opportunity for all Americans. In addi
tion, we have long recognized the critical importance of access to the tools of eco
nomic development and to housing affordability.
In the area of civil rights, in recent years NCLR has done research and policy
analysis on such subjects as employment discrimination and Government enforce
ment of civil rights protections tor Hispanics. In December of 1993, NCLR issued
a report, The Empty Promise: Civil Rights Enforcement and Hispanics, which docu
mented the Equal Employment Opportunity Commission’s ineffectiveness in protect
ing the Hispanic community against emplovment discrimination. NCLR has recently
launched a “Know Your Rights” public information and education campaign, a pri
vately funded nationwide effort designed to educate Hispanics about tneir civil
rights and to develop improved, community-based models to increase the effective
ness of civil rights enforcement. NCLR completed a complementary Fair Housing
Education and Outreach project, designed to help raise awareness within the His
panic community about housing discrimination and the fair housing enforcement
process. In cooperation with five of our affiliates, the project was undertaken in Chi
cago, San Diego, Kansas City, El Paso, and the State of Nebraska.
In the field of housing and community development, NCLR has since 1972 pro
vided direct on-site technical assistance to scores of communities and communitybased organizations in the planning, financing, development, and management of
housing construction and rehabilitation programs. These projects include a series of
self-help rehabilitation programs in Texas financed by the Farmers Home Adminis
tration (FmHA); one of a very few FmHA-financed Hmited-equity cooperatives in the
United States (in Arizona); Community Development Block Grant-financed tene
ment rehabilitation programs in urban areas of Texas; a self-financed and self-sufficient farmworker mobile home cooperative in Florida; litigation-supported anti-dis
placement programs in Virginia; and a series of Department and Housing and
Urban Development-funded housing developments for low-income, elderly, and
handicapped persons in Texas, Arizona, and California. In 1991, NCLR helped local
groups to leverage over $5 million for the construction and rehabilitation of more
than 190 housing units, including 45 units for farmworker families. NCLR is also
helping six low-income communities in Arizona with the financing and design of safe
drinking water and sewer system. Currently, NCLR is in the initial phases of a
major new effort, the Southwest Community Development Initiative. The Initia
tive is' a multi-year program designed to improve the socio-economic conditions of
Hispanics in the Southwest through comprehensive support and assistance to its
traditional constituency, Hispanic-controlled community development corporations
(CDC’s) and other nonprofit community-based organizations. Targeting urbanized
areas in Texas, New Mexico, Arizona, California, and Colorado, the Initiative will
help about 20 selected Hispanic community development entities to increase their
staff and organizational capacity, resources, and linkages, and to achieve multiple
78
community impacts ranging from housing and community facility development to
local political empowerment and improved human services. The Initiative estab
lishes NCLR as a national and regional intermediary, working with other national,
regional, and local entities to achieve broad community development objectives.
NCLR’s experience has led it to an increased focus on the nexus between the civil
rights and housing development fields—the issue of fair lending and disinvestment
and the issue of discrimination in insurance and other crucial business and
consumer services. In 1987, we joined the Center for Community Change, ACORN,
and others in supporting the permanent authorization of the Home Mortgage Disclo
sure Act (HMDA). During the debate over the Financial Institutions Reform, Recov
ery, and Enforcement Act in 1989, NCLR strongly supported the provisions that ex
panded HMDA reporting requirements. Since 1990, we have conducted a series of
training sessions for our affiliates on the Community Reinvestment Act (CRA) and
HMDA. NCLR not only recognizes the crucial link between civil rights and housing
development, but has througn its varied activities attempted to make the link con
crete. We, therefore, have a profound interest in the subject of insurance discrimina
tion, particularly homeowners’ insurance discrimination.
II. Impact of Discrimination on the Hispanic Community
A . O v e r v ie w
The history of Hispanic Americans is, in large part, a history of discrimination.
The beginnings of the Mexican American experience are rooted in conquest, conflict,
and hostility. After the end of the Mexican War in 1848 thousands of Mexicans were
murdered. During this time, executions without trial and lynching were a common
thread of life in the Southwest. It is not surprising that, then and since, Mexican
Americans have been subject to enormous discrimination in education, employment,
housing, and the administration of justice.
When Puerto Ricans became U.S. citizens in 1917 as a result of the Jones Act,
they too suffered the consequences and stigma of being a conquered minority group.
Arriving to the mainland as agricultural laborers ana later as industrial workers,
Puerto Kicans have faced high levels of social and economic discrimination in edu
cation, housing, and employment.
Other Hispanics, including Cuban and Central Americans, have also been sub
jected to severe levels of discrimination. The stereotypes associated with “Hispanic”
are applied indiscriminately to Mexicans, Puerto Ricans, Cubans, and other Spanish
speakers. The literature of the U.S. is replete with derogatory references to His
panics, as “mongrels,” ‘lazy,” “ignorant, illiterate and non-moral,” etc. The public
war against drugs in the 1980’s added to hostility against Hispanics as the media
focus almost exclusively on South American druglords and traffickers. These ref
erences have formed the basis for a stereotype of Hispanics as being inferior, primi
tive, and dishonest; such stereotypes lead, directly or indirectly, consciously or un
consciously, to discrimination.
While many would prefer to believe that prejudice and bigotry have been all but
eradicated—and indeed, there has been mucn progress—the impact of continued dis
crimination should not be underestimated. The effects of such discrimination trans
late directly into social and economic behaviors and conditions which are destructive
to all of society. We cite below some historical and recent studies documenting the
scope and degree of the more “traditional” forms of employment and housing dis
crimination against Hispanic Americans.
B . E m p l o y m e n t D is c r im in a t io n
Employment discrimination serves as a useful, if imprecise, indicator of discrimi
nation against Hispanics in other areas. And in a more obvious way, employment
discrimination against Hispanics often delineates the social and economic bound
aries within whicn Hispanics operate, including where they live.
Mounting evidence indicates that negative attitudes about Hispanics affect their
employment opportunities. A 1989 study conducted by the Urban Poverty and Fam
ily Structure Project at the University of Chicago analyzed the manner in which em
ployer’s perceptions of ethnicity and race affect hiring decisions. The study, based
on interviews with 185 Chicago-area employers, found that 70 percent of those sur
veyed made distinctions among employees or potential employees based on ethnicity
and race. According to the employers, being Hispanic and/or Black was perceived
as being “lower class” and being White meant “middle class.” The study confirmed
the tendency of employers to generalize about the meaning of ethnicity and race
with regard to the quality of the work force and to rely on these generalizations in
their hiring practices.
A 1989 study by the Urban Institute indicates that unequal treatment of Hispanic
and Black jobseekers is entrenched and widespread. This “hiring audit” enlisted re
79
cent college graduates who were over-qualified for positions advertised through
newspapers; auditors were articulate and poised, spoke and dressed conventionally,
and posed as having prior job experience. Auditors were carefully matched on all
characteristics that could affect a hiring decision, and were trained to behave as
similarly as possible in an interview setting.
The 360-audit study conducted in Chicago and San Diego indicates that His
panics, like Blacks, are systematically denied equal opportunity in the hiring proc
ess. The study found:
• White applicants received 33 percent more interviews and 52 percent more job of
fers than the Hispanic applicants; and
• 31 percent of the Hispanic applicants encountered unfavorable treatment in the
hiring process, compared to only 11 percent of the Anglo applicants.
Similar results were found in a 1992 hiring audit conducted in the Washington,
DC—metropolitan area by the Fair Employment Council of Greater Washington
(FEC). In its study, the FEC also used closely matched pairs of Hispanic and Aiglo
testers to inquire about and apply for advertised job openings by botn telephone and
mail. The study found that Hispanic testers encountered discrimination due to their
national origin in more than one job application in five, or about 22 percent of the
time.
Survey research conducted by the General Accounting Office (GAO) documented
additional evidence of national origin discrimination against Hispanics. In a March
1990 report on the Immigration Reform and Control Act of 1986 (IRCA), the GAO
reported the results of a survey of 4,362 employers concerning the effects of IRCA’s
employer sanctions provisions on their hiring practices. The GAO found that:
• An estimated 10 percent of the employers surveyed reported discriminating
against employees or job applicants solely on the basis of national origin charac
teristics;
• An estimated 5 percent began a practice of refusing to hire persons based on “for
eign” appearance or speech accent; and
• An estimated 8 percent required only “foreign-looking” and “foreign-sounding”
persons to comply with IRCA s employment verification requirements, rather than
requiring all job applicants to comply as mandated by IRCA.
The effects of employment discrimination may also be measured in terms of lost
wages. Labor market research provides substantial evidence that, throughout the
1980’s, Hispanics continued to experience high-levels of employment discrimination.
At least five independent studies have found that, even after controlling for factors
known to affect employment and earnings, such as age, occupation, and educational
attainment, a significant proportion of the “earnings gap” between Hispanics and
Anglos appears to be attributable to employment discrimination.
• A 1982 National Council of La Raza study, The Effects of Discrimination on
the Earnings of Hispanic Workers: Finaings ana Policy Implications, using
data from the U.S. Bureau of the Census, March 1981 Current Population Survey,
found that 14 percent of the earnings gap between White males and Hispanic
males and 29 percent of the gap between White males and Hispanic females Were
due to ethnicity alone, suggesting serious levels of employment discrimination;
• A 1982 U.S. Commission on Civil Rights report, Unemployment and Underem
ployment Among Blacks, Hispanics ana Women, using data from the March 1980
Current Population Survey, found that, while disparities in unemployment and
underemployment between Hispanic and Anglos could be explained to some extent
as reflections of differences in education, training, and age, substantial disparities
remained even after controlling for these factors. The Commission concluded that
sufficient evidence exists to suggest that discrimination continues to be a signifi
cant, if not precisely quantifiable, factor in employment disparities between
Whites and Hispanics.
• A 1985 University of Colorado study, using Census data to analyze the causes of
the disparity in earnings among Hispanic, Anglo, and Black males, found that in
1980 discrimination and labor market segmentation accounted for 18 percent of
the difference between Hispanic and Anglo male earnings.
• A 1990 study sponsored by the Inter-University Program (IUP) for Latino Re
search, using data from the 1940, 1950, 1960, 1970, and 1980 Censuses, and the
1983, 1986, and 1988 Current Population Surveys, analyzed and compared Latino
and White incomes from 1939 to 1987. After controlling for demographic, occupa
tional, and human capital differences between Hispanics and Whites, the re
searchers estimated that employment discrimination accounted for:
☆ Approximately 10-16 percent of the gap between Latino male and White male
incomes from 1973 through 1987; and
80
☆Approximately 30-40 percent of the Latino female-White male income gap over
the same period.
The study concluded that, although significant progress was made in the 1960’s,
inequality due to discrimination in the labor market has not declined, and for
many Hispanics has actually increased over the past 20 years. According to the
IUP study, discrimination has actually increased for Latino women compared to
White women, for Mexican native-born males compared to White males and for
Mexican immigrants given similar patterns of industrial employment compared to
White men. This persistence and/or renewal of discrimination coincides with the
end of major initiatives and some reversals in the area of affirmative action.
• Finally, a studv reported in 1991 by Edwin Melendez of the Massachusetts Insti
tute of Technology compared the hourly wages of Hispanics and non-Hispanics in
New York City. The study, which used 1980 Census Public Use Microdata Sample
for New York City, found that while labor market location was responsible for a
substantial proportion of the Hispanic/non-Hispanic wage differential, discrimina
tion was still a very significant factor. The study concludes that discrimination in
employment accounted for:
☆ One-third of the wage gap for Mexican, Puerto Rican, and Cuban men and onehalf of the wage gap for Other Hispanic men in the New York City labor mar
ket; and
☆ Between one-fifth and one-half of the wage gap for all Hispanic women in New
York City.
Taken together, the results of these studies provide compelling evidence of persist
ent, severe employment discrimination against Hispanics. Despite the fact that the
studies relied on a number of different data bases and used somewhat different
methodologies, the research findings are remarkably consistent:
• The percentage of the Hispanic male-Anglo income gap attributable to employ
ment discrimination falls within a 10-18 percent range;
• The percentage of the Hispanic female-Anglo male income gap attributable to em
ployment discrimination falls within 30-40 percent range.
Considered as a whole, the combination of the cited studies offer persuasive evi
dence that Hispanics suffer from high levels of discrimination in the labor market.
At a minimum, the evidence indicates that true equal employment opportunity for
Latinos remains an unfulfilled goal.
C . H o u s in g D is c r im in a t io n
In addition to employment discrimination, a growing body of evidence documents
the scope and degree of housing discrimination against Hispanics:
• Dallas: A 1979 HUD audit in Dallas was the first HUD research to focus on the
extent of housing discrimination against Hispanics. It had previously been as
sumed that Hispanics would suffer less than Blacks from housing discrimination
for two reasons: Hispanics were a smaller subpopulation, and the income of His
panic renters was somewhat greater than that of Black renters. The results of the
audit showed otherwise.
The Dallas study found that 42 percent of dark-skinned Mexican Americans and
16 percent of light-skinned Mexican Americans were given false information on
the availability of rental units. The chance of dark-skinned Mexican Americans
experiencing at least one instance of discrimination in a typical housing search
was 96 percent, and the probability of light-skinned Mexican Americans experi
encing similar discrimination was about 65 percent.
The incidence of discriminatory treatment against dark-skinned Mexican Ameri
cans was found to be far greater than that against either Blacks or light-skinned
Mexican Americans. Discriminatory treatment against dark-skinned Mexican
Americans regarding availability of a unit was two and one-half times as great
than against either Blacks or light-skinned Mexican Americans. Furthermore, dis
criminatory treatment against dark-skinned Mexican Americans on terms and
conditions of the contract was far greater.
• Denver: A 1982 HUD-funded study in Denver also found evidence of housing dis
crimination against Hispanics. When teams of White and Hispanic auditors were
sent out to homes that were advertised for sale, the White auditors received con
siderably more information when they asked about homes similar to the adver
tised house in the same general area or in other locations. One in three Hispanic
auditors was told that there were no homes available which were similar to the
advertised home and in the same general area, compared to one in five White
auditors. Furthermore, 60 percent of Hispanic auditors were told of no similar
homes in other areas, compared to 31 percent of White auditors. These differences
81
in the degree to which Hispanic and White auditors were informed of housing al
ternatives were sizable ana statistically significant.
Discrimination against Hispanics regarding housing availability was not found
to be as widespread in the rental housing market in Denver, but some differences
in the treatment of Hispanic and White auditors strongly suggest discrimination.
For example, Hispanic auditors were twice as likely as Whites to be told that the
advertised units were unavailable, and also twice as likely not to be told of other
rental housing possibilities which might meet their needs. In neither case, how
ever, were the differences statistically significant.
• Boston: A Boston telephone survey found major differences in treatment of Whites
and minorities. In a HUD-funded study in 1981, a telephone survey was done of
selected realtors who were advertising units for rent. Tne test teams consisted of
three persons: one that would normally be identified by voice as White, one as
Black, and one as Hispanic. In the 42 tests conducted, all White testers were in
vited to come to the office to be shown a unit, while Black and Hispanic renters
were informed that no units were available 31 times.
Site visits were carried out on 47 rental firms, including those that were al
ready covered by the telephone survey and continued to advertise units for rent.
In 23 of 47 tests, only the White testers were shown units while Black and His
panic testers were told nothing was available.
• Housing Discrimination Study: In 1989, HUD sponsored a national fair housing
audit study which was conducted by The Urban Institute. Results were based on
3,800 fair housing audits (paired tests) conducted in 25 metropolitan areas. Pairs
of auditors—one White and the other minority—posed as otherwise identical
homeseekers. They responded separately to advertisements randomly selected
from the major newspapers of 40 metropolitan areas, and recorded tneir treat
ment by real estate and rental agents. According to THE study, Hispanics seeking
homes experienced discrimination in at least half of their encounters with both
sales and rental agents. The HUD study found that the incidence of discrimina
tion is 56 percent for Hispanic homebuyers and 50 percent for Hispanic renters.
Given this large and growing body of evidence, the National Council of La Raza
believes that housing discrimination experienced by the Nation’s 25 million Ameri
cans of Hispanic descent is massive in scale. The effects of such discrimination are
examined below.
IQ. Effects of Housing Discrimination
A. O v e r v ie w
While the effects of employment discrimination can be quantified in terms of lost
wages, similar calculations with respect to housing discrimination are necessarily
more complex and imprecise. One obvious effect is that Latinos almost certainly pay
more for comparable nousing than non-Hispanics. Given that one effect of housing
discrimination is to artificially limit the portion of the housing market accessible to
the Latino community, the laws of supply and demand would dictate that the rel
ative price of housing available to Hispanics would increase. And indeed, survey
data suggest that Hispanic families pay an excessive portion of their incomes for
housing. Between 1978 and 1991, the proportion of poor Hispanic households spend
ing at least 30 percent of their income on housing grew from 69 percent to 80 per
cent. In 1991, 48 percent of all Hispanic households spent at least 30 percent of
their income on housing, and 24 percent of all such households paid at least 50 per
cent of their income on housing. By comparison, 31 percent of non-Hispanic house
holds spent at least 30 percent of their income on housing, and 14 percent paid at
least 50 percent of their income on housing during the same period.
Given the complex relationship between housing location, the quality of Govern
ment services, and the nature of the economic development process, nousing dis
crimination has more deleterious—if somewhat more subtle—impacts on Hispanics.
These impacts include inferior schooling and stunted opportunities for economic de
velopment.
B . S c h o o l S e g r e g a t io n
The significance of discrimination against Latinos seeking housing cannot be over
emphasized. One effect of housing discrimination is school segregation. Analyses of
Census data reveal that Hispanic students are more likely than other minority stu
dents to attend predominately minority schools; for example, in the 1991-1992
school year, almost three-quarters (73.4 percent) of all Latino students attended
schools that were predominately minority, compared to two-thirds (66.0 percent) of
Black students. This study, which was commissioned by the National School Board
Association, documents the extent and severity of school segregation, noting the
“educational damage associated with racial segregation.” Segregation by race is
82
strongly related to segregation by poverty. Students in segregated schools not only
face discrimination and stereotypes about minority schools but also attend—schools
struggling with the much greater concentration of health, social, and neighborhood
problems that are associated with high levels of poverty.
Problems of race and poverty segregation are the direct result of housing policies
and housing discrimination. Unless we address the issue of discrimination in the
housing market, we are unlikely to successfully resolve the problems associated with
school segregation.
C . E c o n o m ic D e v e l o p m e n t S t u n t e d
A re la te d and eq u ally d ev a sta tin g effect o f h ousin g d iscrim in ation is its im p act
on th e econom ic developm ent o f a neighborhood. E conom ic iso lation is in larg e p a rt
th e d irect re s u lt o f d iscrim in ato ry hou sin g p ractices w hich lead to th e co n cen tratio n
o f m in o rities in ce r ta in neighborhoods, w hich in tu rn a re associated w ith th e d is
in v e stm e n t o f b a n k in g in s titu tio n s a s w ell a s th e exodus o f th e m y riad co n su m er
and b u sin e ss serv ices th a t a re cru cial to th e econom ic w ell bein g o f in d iv id u als and
co m m u n ities. In d ivid u als and e n tire neighborhoods su ffer th e b ru n t o f d isin v e st
m en t. D isin v estm en t by econom ic e n titie s fosters an atm osphere rife w ith op p ortu n i
tie s for exp lo itatio n . In m any L a tin o neighborhoods, w hile th ere is lim ite d o r no a c
cess to b a n k in g in stitu tio n s, th e r e is no sh ortag e o f usurious ch eck c a sh in g e s ta b
lish m e n ts, “cu rren cy ex ch an g es,” loan sh a rk s, and oth er shady so u rces o f fin a n cin g
and cre d it, an d no sh ortag e o f cu sto m ers in need o f th e ir services. So ciolog ist W il
lia m J u liu s W ilson arg u es t h a t a s com m un ities are iso lated from m a in stre a m eco
nom ic and so cial in stitu tio n s, so too a re th ey iso lated from m a in strea m p a tte rn s and
n o rm s o f b ehav ior. E con om ic iso latio n is th u s in ex tricab ly in tertw in ed w ith p e rsist
e n t poverty, neighborhood d eclin e, and all th e social problem s a ttr ib u ta b le th e re to .
IV. Home Ownership Opportunities Denied
A . O v e r v ie w
In 1991, the American Housing Survey (AHS) indicated that only 39 percent of
the Hispanic population in the U.S. owned a home; in contrast, 66 percent of the
non-Hispanic population owned a home. From a different perspective, approximately
85 percent of all homeowners were non-Hispanic Whites, 8 percent were Black, and
only 4 percent were Hispanic.
The social and economic consequences of this disparity are enormous. High rates
of home ownership among Americans—frequently encouraged and facilitated by var
ious public policies and subsidies—have been cited by historians and economists
alike as being responsible for the “fluidity” or “upward mobility” characteristic of
American society. In addition to the clear economic advantages accrued to home
owners in terms of savings, accumulation of wealth, and tax benefits, the very idea
of home ownership has had powerful symbolic significance for generations of Ameri
cans. If you worked hard and “played by the rules,” every American family could
expect to own their own home. This ideal, and the fact—that it was true for most
Americans, played a major role in assuring that every American “bought into” our
social, economic, and political system.
The denial of home ownership opportunities to Hispanics and others based on race
or national origin is thus twice damaging—it directly harms the victims of discrimi
nation and also undermines the very ideals upon which our society functions.
B . D is c r im in a t io n
in t h e
M o rtg ag e M a rket
The significantly lower levels of home ownership among Latinos are attributable
in part to their relatively high rates of poverty. Yet, poverty alone does not account
for the consistently low rates of Hispanic home ownership across income, geography,
and over time. Discrimination against Hispanics in the private housing finance mar
ket appears to play a significant role; for example:
• The 1990 Home Mortgage Disclosure Act (HMDA) data show that within each in
come category and for every type of loan documented—Government-backed, con
ventional, refinancing, home improvement—Hispanics were significantly less like
ly to receive loan approvals than Whites with similar incomes; specifically:
☆A greater percentage of Zoio-income Whites (69 percent) obtained conventional
mortgage loans than moderate-income Hispanics (68 percent).
☆ Low-income Whites had significantly greater approval rates than wpper-income
Hispanics for refinancing and home improvement loans.
• A Department of Justice investigation of a well-known mortgage company in Bos
ton found that in 1992, the denial rate was 11 percent for white mortgage appli
cants, 21 percent for blacks, and 24 percent for Hispanics.
83
• According to a 1993 study by Aveiy, Sniderman, and Beeson who analyzed 1991
HMDA data, the denial rate for mortgage loan applications from Hispanics is 50
percent higher than for White applicants with the same income characteristics.
A common industry response to higher denial rates of mortgage applications from
Hispanics is the assertion that Hispanics have higher loan-to-value ratios and are
therefore presumed to be more likely to default on a loan. Yet a 1993 study by
Berkovec, Canner, Gabriel, and Hannan found that the default rate for Hispanic
households does not differ significantly from the rate for White households.
In short, discrimination in the mortgage lending process is a serious problem for
Hispanics, and cannot easily be accounted for by market factors.
C. I n su r a n c e D is c r im in a t io n
NCLR believes that discrimination against Hispanics in the insurance market is
similarly pervasive, although there are far fewer studies focusing exclusively on in
surance discrimination in general or homeowners’ insurance discrimination in par
ticular. However, anecdotal evidence suggests that insurance companies “redline”
minority neighborhoods in much the same way that banks “redline” minority neighborhoods; for example:
• In 1993, the Texas Department of Insurance (TDI) analyzed participation trends
in the State-sponsored insurance plan to which rejected insurance applicants are
typically referred. The TDI study found that those areas of high Hispanic con
centration (including Dallas, Houston, and the Rio Grande Valley) had signifi
cantly higher rates of participation in the State-sponsored insurance pools imply
ing disparities in rejection rates of Hispanic applicants. The study also revealed
that those areas without insurance agent operations had high concentrations of
Hispanics.
• An ACORN study of five cities indicates that minority and low-income neighbor
hoods were underinsured by as much as 48 percent in some low-income neighbor
hoods compared to non-minority areas.
• Just this year, the National Fair Housing Alliance (NFHA) made public the re
sults of a 3 year, in-depth investigation of insurance company practices. Using
testers from White and minority neighborhoods, NFHA found widespread dis
crimination against minority testers in their attempts to obtain homeowners’ in
surance. In the city of Chicago, the incidence of discriminatory treatment of
Latinos more than 95 percent. That is, in over 95 percent of the attempts by a
Latino tester to obtain insurance, the insurance company:
(1) did not return phone calls to offer either a quote on policies or to provide
needed information;
(2) would sell only lower-quality insurance coverage and fail to inform the
Latino tester that the coverage was substandard while testers in White neigh
borhoods were told about top-of-the-line policies; or
(3) applied stricter standards, such as minimum value requirements.
The implications of insurance discrimination against Latinos are far-reaching.
Without homeowners’ insurance, Latinos cannot qualify for mortgage loans and
therefore cannot become homeowners. Many lenders require replacement value cov
erage—which, as the NFHA study revealed, is infrequently offered to testers from
minority neighborhoods. In addition, the higher costs for insurance coverage in
Latino neighborhoods directly affects the level of income disposable toward the pur
chase of a home. Home ownership is crucial to maintaining or furthering the eco
nomic and social viability of neighborhoods. Hence, discrimination by insurance
companies can have the effect of stunting economic development and contributing
to the decline of neighborhoods in much the same way as housing and mortgage dis
crimination.
Given the historical evidence regarding discrimination against Hispanics in the
employment, housing, and mortgage markets, the strong indications that similar
discrimination may be taking place in the insurance market, and the severe con
sequences such discrimination would have on Latinos’ housing choices and rates of
home ownership, NCLR believes the time has come for the Congress to enact insur
ance disclosure legislation. Described below are what we believe to be the essential
elements of such legislation.
V. Recommendations
The National Council of La Raza supports the enactment of strong, landmark in
surance disclosure legislation. To be effective, such legislation must include four
basic provisions:
• Zip + 4: It is essential to report information about the availability, cost, and qual
ity of insurance policies on a zip + 4 basis. Reporting on a zip + 4 basis would
84
allow use of demographic data from the Census Bureau because such data are
easily translated into census tract units. Moreover where Hispanics are concerned,
the spatial and geographic patterns of Hispanic households, with few exceptions,
are not accurately captured by zip codes. By using zip + 4 as the reporting unit,
experts and practitioners would t>e able to analyze whether insurance coverage
amount, price, and quality vaiy according to the demographics of neighborhoods,
in particular whether it varies according to the race and income of a neighbor
hood.
• Loss Data: In order to determine whether differences in the cost of insurance re
flect real loss experience or racial prejudice and stereotypes, information regard
ing losses must be made available for analysis.
• Coverage: The scope of insurance disclosure legislation should be broad. There is
no substantive basis for limiting disclosure of insurance practices to a few urban
areas. Excluding small cities and rural areas from the scope of the legislation
would amount to a Congressional finding that discrimination only occurs in large,
urban areas.
• Race and Ethnic Data: This information is essential for civil rights enforcement
purposes.
The Los Angeles riots were a painful reminder of the disaffection of minority com
munities who feel shut out, who have no access, and who feel that American institu
tions do not offer them the protections other Americans receive. It is the duty of
Congress to take immediate, substantive, pro-active steps to address illegal insur
ance discrimination and the consequences associated with denial of insurance cov
erage.
On behalf of the National Council of La Raza, we thank the Committee for its
consideration of our views, and we are ready to answer any questions you may have.
85
“ 10-94
TUE
18=33
ACORN
Kay 9, 1994
The Honorable Donald W. Riegle
Chairman
Committee on Banking, Housing, and Urban Affairs
U.S. Senate
Washington, D.c. 20510
Dear Chairman Riegle:
We write to urge you to support and move legislation this
year that would require insurance companies to disclose
information regarding where they write property insurance,
comparable to the information reported by banks and thrifts under
the Home Mortgage Disclosure Act (HMDA)
In 1968, the National Advisoi^ Panel on Insurance in Riot
Affected Areas found that ‘communities without insurance are
communities without hope." After the riots in Los Angeles in
1992, surveys found that the insurance crisis in minority and lowincome areas had not materially abated. Numerous studies by
academics, state insurance departments, the federal government and
advocacy organizations and recent litigation have demonstrated
that discrimination and redlining are pervasive in the insurance
industry.
Redlining on the basis of geographic location by the
insurance industry has a devastating impact on individuals and
communities. It crushes opportunities for homeownership and
economic mobility, and is a major impediment to the creation of
affordable housing and community development in minority and low
- and moderate-income neighborhoods.
We commend you for requesting a study by the General
Accounting Office on what data should be collected to examine
disparities in the availability, affordability, and accessibility
of property insurance. We believe that that recently released
report --together with mounting evidence of insurance
discrimination-- justify a forceful and immediate response by
Congress.
We believe that data collection including race, national
origin, and gender data would serve several purposes. First, it
would allow the industry, the Congress, and the public to assess
the scope and extent of disparities in the provision of property
insurance by neighborhood.
Second, it would contribute co an objective assessment of the
underlying causes of any such disparities. For over two decades,
the insurance industry has dismissed charges of redlining by
insisting that any disparities in availability or price of
insurance reflect differences in risk. Civil rights, community,
and consumer organizations, citing evidence based on statistical
data on losses, litigation, and testing, have argued that
86
Y-10--94
TUE
1 8= 34
ACORN
discrimination based on the racial characteristics of
neighborhoods exists, and is pervasive. Data collection will
allow this debate to be settled conclusively, and allow for
considered responses by policy makers at the state and federal
levels.
Third, data collection can materially assist efforts to
enforce state and federal laws prohibiting discrimination in the
provision of insurance. As a leading Congressional supporter of
the Home Mortgage Disclosure Act (HMDA) t we need not tell you how
useful this data can be to the banking agencies, HUD, and the
Justice Department in their efforts to eradicate lending
discrimination. Like HMDA data, data on insurance company
activities won’t by itself prove discrimination by individual
companies. It will, however, assist regulators and civil rights
organizations in targeting companies for investigation and
enforcement.
Finally, in the same way that the sunshine created by HMDA
has produced a revolution in mortgage lending, we are convinced
that simply requiring public disclosure of policy information will
result in significant changes in insurance industry practices.
We commend you for your staunch commitment over the years to
civil rights, community economic development, and consumer
protection. We hope that, under your leadership, strong and
comprehensive legislation in this area can be enacted this year.
87
- 94
TUE
18=34
ACORN
Sincerely,
Alliance
to End
American
Civil
Childhood Lead
Liberties
Union
Community
Organizations
Association
(ACORN)
of
Center
Community
for
Consumer
Federation
Consumers
Lawyers
Committee
of
for
Clv^il Rights
Civil
for
Rights
Defense
and
of La
the Advancement
Educational
Council
National
Fair
National
Insurance
Consumer
National
League
Cities
National
Low-income
Housing
of
Sousing
Coalition
Neighborhood
Coalition
Rican
Coalition
National
Urban
League
NETWORK:
A
Citizen's
United Xethodlst
Society
Public
Catholic
a New
Church ,
Social
Equality
Congress
Interest
inc.
organization
Puerto
for
Colored
Alliance
National
National
Fund,
of
Raza
National
organization
Under Law
Institute
National
U.S.
Vow
America
on
Association
NAACP Legal
Public
Reform
Change
Conference
XcAuley
National
(NAACP)
for
Union
Leadership
riie
Poisoning
Justice
Lobby
(O.N.E.)
watch
General
Research
Board
Group
of
Church
and
People
An Analysis of Zip Code Distribution of
State Farm and Allstate Agents and Policies
in Chicago
ILLINOIS PUBLIC ACTION
April, 1993
89
AN ANALYSIS OP ZIP CODE DISTRIBUTION OF
STATE FARM AND ALLSTATE AGENTS AND POLICIES IN CHICAGO
Summary of Findings
STATE FARM AND ALLSTATE AGENTS DISPROPORTIONATELY CONCENTRATED
IN NORTHWEST AND SOUTHWEST SIDE ZIP CODES
State Farm and Allstate, the nation's two largest providers
of automobile and homeowners insurance, claim not to discrimi
nate in marketing or to redline Chicago neighborhoods.
However
their agents are disproportionately located on the city's
northwest and southwest sides:
* Nine northwest and three southwest side zip codes had five
or more State Farm agents while seven west side and seven
south side zips had one or no agents.
* Nine northwest, four southwest, and two far-south side zip
codes had five or more Allstate agents while seven west
side and seven south side zips had one or no agents.
STATE FARM AND ALLSTATE POLICIES ARE DISPROPORTIONATELY CON
CENTRATED IN NORTHWEST AND SOUTHWEST SIDE ZIP CODES
Both State Farm and Allstate contend that the ,ack of agents
does not mean a community is underserved.
However the data on
auto and home policies provided by the two companies documents a
similar pattern of disproportionate distribution:
* Ten northwest and five southwest side zip codes had the
heaviest concentrations of State Farm policies while seven
west side and eight south side zips had the least.
* Six northwest, four southwest side, and four far-south
side zip codes had the heaviest concentrations of Allstate
policies while four west side and three near-south side
zips had the least.
POLICY DISTRIBUTIONS LARGELY REFLECT RACE AND INCOME PATTERNS OF
CHICAGO COMMUNITIES
The disproportionate distribution of both agents and
policies genera M y corresponds to the race and income make-up of
Chicago community areas:
* All of the seventeen zip codes with the greatest concen
trations of State Farm policies are in majority white
communities, while all but one of the sixteen zip codes
with the least concentration, outside the downtown area,
are in majority black or hispanic comoninities.
90
* Allstate had a similar concentration in the majority white
communities on the northwest and southwest side although a
significant proportion of its policies are in the higher
income majority black communities.
However most zip codes
with the least concentration of policies, outside the
downtown area, are in low income communities that are
majority black or hispanic.
STATE FARM AND ALLSTATE AGENTS AND POLICIES ARE DISPROPORTION
ATELY DISTRIBUTED BY CONGRESSIONAL DISTRICT
Congressional districts, which in Chicago largely adhere to
racial boundaries, show the same pattern of disproportionate
distribution of auto and home policies:
* Hie seven zip codes, excluding the Loop, in the
predominantly black Seventh District had 6 State Farm
agents and an average of 3,568 policies while the six zip
codes in the neighboring but predominantly white Fifth
District had 56 agents and average of 18,250 policies.
Allstate had 8 agents and an average of 3,124 policies in
the Seventh District zips but 42 agents and an average of
7,583 policies in the Fifth District zip codes.
* *nie four zip codes in the predominantly black Second
District had 4 State Farm agents and an average of 6,233
policies while the four zip codes in the neighboring but
predominantly white 'Riird District had 25 agents and
average of 1 1 , 1 1 1 policies.
Allstate had 10 agents and an average of 8,574 policies in
the Second District zips but 3 6 agents and an average of
9,154 policies in the Third District zip codes.
91
AN ANALYSIS OF ZIP CODE DISTRIBUTION OF
STATE FARM AND ALLSTATE AGENTS AND POLICIES IN CHICAGO
Introduction: AGENT DISTRIBUTION AND REDLINING
In testimony presented to the
Commerce, Consumer Protection and
and in previous studies, Illinois
the unequal distribution of State
Chicago neighborhoods.
U.S* House Subcommittee on
Competitiveness last month,
Public Action has documented
Farm and Allstate offices in
These analyses have consistently found a disproportionate
concentration of such offices in northwest and southwest side
communities that are predominantly white and middle income, and
a corresponding lack of those offices in communities on the west
and south side that are predominantly black and hispanic,
We have contended that this pattern of office location is a
result of discriminatory marketing, often referred as
"redlining", that makes it difficult for inter-city consumers to
shop for and purchase insurance coverage from State Farm and
Allstate. These two firms, who dominate market share both in
rilinois and nationwide for automobile and homeowners insurance,
generally charge the lowest rates.
As a consequence, many low-income minority Chicagoans are
forced into the hands of the "non-standard" carriers which
typically have the highest rates and the poorest records of
consumer service. With auto liability coverage mandated by law
and lenders requiring coverage for home mortgages, the poorest
consumers are forced to pay the most for their insurance,
Both State Farm and Allstate have challenged these analyses
and they insist that they do not engage in 'Useriminatory
marketing or other redlining practices. As tney have recently
stated to the Subcommittee;
"State Farm strives to offer our insurance products to the
general public in a manner that does not discriminate on the
basis of race, color, religion or national origin." (1)
■Allstate feels that it is accessible in every neighborhood
in the city of Chicago." (2)
Further, both firms claim that because agents are not
restricted to selling policies in their i.Tjneaiate vicinity and
are willing to provide pricing information over the phone,
office locations are not meaningful:
92
2
■Just because there is not an agent's office in a particular
area does not mean the service provided to consumers living
in that area is inf erior....Regardless of whether or not a
State Farm agent's office is located in a particular zip
code. State Farm insurer cars and homes on a non-discriminatory basis throughout the city." (3)
■Simply because an agent is not located on a given block or
in a particular neighborhood, it should not be construed
that we (Allstate) do not write or are seeking to avoid
business in that particular location." (4)
However the following analysis substantially contradicts the
claims of both companies.
Based on the zip code data on
policies and agent locations provided by State Farm and
Allstate, it is clear that:
* the marketing of policies is closely correlated with the
geographic distribution of agents and offices;
* that policies are correspondingly concentrated in the
communities with the most agents and lacking in those
communities with few or no agents; and
* this uneven distribution of agents and policies largely
reflects the r a c i a l and income make-up of those
communities.
AGENT DISTRIBUTION BY ZIP CODE
As indicated by previous analyses of advertised office
locations, both State Farm and Allstate have widely uneven
distributions of agents in Chicago communities.
State Farm lists 186 agents ("independent contractors" and
trainee agents) as having had offices in 5 8 Chicago zip codes
curing the last five years.
Although this was an average of
slightly more than three per zip, the distribution ranges from
no agents in a dozen west side and south side zip codes to as
many as twenty in one northwest side zip code.
Overall most
State Farm agents are concentrated in ten north and northwest,
three southwest and one far southeast side zip codes. (Map I)
Allstate lists 215 agents as of February 28, 1993 located in
57 Chicago zip codes (or an average of slightly less than four
per zip code.)
However a similar pattern of uneven distribution
is evident -- ranging from several zip codes on the northwest
and southwest sides with more than a dozen agents to ten zip
codes on the south and west sides with no agents.
93
3
As with State Farm, most Allstate agents are ,oca ted on the
northwest side followed by the southwest side zip codes;
although Allstate has a substantially larger number of agents on
the far south side side.
(Map II)
Agent distribution is closely correlated with policy
distribution for both companies. Although zip codes are not a
precise designation of the immediate service area (not
infrequently an office located on the border of one zip code
markets to policyholders across the street in the neighboring
zip)
the zip codes with the most agents generally have the most
policies, and these with the fewest agents have fewer policies.
TABLE I -- AGENTS AND POLICIES PER ZIP CODE
State Farm (auto/1991)
(#)
Total
Policies
Zip Codes with
0 to 1 agents
19
60,200
3 ,168
Zip Codes with
2 to 4 agents
19
107.139
5,639
Z ip Codes w ith
5 or more a g en ts
14
147,503
10,536
Average
per Zip Code
Allstate (auto 1 iability/1992)
(*)
Total
Policies
Average
per Zip Code
Zip Codes with
0 to 1 agents
18
28,048
1,558
Zip Codes with
2 to 4 agents
18
57,159
3,572
Zip Codes with
5 or more agents
16
63,958
3,762
NOTE •- data for both firms excludes the downtown
zip codes 60601-60606 and O'Hare (60666)
SOURCES: State Farm, Allstate —
see Appendix A
94
95
96
While it snoulc come as no surprise that ¿cents ¿r.d offices
are .ocatea in neignbornooas with the most pol icyr.clcers, both
State Farm and Allstate nave maintained that agent location does
not indicate discriminatory marketing.
However the zip code
data provided by these companies clearly demonstrate that the
number of policies is closely correlated with the geographic
distribution of agents and offices.
POLICY DISTRIBUTION BY ZIP CODE
State Farm Policies by Zip Code
Based on the exposure data provided to the Subcommittee,
State Farm Mutual had 316,710 policies in the City of Chicago
during 1991.
Map III illustrates the distribution of those
policies by zip code.
As is immediately apparent, State Farm policyholders are
concentrated in the same communities as their agents.
Fourteen
north side, five southwest side, one far-south side and one
southeast side zip codes had 125% or more of the average number
of policies. At the same time, ten south side and eight west
side zip codes had less than 75% of the average.
State Farm Fire & Casualty lists 176,843 homeowners policy
exposures in 1991 for Chicago.
These were distributed similarly
as seen on Map IV, Twelve north side, four southwest side and
one southeast side zip codes had more than 125% of the average
number of policies.
At the same time, nine south side and six
west side zip codes had less than 75% of the average,
Racial and income breakdowns by zip code are not readily
available, but there are data for corresponding community areas.
(See Appendix B) Zip codes do not precisely correspond with the
community areas, and several include diverse communities, eg.
60608 includes both the white and average income community of
Bridgeport and the hispanic and lower-income lower West Side
community (Pilsen) . However the community area data allow for
some comparisons of policy distribution by race and income.
The zip codes with the highest concentration of State Farm
policyholders are ail majority white community areas, while
almost ail of zip codes with the fewest average policies
(outside the downtown Loop and Near North areas) are either
majority black or hispanic,
Most of the zip codes with the fewest State Farm policies
are low-income.
However there is no necessary correspondence-
97
8 4-05 1 0
-
9 4 - 5
98
99
9
with income: eigne of the northwest side zip codes w: zr. the
highest concentrations of policies are average income, while
only one cf the five far-south side zip codes representing
average income areas had a high concentration of policies.
Although there is a larger number of State Farm policies in
average income majority black communities than low-income areas,
the numbers are not at all comparable to these in similar income
majority white communities.
Allstate Policies by Zip Code
Allstate auto and home policies also closely track agent
locations. Allstate had 149*930 automobile liability policies,
as measured in written car years, in Chicago during 1992,
distributed over 57 zip codes as illustrated by Map V,
Again there was a heavy concentration in the same northwest
side, the southwest side and southeast sice zip codes as with
State Farm. However Allstate also had a larger proportion on
its policies in the five far-south side zip codes and a smaller
percentage on the far north side.
The distribution of residential property insurance,
including both homeowners and renters insurance, was virtually
the same based on the data provided to the Subcommittee.
Allstate had some 151 ,493 such new or renewed policies ir.
Chicago in 1992 — see Map VI.
In general, Allstate's policy distribution corresponds more
closely with income distribution than race. While almost all of
the zip codes (outside of downtown) with the lowest percentage
of Allstate policyholders were either majority black or
hispanic, these were also communities with the lowest incomes.
There was a higher concentration of both auto and home
policies in the majority black zip codes representing average
income communities.
(As indicated above, Allstate has a higher
number of agents in those areas while having only a handful in
the lower-income zip codes.)
POLICY DISTRIBUTION BY CONGRESSIONAL DISTRICT
Another revealing approach to analyzing the zip code policy
data provided by State Farm and Allstate is by congressional
district.
Following last year's redistrieting, congressional
districts closely follow racial boundaries, particularly in
Chicago.
i ni s
i i a si i t iUMÀMàLÈ
**J*H
100
101
102
-ip code alignments, of course, do not necessarily
correspond to the boundaries of the new congressional districts,
and often a zip code nay be part of two or more sucn districts.
As a consequence, we examined che number of policies ir. only
che 47 zip codes which are wholly or predominantly ir. a
congressional district; the information was not available to
accurately apportion the multi-district zip codes.
Tables II and III summarize the policy information by
congressional district (see Appendix C for full data.)
While
the seven Chicago congressional districts represent varying
proportions of the cities (and thus differing number of zip
codes)
several interesting comparisons are possible.
In the four south side zip codes of the predominantly black
Second District, there were just four State Farm agents, and an
average of 3^767 auto policies and 2,466 home policies.
By
contrast, the four southwest side zip codes in the
overwhelmingly white Third District had 25 agents and averaged
llr766 auto and 6,011 home policies -- nearly three times that
of the Second District.
Similarly, in the predominantly black Seventh District,
excluding the Loop six zip codes but even including the Near
North communities, there were just 6 State Farm agents in seven
zip codes.
These zips averaged 2,041 auto policies and 1,527
home policies.
The five zip codes in the adjacent and predominantly
nispanic Fourth District had just eight agents? they averaged
3,96 5 auto and 2,5 60 home policies.
However the neighboring six
zip codes in the overwhelmingly white Fifth District had 56
agents? and an average of 11,83 5 auto and 6,415 home policies —
more than five times the average per zip code in the Seventh
District and two-and-a-half times the average in the Fourth.
The pattern is similar, if not as extreme, for Allstate.
While the number of agents was greater in the Third District zip
codes that the Second, and the average number of policies, both
home in auto, was much closer,
However the non-Loop zip codes in the Seventh District had
only Allstate eight agents, averaging just 1,325 auto and 1,799
home policies.
In the Fourth District, there were six agents
and an average of 2,417 auto and 2,677 home policies.
By contrast, in
an average of 3.924
more than twice the
greater than in the
the Fifth District there were 42 agents and
auto policies and 3,6 59 home policies -average in the Seventh and more than 5 0%
Fourth.
103
13
TABLE II -- STATE FARM POLICIES BY CONGRESSIONAL DISTRICT IN
CHICAGO
AUTO/1991
Cong.
District
Zip Codes
in District*
Total
Policies
Average
Pol./ZiD
Listed
Aaents
1
2
7
25,629
3,661
10
4
15,067
3,767
4
3
4
47,065
11,766
25
4
5
19,825
3 ,965
5
6
71,008
11,835
7
6 downtown
1,868
9
8
56
7 other
14,290
311
2,041
6
8
51,730
6,466
28
Zip Codes
in District*
Total
Policies
Average
POI./Zip
Listed
Agents
9
HOME/1991
Cong.
District
1
2
7
12,4 87
1,784
10
4
9,863
2,466
4
3
4
24,044
6,011
25
4
c
12,801
2,560
5
6
38,491
6,415
56
7
6 downtown
9
8
1,470
10,688
245
1,527
9
7 other
8
27 ,212
3,402
28
6
Includes zip codes only whoil,y or predominantly in a
congressional district; does not include partial zip codes.
104
TABLE III --
AUTO
ALLSTATE POLICIES BY CONGRESSIONAL DISTRICT IN
CHICAGO
(Liability)/1992
Cong.
District
Zip Codes
in District*
Total
Pol icies
Average
Pol./Zip
Listed
Aqents
1
7
20,221
2,889
14
2
4
17,130
4,282
10
3
4
20,028
5,007
36
4
5
12•085
2,417
6
5
6
23,546
3,924
42
770
9,278
128
1,325
20
8
15,240
1,905
35
Total
Policies
Average
Pol./Zip
Listed
Agents
7
9
6 downtown
7 other
8
HOME (Homeowners & Renters)/1992
Cong.
District
Zip Codes
in District*
1
7
21,405
3,058
14
2
4
17,170
4,292
10
3
4
16,588
4,147
36
4
5
13,387
2,677
6
5
6
21,952
3,659
42
853
12,595
142
1,799
20
8
16,482
2,060
35
7
9
6 downtown
7 other
8
* Includes zip codes only wholly or predominantly in a
congressional district; does not include partial zip codes.
105
15
As with the community area data, the congressional district
analyses demonstrate the same pattern of uneven distribution of
agents and policies, which largely reflect the racial and income
make-up of those districts.
A Concluding Note
Only with policy information by census tract could a more
precise correlation between agents, policies, race and income be
possible. However the zip code data presented to the Subcommit
tee does provide a clear picture of the disproportionate
concentration of policies in predominantly white and middleincome northwest and southwest side communities and a
corresponding lack of those policies in communities on the west
and south side that are predominantly black or hispanic.
And
that disproportionate distribution closely reflects the office
locations of State Farm and Allstate agents.
NOTES
(1) Quoted from correspondence to Representative Cardiss
Collins, Chair of the Subcommittee on Commerce, Consumer
Protection and Competitiveness, U.S. House of Represen
tatives from Cranford A. Ingham, Vice-President and General
Counsel. State Farm Mutual Automobile Insurance Company,
dated March 30, 1993 ,
(2) Quoted from correspondence to Representative Cardiss
Collins, Chair of the Subcommittee on Commerce, Consumer
Protection and Competitiveness, U.S. House of Represen
tatives from Allstate Insurance Company, March, 1993
(3) State Farm, op cit,
(4) Allstate, op cit.
All data subsequently cited on agent locations, automobile
and home insurance policies by zip codes from .information
provided to the Subcommittee in these correspondences.
Data in
Appendix B on community areas from the Census Bureau, U.S.
Department of Commerce.
106
APPENDIX A —
Agents and Policies by Zip Code
STATE FARM (auto/1991)
Zip Codes with
0 to 1 Agents
Policies
Zip
Zip Codes with
2 to 4 Agents
Policies
Zip
Zip Codes with
5 or More Aaents
Zip
Policies
60607
60608
60612
60620
60621
60622
60624
60627
60633
60636
60637
60642
60644
60648
60649
60651
60653
60657
60658
1105
4076
1016
5899
1026
3818
1017
2148
2669
1829
2516
5184
1961
8410
2033
3098
436
7430
4529
60609
60610
60611
60613
60615
60616
60619.
60623
60626
60628
60632
60635
60640
60643
60645
60647
60655
60656
60660
60614
60617
60618
60625
60629
60630
60631
60634
60638
60639
60641
60646
60652
60659
9013
10303
11001
9203
13739
11809
7175
18925
14171
7258
10398
9417
9769
5322
(14)
147,503
(19)
60,200
Average : 3,168
(19)
3942
3506
2587
5565
3303
3861
4614
3240
4000
5191
9386
9021
5563
7808
7322
4749
7543
11660
4278
107,139
Average : 5,639
Average: 10,536
107
APPENDIX A /2
ALLSTATE
(auto 1 iabil
ty/1992)
Zip Codes with
0 to 1 Agents
Zip
Policies
Zip Codes with
2 to 4 Agents
Zip
Policies
Zip Codes with
.5 or More Agents
Zip
Policies
60608
60609
60611
60612
60615
60621
60622
60623
60624
60626
60627
60636
60637
60642
60644
60648
60651
60653
2196
2150
890
824
2297
1988
1569
2414
1536
1513
181
2411
2321
211
2060
206
2305
976
60607
60613
60616
60617
60619
60625
60628
60630
60633
60638
60640
60645
60647
60649
60657
60660
417
2576
2074
6607
87 01
3576
7082
4542
626
4828
2154
2341
3756
3038
3081
1706
60610
60614
60618
60620
60629
60631
60632
60634
60635
60639
60641
60643
60646
60652
60655
60656
60659
1246
3049
5642
7456
67 02
2285
4773
5421
1430
3699
4622
4469
2406
3729
2670
2306
2053
(16)
57.159
(17)
63,958
(18)
28,048
Average: 1,55 8
Average : 3,57 2
Average:
3,7 62
108
109
110
APPENDIX C
Chicago Agents and Policies by Congressional
District
State Farm (1991)
FIRST DISTRICT
Agents
Auto
Policies
Home
?olicies
60615
60619
60637
60642
60649
60653
60655
2
4
0
0
I
0
3
3303
4614
2516
5184
2033
43 6
7543
1608
2382
1 102
2554
1094
153
3594
TOTAL
1.1
25629
12487
3661
1784
Agents
Auto
Policies
Home
Policies
60620
60627
60628
60636
1
0
2
1
5899
2148
5191
1829
3662
1134
3757
1310
TOTAL
4
15067
9863
Av./Zip
1
3767
2466
Agents
Auto
Policies
Home
Policies
Zip Code
Av./Zip
SECOND DISTRICT
Zip Code
THIRD DISTRICT
Zip Code
60629
60632
60638
60652
7
4
5
9
13739
9386
14171
9769
7639
5230
6399
4776
TOTAL
25
47065
24044
A v ./Z ip
6.2
11766
6011
Ill
APPENDIX C /2
FOURTH DISTRICT
Agents
Auto
Policies
Home
Pol i d €
60608
60609
60622
60623
60647
1
3
0
2
2
4076
3942
3818
3240
4749
2192
2356
2026
2640
3587
TOTAL
8
19825
12801
3965
2560
Auto
Policies
Home
Policies
Z ip Code
1.6
Av./Zip
FIFTH DISTRICT
Zip Code
Agents
60614
60625
60630
60634
60641
60656
5
6
12
20
10
3
9013
9203
11809
18925
10398
11660
7432
4247
6144
9577
5459
5632
TOTAL
56
71008
38491
Av./Zip
9,3
11835
6415
SEVENTH DISTRICT
Home
Pol ici<
Zip Code
Agents
Auto
Policies
60601-06
9
1868
1470
60607
60610
60611
60612
60624
60644
60651
0
2
3
0
0
1
0
1105
3506
2587
1016
1017
1961
3098
7 07
2991
2268
563
47 5
1160
2524
TOTAL (non-Loop)
6
14290
10688
0 .9
2041
1527
A v ./Z ip
(n o n -L o o p )
112
APPENDIX C/3
NINTH DISTRICT
Zip Code
Agents
Auto
Policies
Home
Policies
60613
60626
60631
60640
60645
60646
60648
60660
2
3
6
2
3
10
0
2
5565
4000
7175
5563
7322
9417
8410
4278
3596
2407
3603
2589
3945
4558
4033
2481
TOTAL
28
51730
27212
Av./Zip
3.5
6466
3402
113
APPENDIX C /4
Allstate (1992)
FIRST DISTRICT
Agents
Auto
Policies
60615
60619
60637
60642
60649
60653
60655
1
3
0
0
3
0
7
2297
87 01
2391
211
3038
97 6
2607
3929
8064
2297
1432
2933
655
2095
TOTAL
14
2 0221
21405
2
2889
3058
Agents
Auto
Policies
60620
60627
60628
60636
7
0
2
1
7456
181
7082
2411
6219
1080
6743
3128
TOTAL
10
17130
17170
2.5
4282
4292
Agents
Auto
Policies
Zip Code
Av ./Zip
Home
Polici<
SECOND DISTRICT
Zip Code
Av./Zip
Home
Pol ici<
THIRD DISTRICT
Zip Code
Home
Pol ici'
60629
60632
60638
60652
8
8
4
16
6702
4773
4824
3729
6272
3893
3633
2790
TOTAL
36
20028
16588
9
5007
4147
Av./Zip
114
APPENDIX C /5
FOURTH DISTRICT
Zip Code
Agents
Auto
Policies
Home
Policies
60608
60609
60622
60623
60647
2
0
1
0
3
2196
2150
1569
2414
3756
2773
2441
1916
2733
3524
TOTAL
6
12085
13387
2417
2677
Auto
Policies
Home
Policies
4264
3448
3188
4742
3442
2868
1. 2
Av./Zip
FIFTH DISTRICT
Zip Code
Agents
60614
60625
60630
60634
60641
60656
3
13
5
3049
3576
4542
5451
4622
2306
TOTAL
42
23546
21952
7
3924
3659
Agents
Auto
Policies
Home
Policies
9
2
10
Av./Zip
SEVENTH DISTRICT
Zip Code
60601-06
606 07
60610
60611
60612
60624
60644
60651
TOTAL
(non-Loop)
A v ./Z ip
(n o n - L o o p )
20
770
853
2
5
1
0
0
0
0
417
1246
890
824
1536
2060
2305
395
2047
1476
480
1604
2454
4141
8
9278
12595
1.1
13 25
1799
115
APPENDIX C/ 6
NINTH DISTRICT
Zip Code
Agents
Auto
Policies
Home
Policies
60613
60626
60631
60640
60645
60646
60648
60660
4
i_
14
5
2
7
0
2
2576
1513
2285
2153
2341
2406
206
1760
2 877
1427
2432
2131
2499
2402
1056
1658
TOTAL
35
15240
16482
Av. / Zip
4.4
1905
2060
116
May 10, 1994
The Honorable Donald W. Riegle
Chairman Committee on Banking, Housing and Urban Affairs
The United States Senate
Washington, D.C. 20610-8076
Dear Senator Riegle:
It is with deep personal regret that I am unable to attend the
hearing on Insurance Redlining scheduled for tomorrow, due to my
commitments concerning health care reform measures in this state.
I am enclosing our March 1993 study as you have requested.
In
addition, my staff has prepared new tables which will be of
interest to your committee.
Since the Missouri Department of Insurance has completed several
market conduct exams on the issue of affordability and availability
of insurance. Below are some observations drawn from those exams.
First, there is an essential need for data.
Missouri began to
collect homeowners insurance and private passenger automobile
insurance data by zip code since 1979 under our statute, Section
374.400, Revised Statutes of Missouri.
Without an adequate data
base, it is difficult to monitor the degree of competition in the
individual neighborhoods (zip codes) of this state and to know
which companies ,write what type of coverage in what geographic
location.
The department needs to be able to direct its efforts
intelligently to the problem areas or target the companies that
appear to reduce coverage for protected classes of business.
Second, there is a need to monitor competition for personal lines.
In accordance with the McCarran Ferguson Act of 1945 we believe
that if open competition is employed by a state for rate
regulation, then "active supervision'1 must be implemented.
It is
by monitoring competition and its effects that we are able to
ensure that all markets are being served in this state effectively.
Our mode of monitoring competition and policing the protections
provided in the Unfair Trade Practices Act has been by zip code
analysis.
^here is pending legislation before the U.S. Congress
that supports this position. (Kennedy H.R.1257 and Collins HR
1188)
From our analysis and market conduct examinations of
insurers we have seen that there is a problem in the accessibility
Harry S Trum an State O ffice Building. Room 6 3 0 • 301 W est High • Jefferson City, M issouri 651 0 1
Telephone 3 1 4 /7 5 1 -41 26 or TDD 3 1 4 /5 2 6 -4 5 3 6 (Hearing Impaired)
117
The Honorable Donald W. Riegle
May 10, 1994
Page 2
of insurer placed agents, in the availability of coverage due to
underwriting rules and in the affordability of coverage fairly
priced.
Using our data on agent appointments by zip code and consequent
number of policies written,
we found that a strong significant
positive correlation existed in that the more agents in a locale,
the more policies written.
After targeting several exams of major writers of homeowners
coverages in our state we have learned that several underwriting
variables are being used as surrogates for valid criteria.
Variables such as the age of the home is used as a substitute for
the condition of residence. The condition of the property has an
intuitive fairness in that it is tied to the probability of loss;
however, a false assumption exists in that older homes must be in
poor condition.
Such a standard overlooks the effects of
rehabilitation of older properties such that their heavier
construction is more resistant to high winds than a new home.
Another variable used is the minimum amount of insurance coverage
which presumes that a smaller home is more likely to suffer a loss
than a larger, more expensive property. Tornadoes, of course, are
insensitive to the value or size of a home.
Larger homes present
greater exposure to risk, as any insurer's rate table will reflect.
Another variable is the relationship of replacement cost to market
value of a property.
This is unnecessary since market value
policies have been developed to address this issue.
Most insurers have undertaken to use variables to accept or reject
a risk; variables that are convenient though are in no way accurate
in assessing the risk of the loss.
Use of such variables create
dislocation
in the competitive market
by
leaving
various
populations without coverage with little or no concern as to the
impact upon neighborhoods since these variables are thought of as
"objective". No test of disparate impact has ever been applied in
any states' Unfair Trade Practice laws for insurance.
As to the matter of affordability of coverage, we have done some
initial evaluations of territorial rating structures in light of
five years of loss cost data from over 95% of the Missouri market.
We have employed analysis of variance techniques and multiple
regressions in order to defend territorial assignments.
To date,
our conclusion is that the primary force in territorial assignments
118
Honorable Donald W. Riegle
May 10, 1994
Page 3
is a market driven desire to seek and avoid segments of the
population through pricing mechanisms. This type of analysis can
only be accomplished through the collection of loss data by zip
code as well as exposure data.
Since states are unwilling or unprepared to monitor competition in
their markets, including the collection of loss data to evaluate
the fairness of rate structures as they exist, we strongly support
the concept of a federal mandate as in H.R. 1257 and H.R. 1188.
si# e5y1
jfk Y ANGOFF
Director of Insurance
State of Missouri
JA:dh
Enclosure
Missouri Department of Insurance
Market Excluded by Minimum Value
Policy Underwriting Rule
ZIP CO DE
C ategory
% Houses
Less Than
$40,000
100% Urban
100% Rural
> 80% Non-white
Statewide
Source: Bureau of the Census
20.69
55.19
67.06
26.87
1990 Census
Missouri Department of Insurance
Market Excluded by Minimum Value
Policy Underwriting Rule
ZIP CO DE
Category
% Houses
Less Than
$50,000
100% Urban
100% Rural
> 80% Non-white
Statewide
Source: Bureau of the Census
34.14
67.88
83.03
38.83
1990 Census
Missouri Department of Insurance
Market Excluded by Age of Home
Using 35 Year Underwriting Rule
ZIP C O DE
C ategory
% Houses
Built Before
1960
100% Urban
100% Rural
> 80% Non-white
Statewide
Source: Bureau of the Census
68.67
38.17
90.02
43.48
1990 Census
ZIP CO DE
C ategory
100% Urban
100% Rural
> 80% Non-white
Statewide
% Houses
Built Before
1970
85.99
53.37
95.79
61.37
In our preliminary analysis of territory base rates, we
have found a surprisingly low correlation between
territorial base rates and territory loss experience
122
Missouri Department of Insurance
Market Excluded by Age of Home
Using 25 Year Underwriting Rule
123
PREPARED STATEMENT OP JAY ANGOFF
D ir e c t o r , M i s s o u r i I n s u r a n c e D e p a r t m e n t
I am vexy pleased to announce that almost all of the leading homeowners’ insur
ers in Missouri have voluntarily agreed to allow us to disclose the ratio between
their statewide market share and their high-minority zip code market shares. I aplaud them for their positive and helpful responses and for the spirit of cooperation
ley have shown.
They have also raised some important issues that should be given serious consid
eration by anyone interested in now the homeowners’ insurance market functions.
These issues include:
1. Whether an insurer who is located in a certain region of the State should be
required to insure people throughout the State. Some insurers specialize in insuring
people in the region in which they’re located. They employ people in that region ana
they know that region well. Is there anything wrong with such insurers limiting
themselves to insuring people in their own region?
2. Whether cost per $1,000 of premium is the best way to measure the cost of insur
ance. Several insurers accurately point out that lower amounts of insurance cost
more per $1,000 of insurance than do higher amounts of insurance; and that lowincome people, who presumably buy lesser amounts of insurance, would therefore
pay more per $1,000 of coverage. On the other hand, this argument does not justify
lower loss ratios (i.e., amount paid out per premium dollar) for lower amounts of
coverage than for higher amounts of coverage.
3. Whether Missouri law authorizes insurers to refuse to write in certain zip codes.
Subsection (c) of section 375.936(11) of the Missouri code prohibits refusing to insure
based on geographic location. Subsection (g) prohibits refusing to insure based on
race. Subsection (i)a, however, provides that the prohibitions contained in sub
sections (c) and (g) do not apply if the refusal to insure “is for a business purpose
which is not a mere pretext for unfair discrimination.” If some insurers do refuse
to write in certain zip codes, therefore, are they merely doing what the law permits
them to do?
4. Whether insurers who do have a substantial presence in urban areas should be
penalized for “cheny-picking” the market If a limited number of insurers are willing
to write substantial business in urban areas, is it fair to them to require them to
insure all uifcan business?
5. Whether, as a practical matter, insurers can be forced to write business in areas
in which they don't wish to write. Regardless of what the law provides, and regard
less of the context to which the State disseminates comparative price information
or information about how to contact insurers, if an insurance company has no
agents in certain areas will residents of those areas buy insurance from that com
pany?
The Department believes that all the above are important questions which do not
have easy answers. We look forward to a continued discussion on homeowners’ in
surance issues in hopes that these issues can be resolved in a manner that will ben
efit both the public and the insurance industry.
S
HOMEOWNERS9INSURANCE IN MISSOURI
Introduction
Statistical data on the availability and affordability of homeowners’ insurance in
Missouri has been available for years, but has never been extensively studied.
This report is a synopsis of an initial study undertaken by the Missouri Depart
ment of Insurance to analyze the information reported by the insurance industry to
the department regarding rates charged, claims paid, and types of policies sola in
low-income areas.
The report compares the experience in low-income zip codes in St. Louis and Kan
sas City to statewide market experience. It also compares what is happening in the
insurance market in predominantly black low-income neighborhoods and predomi
nantly white low-income neighborhoods.
Background
Missouri is one of the few States in the Nation with extensive and detailed data
on homeowners’ insurance.
In 1978, the Missouri legislature enacted a law directing the Missouri Department
of Insurance to collect data on homeowners’ insurance on an annual basis. The fol
lowing year a similar law was enacted requiring collection of auto insurance data.
124
These laws provide that each insurer writing business in the State is required to
report all premium and loss data to the director of the department. The laws also
state that the director shall establish the statistical base for the reporting of the
information.
Pursuant to those laws, Missouri has required that insurance companies report
statistical information by zip code. The data to be reported includes the number of
policies a company writes, the amount of premium it receives, the amount of claims
it pays, and the amount it pays out in claims.
Neither of the statutes regarding the collection of the data state that it will re
main confidential. In fact, the automobile insurance disclosure law expressly states
that the director shall make reports of the data available to the public. Neverthe
less, due to the claims of insurers that this information was proprietary, prior Ad
ministrations verbally agreed that information that identified companies would re
main confidential.
That agreement was put in writing in 1987 in a departmental bulletin issued by
Director Lewis Crist. This bulletin, which does not have the force of law, stated that
homeowners’ and auto insurance data submitted to the department "will not be dis
tributed to anyone other than the submitting insurer or a member of its group ex
cept on an aggregate (total of all insurers reporting) basis.”
In view ofth e understanding that has existed Detween the department and the
insurance industry, the department believes that at this time it would be unsporting
to identify individual company data, or to release the data in a manner that would
enable the reader, by using other publicly available data, to identify individual com
pany data. Therefore, this report contains no data that would enable a reader to
identify any individual insurer data.
Findings
This study is based upon the last 5 years of information filed by insurance compa
nies writing homeowners’ insurance in Missouri.
Table 1 sets out the average rates charged, the types of policies sold, and the
losses paid in minority and white low-income zip codes in St. Louis. That data is
compared to statewide data for all incomes and all zip codes.
As this table demonstrates, St. Louis policyholders in the eight minority low-income urban zip codes paid an average o f $6.15 for every thousand dollars of insur
ance coverage. Policyholders in the three white low-income zip codes paid $4.70 per
thousand. On a statewide basis, policyholders paid $4.27.
In addition, differences exist in coverage costs within the minority zip codes. For
example, the cost per thousand ranged from a low of $5.64 in zip code 63133, which
is 82 percent minority, to a high of $7.77 per thousand in zip code 63106, which
is 94 percent minority.
Substantial differences also exist in the types of insurance policies sold in the mi
nority and white low-income zip codes. For example, 74 percent of all policies sold
in minority zip codes were limited policies, whereas, 41 percent of policies sold in
white zip codes were limited policies. Only 23 percent of policyholders bought lim
ited policies on a statewide basis.
So-called limited policies limit coverage in some way. Typically, a limited policy
insures the actual cash value of the home rather than its replacement cost. It may
also limit liability, theft, or contents coverage. In contrast, a standard policy typi
cally covers the costs of replacing the dwelling and its contents, and provides com
prehensive liability coverage.
Finally, the amount of money insurers paid out to policyholders in claims differed
by zip code. For example, for every dollar insurers received in premium from policy
holders in minority low-income zip codes, they paid out 64 cents in claims—in insur
ance jargon—their loss ratio was 64 percent. In contrast, insurers paid out 66 cents
on the dollar in white low-income zip codes, and 58 cents on the dollar statewide.
Table 2 breaks out St. Louis cost and pay out information further by type of policy
purchased. For example, the average cost for a limited policy in minority zip codes
was $7.30 compared to $4.65 in white zip codes and $5.45 statewide. Yet the 57 per
cent loss ratio in minority zip codes was substantially lower than the 72 percent loss
ratio in white zip codes.
Tables 3 and 4 set out the same type of data for Kansas City as do tables 1 and
2 for St. Louis. In Kansas City, as in St. Louis, residents of minority low-income
zip codes paid more for coverage than both residents of white low-income zip codes
and Missouri residents in general. Also, insurers sold more limited policies in minor
ity low-income zip codes than they did in white low-income zip codes or statewide.
And their loss ratio on those policies—45 percent—was lower than it was in either
white low-income zip codes or statewide.
125
Table 5 ranks the top 20 homeowners’ insurers by ratio of statewide market share
to market share in minority low-income zip codes.
As the table indicates, Company A’s market share statewide was 62 times greater
than its market share in minority zip codes. Eleven insurers had statewide market
shares that were more than three times their minority market shares.
On the other hand, eight carriers had higher market shares in the minority zip
codes than they did in the State as a whole.
TABLE I
ST. LOUIS MSA
LOW INCOME URBAN ZIPS
S B Minority (Greater than 8 0 % Non-White)
63106
$12,324
63107
$19,113
$24,251
63112
63113
$19,086
$21,957
63115
63120
$22,143
63133
$23,601
63140
$16,507
Aine* t o ta ls *
White (Less than 2 0 %
63111
63118
63143
Non-White)
$22,527
$22,130
$25,509
94%
87%
86%
99%
99%
96%
82%
99%
5%
19%
16%
$7.77
$6.61
$6.32
$6.44
$5.72
$6.05
$5.64
$7.15
? n r * * »
12%
13%
31%
16%
31%
29%
50%
26%
$4.73
$4.84
$4.36
62%
46%
80%
88%
87%
69%
84%
69%
71%
50%
74%
i v r x m m ’n
ArenTotals
STATEWIDE
ALL INCOME - ALL ZIPS
38%
54%
20%
3.89%
4.73%
5.82%
742%
7.19%
12.97%
6.60%
5.23%
\¡ Ä 7 J
7.71%
7.23%
5.87%
gg i l l
ÙTVÌ1
iir r
$3,035
$2,585
$2,402
$1,643
$1,755
$1,343
$1,703
$3,571
72%
63%
61%
64%
58%
85%
52%
89%
'M l
i^ 6 4 %
$1,765
$2,021
$2,050
65%
72%
55%
127
TABLE 2
ST. LOUIS MSA - LOW INCOM E URBAN ZIPS
Minority (Greater than 80% Non-White)
$6.15
7.10%
All Policies
$4.77
9.70%
Standard Policies
$7.30
6.17%
Limited Policies
White (Less than 20% Non-Whitei
$4.70
7.15%
All Policies
$4.73
8.71%
Standard Policies
$4.65
4.95%
Limited Policies
$1,863
$2,172
$1,689
64%
76%
57%
$1,922
$1,677
$2,533
66 %
63%
72%
$1,838
$1,827
$1,913
58%
61%
47%
Statewide - All Z ìds
All Policies
lfandard Policies
ILimited Policies
$4.27
$4.10
$5.45
7.83%
8.93%
4.20%
TABLE 3
KANSAS CITY MSA
LOW INCOME URBAN ZIPS
I » « » CASHFLOW
to s s ,
'IBSl^FIm m c Æ i î\miÆraWflRATtO‘
mou$hho0>|M^p
zw
m c o m ^ r fo p h
Minority (Greater than 80% Non-White)
64130
$23,355
9 5%
$6.22
91%
$6.52
ju s jArc* Totals
"
$21*934 ' , >/' ' " 1!»«/*•>■ j M
White (Less than 20% Non-White)
16%
$5.65
64124
$22,682
64128
64120
64125
64123
64053
Area Totals
$20,512
$21,508
$23,547
$25,346
$23,682
15%
8%
8%
4%
$6.06
$5.99
$5.40
$4.97
45%
55%
8.58%
$1,823
35%
65%
8 .95%
$1,452
aîW
55%
'' * ;
MWÏ
54%
3 5%
4 6%
63%
65%
63%
7.34%
1.95%
7 .78%
8.0 8 %
5.73%
46%
65%
54%
37%
35%
$3,309
$4,426
$2,043
$2,846
____$1,950
60%
100%
4 1%
72%
89%
53%
$23,353
STATEWIDE
ALL INCOME - ALL ZIPS
¡Statewide
J $33AM'
* Derived from 1990 Census Data
-
tW
A
V
"* * *. * S*% |
129
TABLE 4
KANSAS CITY MSA - LOW INCOME URBAN ZIPS
Minority (Greater than 80% Non-White)
8.71%
$6.32
All Policies
Standard Policies
$5.45
11.44%
Limited Policies
$7.65
6.79%
White (Less than 20% Non-White)
7.14%
All Policies
$5.46
Standard Policies
$5.32
9.81%
$5.75
3.45%
Limited Policies
Statewide - All Zips
All Policies
$4.27
7.83%
Standard Policies
$4.10
8.93%
Limited Policies
$5.45
4.20%
8 4-05 1 O - 94 - 6
$1,685
$2,011
$1,298
60% 1
75%
45%
$2,807
$2,515
$3,957
84%
90%
72%
$1,838
$1,827
$1,913
58%
61%
47%
130
TABLE 5
Top 20 Writers
Company A
Company B
Company C
Company D
Company E
Company F
Company G
Company H
Company 1
Company J
Company K
Company L
Company M
Company N
Company 0
Company P
Company Q
Company R
Company S
Company T
Statewide
to
Minority
index
62.00
41.25
36.33
22.24
15.13
11.67
6.32
3.64
3.49
3.33
3.08
2.77
0.91
0.70
0.63
0.58
0.55
0.37
0.11
0.06
Statewide to Minority index - Statewide Market Share divided by Minority Market Share
Company A's market share is 62 times larger than its minority market share.
Table does not rank insurers by size.
131
STATEMENT OF THE
NAACP LEGAL DEFENSE AND EDUCATIONAL FUND, INC.
May 18,1994
This statement is submitted by the NAACP Legal Defense & Educational Fund,
Inc. (LDF) for the Senate Banking Committee’s Hearing on Homeowners’ Insurance
Discrimination.
LDF has been widely recognized for its contribution to civil rights enforcement
through litigation efforts to eradicate the causes and effects of racial discrimination
and segregation in all aspects of American life. Through litigation and legislative
activities, LDF has demonstrated a steadfast commitment to ensuring equal oppor
tunities in housing. LDF played an integral role in the passage of the 1988 Amend
ments to the Fair Housing Act and continues to enforce the Act through private liti
gation throughout the United States.
In accord with its commitment to guarantee equal opportunities in housing, LDF
is concerned that all Americans have access to homeowners’ insurance without re
gard to the race of individuals or the race or socio-economic status of the residents
within a particular neighborhood. LDF has gained a keen awareness of the exist
ence of widespread claims of discrimination in the provision of homeowners’ insur
ance through its work with fair housing groups, attorneys, insurance company em
ployees, ana State Insurance Commissioners.
As the Southern District of Ohio succinctly noted in McDiarmid v. Economy Fire
& Casualty, 604 F. Supp. 105, 107 (S.D. Ohio 1979), “It is elementary that without
insurance, mortgage financing will be unavailable, because a mortgage lender sim
ply will not lend money on the property. Without mortgage financing, homes cannot
be purchased. Thus, the availability of insurance and the ability to purchase a home
go nand in hand and vary, in direct proportion, to one another.” Accordingly, the
ideal of equal opportunities in housing cannot be achieved without ensuring that
Americans are not subjected to discrimination in the provision of homeowners’ in
surance.
Insurance discrimination not only impacts individuals who seek to obtain insur
ance, it also has a devastating effect on entire neighborhoods. Where insurance is
denied because of the racial composition of the neighborhoods, these neighborhoods
are more prone to become areas of desolation. Residents of these areas are less like
ly to attain the dream of home ownership, which ultimately has an adverse impact
on the economic viability of these communities. The few residents of these areas
who become homeowners without insurance are then left without any means for re
storing their property if damage or loss occurs, thus, possibly leaving abandoned or
ill-kept homes within these neighborhoods.
In addition to claims of individual discriminatory treatment by insurance compa
nies and agents, LDF has received information about several alleged practices by
insurance companies that may have the effect of discriminating on tne basis of race.
1. Minimum Value—We have been informed that many insurance companies have
underwriting guidelines which prohibit agents from writing policies for properties
valued under a certain dollar amount, such as $40,000. In a large number or local
ities such minimum value requirements may adversely affect a disproportionate
number of minorities. We are aware of no evidence that would justify use of such
requirements with a disproportionate adverse impact.
2. Maximum Age—Throughout the United States, minorities are concentrated in
urban areas; many of these minority areas encompass the oldest areas within these
cities. Thus, underwriting guidelines under which companies refuse to provide in
surance to individuals who live in older homes may also have a disparate impact
on minorities.
3. Inferior Policies—We have received allegations that individuals who reside in
minority communities are often given inferior homeowners’ policies, as compared
with those who live in predominantly white areas. These inferior policies limit a pol
icyholder's recovery for loss or damage to the amount at which tne property is val
ued on the market, whereas, residents of predominately white areas allegedly re
ceive replacement cost coverage. The insurance industry has not, to our knowledge,
provided any data or evidence to support such discriminatory availability of policy
types.
4. Cost Disparities—LDF also is aware of studies that suggest minorities and resi
dents of minority communities are likely to be charged significantly more for policies
than are residents of predominately white areas.
5. Marketing Practices—We have seen evidence that appears to indicate that in
surance marketing practices do not treat white and minority consumers equally. Al
leged discriminatory practices include failure to locate agents in minority neighbor
132
hoods and failure to advertise in minority areas. In addition, LDF has received com
plaints from insurance agents who were discouraged from writing policies in minor
ity neighborhoods or blatantly directed to discontinue selling in these areas.
LDFs experience and our consultation with other fair housing advocates has iden
tified the practices set out above as serious areas of concern for minority home
owners ana homeseekers. The amount of evidence and the number of complaints of
homeowners* insurance discrimination seem to be increasing. Yet, the insurance in
dustry adamantly denies that any discriminatoxy practices occur with regard to the
provision of homeowners’ insurance. This is precisely the factual dispute that ex
isted between advocates and the mortgage industry prior to the enactment of The
Home Mortgage Disclosure Act (HMDA). Our Nation’s experience under HMDA
leaves no doubt that legislation requiring data collection is needed to provide the
public and Federal agencies with information that would demonstrate the depth of
this problem and provide the basis from which fair housing enforcement efforts may
be made. In addition, data collection will provide the insurance companies with in
formation that will enable them to monitor themselves and will provide incentive
for these companies to change policies which may have a discriminatoxy impact.
The following types of information are essential to enable meaningful analysis of
the facts concerning claims of homeowners’ insurance discrimination:
• Reporting of the race of applicants and policyholders;
• Extensive geographic reporting throughout the country, inclusive of cities with
significant minority populations;
• Reporting within small enough areas to observe neighborhood activity, such as by
census tract;
• Loss data, i.e., amount and number of claims by neighborhood and data that
would justify rating territories (without this information, it would be impossible
to conclude whether or not legitimate risk factors account for any dearth of poli
cies or higher premium rates in minority areas); and
• Type of policies sold in each neighborhood and prices chained for these policies.
Too often, African-Americans are denied the opportunity to purchase homes be
cause of discrimination by realtors, sellers, and mortgage providers. Legislation has
given victims of housing discrimination the tools with which to challenge the dis
criminatory actions of these individuals and entities. However, discrimination in the
provision of homeowners’ insurance has not been as comprehensively studied and
addressed. Congressional intervention is needed now. in the form of strong legisla
tion which would provide the means for monitoring the actions of the insurance in
dustry. Without legislation, it will continue to be very difficult to assess the mag
nitude of the unequal treatment of minority communities by insurance companies
and, in all likelihood, Americans will continue to suffer from this form of discrimina
tion.
133
N A TIO N W ID E INSURANCE COM PANIES
O N E N A TIO N W ID E P LAZA. CO LU M B US . O H 43216
W. CRAIG 2IMPHER
ASSOCIATE VICE PRESI06NT
GOVERNMENT RELATIONS
May 12, 1994
The Honorable Donald W. Riegle
Chairman
Banking, Housing and Urban Affairs Committee
United States Senate
Washington, D.C. 20510
Re: Urban Insurance Availability
Dear Mr. Chairman:
As a corporate policy, the Nationwide Insurance Enterprise staunchly opposes
"redlining" Nationwide is the country’ s fourth largest auto insurer and fifth
largest homeowners insurer. We achieved those high rankings by actively
soliciting customers, not by arbitrarily rejecting them.
We have established underwriting guidelines which assure fair and equal
treatment for all insurance applicants, regardless of where they live, their
race, their age or their gender. Our corporate policy is to treat all applicants
for insurance alike — as prospective policyholders. We write homeowners
policies according to fair and sound underwriting guidelines, not according to
any arbitrary bias or unfair or unlawful discrimination, as some would have
you believe.
This hearing has been advertised as a hearing on "redlining” The more
appropriate term should be "urban insurance problems" Both "redlining" and
"discrimination" are tossed about loosely these days, as insurance industry
critics charge the industry with unfair practices involving the cost and
availability of insurance.
N A TIO NW IDE LIFE IN SU R A N C E CO
N A TIO N W ID E M UTUA L IN SU R A N C E CO
NA TIO N W IO E M UTUAL FIR E IN SU R A N C E CO
134
The Honorable Donald W. Riegle
Page 2
May 12, 1994
Some people confuse insurers’ responsibility for addressing financial security
needs with an unintended role of correcting social deficiencies. As a result,
insurers are commonly asked to insure the uninsurable. Using the insurance
mechanism to correct perceived social inequities will stretch the system
beyond its purpose and capabilities. Possibly the worst deception being
perpetrated today is the one that leads people to believe that social problems
can be solved within the insurance process at a price everyone can or will
want to pay.
Nationwide has a clear-cut and firm corporate policy prohibiting redlining.
The company underwrites all applicants individually to assure equal treatment
to make certain that all policies are properly priced. The company applies its
criteria evenly, no matter where a person lives - in a fancy suburb, a rural
area, a small town, big city, or wherever.
In conclusion, many inner-city areas suffer from certain urban insurance
problems which adversely affect the availability and affordability of insurance
products. These problems are not the result of "redlining'', but are caused by
a combination of historical, social, and economic factors the insurance
industry has not created. Nationwide pledges its support and commitment to
work with you and the others on the Committee to address the problems that
affect our nation’ s insurance market. We would look forward to a
cooperative and mutually respectful process toward that end.
Sincerely,
W. Craig Zimpher
cc:
Banking Committee Members
135
REPORTS
TRADEMARKS REG. U.S. PATENT OFFICE
Published by F-D-C Reports. Inc
55 YEARS
Devoted to Specialized Health Care Information
________ _
Health News Daily.
SPECIALIZED DAILY INFORMATION SERVICE FOR EXECUTIVES AND DECISION-MAKERS IN THE HEALTH INDUSTRIES. PUBLISHED BY R )-C
REPORTS. INC, 5550 FRIENDSHIP BLVD., SUITE ONE, CHEVY CHASEJMD 20815-7Z78. FAX 24-HOUR No. (301)656-309*; PHONE (301>657-9830
Vol. 6, NO. 91
C O N T E N T S CO PYRIG H TED ©
F-D-C REPORTS. INC.. 1994
Wednesday, May 11,1994
a m a ffla a a a e s E s a i
^
H ealth care infrastructure, com plexity of reform proposals m ust be considered in addition
to issue of costs, CBO Director Reischauer urges
Story b e lo w
S e n a te Labor & Hum an Resources C o m m ittee w ill report ou t health reform bll! by
Memorial Day, Sen. Kennedy says May 10
Page 2
S m ithK line B eecham w ill m an ufactu re and m arket generic Tagamet through Penn Labs
subsidiary and Lederle generics division
Page 3
Q uad generic trial: Shah guilty, Bansal acquitted M ay 9 after jury trial in Baltimore federal
court; government loses conspiracy case against former executives_____ _______________ _ Page 4
HEALTH CARE ADMINISTRATIVE COMPLEXITIES AS CRITICAL TO REFORM SUCCESS as cost containment,
Congressional Budget Office Director Robert Reischauer told reporters May 10 in Washington, D.C. Saying that
he is "still waiting" to see comprehensive health reform legislation that can achieve the goals of covering most
uninsured Americans with reasonable cost containment and in a workable framework, Reischauer warned: "The
architects of these health care reform proposals have to be careful that they are not creating systems that require
more institutional capacity and more administrative experience than we are capable of conjuring up/
The CBO director characterized the low-income and small-business subsidies central to President
Clinton’s Health Security Act (HR 3600, S 1757), Rep. Jim Cooper’s (D-Tenn.) Managed Competition Act (HR
3222, S 1579) and Sen. John Chafee’s (R-R.I.) HEART bill (HR 3704, S 1770) as "immensely complicated things
to administer." Senate Labor and Human Resources Committee Chairman Edward Kennedy’s (D-Mass.) proposal,
which will be marked up beginning May 18, could well be "equally if not more complex," Reischauer suggested
based on news accounts of the measure.
Discounts for low-income Americans and small, low-wage businesses "in effect are a welfare program,’'
Reischauer asserted. Noting that "premiums have to be paid to insurers on a monthly or quarterly basis," he
asked: "Are we determining income eligibility on a quarterly or monthly or annual basis? Is it prospective or
retrospective income that we’re worried about?"
Even once such issues have been resolved, "it’s going to be no easy task to make the system work,”
Reischauer declared, given the scope of the subsidies outlined in the major reform proposals. For example, CBO
has estimated that roughly half of the U.S. population could be eligible for discounts under the Administration’s
Subscription to F-O-C Reports 'Health News Oaily' (ISSN 1042-27811 is $1,195 per year.
Additional copies of 'Health News Oaily' $495 each year when mailed in same envelope with $1,195 subscription copy.
This newsletter is available electronically via Newsnet (800-345-1301), OataStar (800-221-7754), Dialog (800-334-2564) and BRS (800-289-4277).
Contents of "Health N e w s D aily“® are protected by the U.S. Copyright Law. Reproduction, photocopying, storage or transmission by m agnetic
or electronic means is s trictly prohibited by law . Authorization to photocopy items for internal or personal use is granted by F-O-C Reports, Inc..
w hen the fee of $2.00 per copy of each page is paid directly to Copyright Clearance Center, 222 Rosewood Or., Danvers, M A 01923 (508) 750-8400.
The Transactional Reporting Service fee code is: 1042-2781/94 $0.00 + $2.00. Violation of copyright w ill result in legal action, including c iv il and/or
crim inal penalties, and suspension of service.
136
2
Health News Daily
Wednesday, May 11, 1994
bill by the time it is fully implemented. Regarding the Managed Competition Act, about 60% of people enrolled
in purchasing cooperatives and 22% of those receiving insurance outside the cooperatives would likely qualify for
subsidies, he said. Reischauer added that the bill’s Health Care Standards Commission would conservatively be
processing 43 mil. applications for assistance every year.
CBO projects that the volume of information generated by subsidy programs would far outstrip the
capacity of current data-collection systems. "New systems are going to have to be developed to ascertain"
information that is not currently gathered, Reischauer said, such as the number of full-time employees working in
a given firm for a specified period of time, in order to determine insurance costs accurately, as well as whom to
bill for what portion of coverage. Such institutions "take a long time to build," Reischauer declared. "This is not
something we can plop into place in six months, or nine months, or a one-year period.*'
In a worst-case scenario, a poorly constructed reform bill hastily passed by Congress would not resolve
the problems of the current insurance market while effectively unraveling that system, Reischauer cautioned.
Noting that repeal of the law could follow, he suggested that lawmakers should take heed from their experience
with the Medicare Catastrophic Care Act and build in "clear benefits" that "materialize quickly for the American
people." Such'a "support group," the CBO director concluded, could help defray some of the inevitable protests
from insurers and providers that would surface following the enactment of comprehensive reform.
"Health reform is different" from almost all other legislation considered by Congress, Reischauer said,
explaining that "it has to be a single, integrated whole." When provisions such as mandatory alliances or an
employer mandate are dropped for political reasons, the result can be a transformation of "the entire workings of
the system you’re trying to put together." With this caveat in mind, Reischauer urged policymakers not to add or
subtract major provisions during floor debate. "We’re going to really need some time in this whole process...to at
least do some lab testing to see whether the thing works," he stated.
On the subject of upcoming CBO analyses, Reischauer reported that his agency would release shortly a
report on the Affordable Health Care Now Act (HR 3080) authored by House Minority Leader Robert Michel
(R-Ill.) and supported by GOP leadership. The analysis of Chafee’s HEART bill may be finished by the end of
June, he said, cautioning that "we are having our problems" modeling the economic impact of the legislation,
given that it would permit the insurance market to reorganize along several different lines. CBO will complete its
work on the House Ways and Means mark the week of May 9, Reischauer forecast, noting that the agency has
not done a "behavioral analysis" of that legislation.
Commenting on the Cooper bill, Reischauer said that although CBO’s recently released report concluded
that the bill’s revenues would not pay for its proposed subsidies, "it’s reparable in lots of different ways." The
issue of why the tax-cap revenues in HR 3222 are substantially lower in this year’s analysis compared to those
calculated for last year’s bill has not yet been resolved, he noted. In general, using tax-cap revenues to finance
health reform raises problems of equity, Reischauer asserted, since high-cost plans do not necessarily imply
high-quality coverage. Rather, some plans are expensive because they cover high-risk individuals, or because they
are located in areas of the country with higher living costs.
The task of designing a tax cap is "not a simple kind of thing, and a lot of people seem to be approaching
it in that way," the CBO director said. "There’s a whole bunch of other factors [besides plan costs] at play, and
until you are able to either make adjustments for that — or even the playing field — it doesn’ t make any sense" to
apply a tax cap to health insurance, he insisted. Overall, "I think there’s more work to be done" on all the plans
that have been introduced, Reischauer concluded.
SENATE LABOR COMMITTEE W ILL REPORT OUT REFORM BILL BY MEMORIAL DAY, Chairman Edward Kennedy
(D-Mass.) said May 10 following a speech before the American Federation of Teachers in Washington, D.C.
Refusing to comment on the voluntary alliance structure in his bill, a key difference with President Clinton’s
Health Security Act (S 1757), Kennedy nevertheless emphasized that his proposal would achieve universal
coverage, the White House’s bottom line.
© F-D-C Reports, Inc., 1994. Photocopying w ithout permission is strictly prohibited. See Page One.
Multiple copy rate: $495 when mailed in the same envelope with $1,195 subscription.______
137
Wednesday, May 11, 1994
Health News Daily
■3 ■
Although Kennedy would have a maximum of only eight legislative days to mark up his bill — considera
tion is scheduled to begin May 18 — committee staffers say the Massachusetts Democrat likely will hold to his
timetable. The fast-track approach indicates that Kennedy already may have the votes to pass his bill.
The 16-member labor panel, whose composition is substantially more liberal then the Finance Committee,
includes Democrats Tom Harkin (Iowa). Paul Wellstone (Minn.), Paul Simon (111.), Howard Metzenbaum (Ohio)
and Barbara M'ikulski (Md.). Republican Sen. Jim Jeffords (Vt), the lone GOP co-sponsor of S 1757 in either
chamber, also is a member of the committee, in addition to moderates Nancy Kassebaum (R-Kan.) and Dave
Durenberger (D-Minn.).
Administration officials note that the level of support Jeffords gives to the President on the Senate floor
will be critical. The Vermont Republican, who supports the state single-payer option in the White House plan,
could be affected by his state legislature’s recent rejection of a Canadian-style delivery system. While Senate
moderates are critical to breaking filibusters, Kennedy said that he does not expect Republican members to
employ the chamber’s parliamentary delay tactics.
"* In the area of federal subsidies, the Kennedy bill would offer support to workers based on individual
wages rather than firm size, the model contained in the President’s proposal. Administration officials say they are
not opposed to the Labor approach, noting that it achieves the goal of targeting needy low-income individuals and
giving employers the incentive to hire them.
A surprise speaker at the AFT meeting was President Clinton, who continued his attack on opponents of
the employer mandate and mandatory alliances. Noting that there are few choices for financing universal coverage
other than requiring firms to contribute to worker premiums, he asked rhetorically: "Should I raise your taxes as
an alternative?"
On the issue of consumer choice within alliances, Clinton maintained that his plan would offer Americans
a "minimum of three choices [of plan] a year, every year." That number could increase over time, he predicted,
stating that "if the evidence of the California [public employees] cooperative buying plan is any indication, we’ ll
get a lot more...They have 15 choices this year of plans, and everybody’s insurance rates went down. This is
about protecting an increasing choice, not about reducing choice."
SM1THKUNE BEECHAM WILL SUPPLY GENERIC TAGAMET TO LEDERLE through SB’ s Penn Labs subsidiary, the
company announced May 10. The generic cimetidine product will be manufactured by SB under the Penn Labs
label; SB will market the anti-ulcer medication to hospitals and managed-care organizations, while American
Cyanamid’ s Lederle Standard Products generic drug division will distribute SB’s generic cimetidine to wholesal
ers and chain and independent pharmacies.
^H^^<^^i2^^Mynagir^"Edhor.John^akotnik*i^ ^ ^ :!^^Assi^ntMa^jng'M^r‘T3eyer^orig"erv
^ ;fl^ ^ w ld h ^ l.^ble^almer%^^Da^dtBlue^^ ^'1^^Pu^^^^^an^MfWerw|^rjyr
^
____
‘^ ^ e a fth Policy &^Bmmedical Research News: Joh n Park^, Usa;White, AnnevPetniska^Kathenne.Pushkar^Kari^hl^d^
NonprescriptioriiPhannaceuticals & Nirtritionais News: Timothy Harrington,-TiilaMichaelutes, C ^ ^ n e 'H e jn i^ ® ®
v,.
¿^M edical Devices, Diagnostics ¿^Instrumentation News: Janrt Coleman/Randy.Buii(hoIdert< ^ ^ ^ t $
wJChnstine Harrington,Jon Dobson^FaithBusHnaq, J u ^ S t it i§ C jM a ^ ..B ^ e r .
( Toiletries. Fragrance & Skin Care News: Susan.Easton^Heather Swain,:Chris G lass^A u dre y.W itte m anv^^^
«. : ^ ^
Circulation Manager Edward Picken, 5550 Friendship Blvd."/Suite One, Chevy Chase, MD 20815-7278, Phohe: (301) 657-9830i
y»vF-D-C Reports Research Service ® ^Inform ation Research & Document Ser^\ce n o w avaUable%
■from F-D-C Reports. Contact Meghanne Malone, Research Services Manager, at <301) 657-9830.^
4
Health News Daily
Wednesday, May 11, 1994
SB will market generic cimetidine under the Tagamet NDA. The generic version will be yellow in order
to distinguish it from the pale green color of the branded product, the company said.
Lederle will begin to distribute the generic version of the H2 antagonist on May 17, when the U.S. patent
on Tagamet expires. SB said that although its corporate strategy "continues to emphasize maintaining the steady
flow of new SB pharmaceutical products," the company also intends to "minimize the anticipated reduction in our
share of the H2-blocker market." U.S. sales of Tagamet in the first quarter were $155 mil.
The announcement follows April 29 tentative ANDA approvals for DuPont Merck and Novopharm to
produce 200 mg, 300 mg, 400 mg and 800 mg strengths of cimetidine; Mylan was granted the first tentative
ANDA approval for the four strengths of cimetidine on Oct 22, 1993. DuPont Merck will distribute cimetidine
through its recently established Endo Labs subsidiary; Endo plans to make the product available to other
companies, which presumably include Merck-Medco’s West Point Pharma generic drugs subsidiary. Mylan
announced a co-promotion agreement with Lilly in mid-April under which Mylan’s generic cimetidine will be
included in a disease-management program marketed to managed care buyers.
'v
As part of SB’s strategy to protect cimetidine market share, the company filed a second trade patent,
trademark and trade dress lawsuit in Philadelphia federal court May 9. The latest suit charges Novopharm and
Schein Pharmaceuticals with "promoting and soliciting orders-for pharmaceutical (cimetidine) tablets copying
SmithKline’s distinctive trade dress and infringing upon SmithKline’s distinctive and unique light green color
trademark" for Tagamet. A similar suit was filed against Mylan May 6.
The similar color "is calculated to and will mislead the consuming public and trade into believing" that
Novopharm’s products are produced or authorized by SmithKline, the suit maintains. Schein is named as a
defendant based on SB’ s belief that they have an agreement to distribute Novopharm’s cimetidine tablets in the
U.S. "SmithKline learned this past week that defendant Schein will be shipping Novopharm cimetidine on May
18, 1994," the suit notes.
Tagamet’ s light green color is the subject of SB’s new print ad campaign in trade magazines with the
headline: "A 17-year record of quality that will have the competition green with envy.' The ads display enlarged
versions of the four different Tagamet tablets and encourage pharmacists to "look for the distinctive light green
color and Tagamet name on every Tagamet tablet you dispense.'
QUAD GENERIC DRUG TRIAL SHAH GUILTY. BANSAL ACQUITTED May 9 after a four-week trial and five days of
jury deliberation before Baltimore federal court Judge Peter Messitte. Dilip Shah was found guilty of making a
false statement to FDA concerning Quad’ s vancomycin hydrochloride under 18 USC 1001 and 18 USC 2. Bansal
was acquitted of conspiracy to defraud the U.S. government under 18 USC 371. He is the only defendant brought
to trial by the Justice Department in its generic drug investigations to be acquitted.
The jury hung on three additional counts against Shah: conspiracy to defraud the government; making a
false statement concerning ritodrine HC1; and making a false statement concerning complaint files. The judge
declared a mistrial on the three counts. It is understood to be unlikely that Shah will be retried, as acts for which
he was not convicted can nonetheless be considered when his penalty is decided at the sentencing, which is
scheduled for early August.
Former Quad Executive VP-Scientific Affairs Dulal Chatterji, a co-conspirator of Shah who testified
against him at the trial, will be sentenced on May 12 by Judge John Hargrove, who presided over all of the
generic scandal trials previous to that of Bansal and Shah.
Beginning with the legal definition of conspiracy - •"an unlawful agreement between two or more
persons" resulting in at least one overt illegal act — the prosecution presented in its opening argument a selection
of the 32 overt acts which Shah allegedly committed, including "normalization" of data; stating that an unsterile
batch of ketamine was sterile; writing three batch records for one batch of vancomycin, an injectable antibiotic;
© F-D-C Reports, Inc., 1994. Photocopying w ithout permission is strictly prohibited. See Page One.
Multiple copy rate: $495 when mailed in the same envelope with $1,195 subscription.______
139
Wednesday, May 11, 1994
Health News Daily
5
and inventing company records. Bansai, charged only with conspiracy, was described as "loyally and corruptly"
carrying out Shah’s illegal instructions.
Prosecution and defense agreed that Bansai played a smaller role than did Shah in the illegal conduct of
Quad pharmaceuticals. Bansal’ s attorney noted that Bansai was mentioned in only four of the 32 overt acts named
in connection with the conspiracy, and that he was a recently hired Quad employee who did not have the same
long-standing relationships as did many other convicted conspirators.
PEOPLE
Norian Corp.: Former Collagen International and Corporate Controller Marc Faeber joins Norian as chief financial
officer and VP-finance..
PUBLIC HEALTH
AZTxise during pregnancy: Public Health Service establishes a task force to "explore the medical and policy
implications" of a recently announced National Institute of Allergy and Infectious Diseases-sponsored study that found
a 67.5% effectiveness rate with AZT in preventing HTV transmission from pregnant women to their infants. The task
force will "address critical questions regarding treatment, testing, monitoring and resource needs, which are raised
as a result of this study," HHS Assistant Secretary for Health Philip Lee explains May 10. Members are: Chair Lynn
Mofenson, MD, National Institute of Child Health and Human Development; James Balsley, MD/PhD, National
Institute of Allergy and Infectious Diseases; Martha Rogers, MD, Centers for Disease Control and Prevention; and
Helene Gayle, MD, Centers for Disease Control and Prevention..
REGULATORY NEW S
Medicare-Medicaid Coverage Databank: Employers may file information on the health care coverage of workers
using scannable paper forms or preformatted diskettes available from the Health Care Financing Administration, or
in an electronic format submitted on magnetic cartridges, according to a preliminary guidance published by HCFA
in the May 10 Federal Register. Established by the 1993 budget law, the databank will be used to determine whether
Medicare and Medicaid enrollees have additional insurance that must first pay for health care services. According to
the preliminary guidance, the information that employers must fumish includes the name and tax identification number
of each worker, the type of group health plan for each individual, and the coverage period. On May 6, HCFA
Administrator Bruce Vladeck said that the Clinton Administration is seeking an 18-month delay in implementation
of the reporting requirement, which now is scheduled to begin in January 1995. A same-say General Accounting
Office report noted that HCFA will have to collect information on 160 mil. Americans, and concluded that the data
bank will cost more than $100 mil. over five years and produce few benefits..
Greenwich Pharmaceuticals: Company has "received notification from FDA that both an independent evaluation
of the efficacy of Therafectin (amiprilose HC1) and the supervisory review of the Pilot Drug Division’s review of
the Therafectin NDA concluded that the application lacks substantial evidence of Therafectin’ s effectiveness in the
treatment of rheumatoid arthritis," Greenwich reports May 10. The company plans to meet with FDA before the end
of May and expects that it will receive "a notice of an opportunity for a hearing"..
PRODUCT NEWS
Bio-Vascular Firm receives FDA marketing clearance for its Peri-Strips device for reinforcing staple lines during
lung surgery, Bio-Vascular announces May 10. The strips are an adaptation of the Supple Peri-Guard pericardium
product and are intended for use with surgical staplers. Supple Peri-Guard, previously cleared for use as a "pericardial
closure" and a "soft tissue patch,' is made of bovine pericardium cross-linked with glutaraldehyde, the St. Paul,
Minnesota-based firm explains. The new Peri-Strips make the staple site "air tight" and shorten hospital stays, the
company claims. Additional uses for the material are under development; further 510(k) applications are expected by
the end of the year..
© F-D-C Reports, Inc., 1994. Photocopying w ithout permission is strictly prohibited. See Page One.
Multiple copy rate: S495 when mailed in the same envelope w ith S1,195 subscription.
140
•6 •
Health News Daily
Wednesday, May 11, 1994
INDUSTRY NEW S
Cardiotronics: Company proposes the acquisition of all outstanding R2 Medical stock for $4 a share in cash or
Cardiotronics stock, the firm reports May 10. The Carlsbad, California-based rapid cardiac resuscitation and support
device firm’s proposal is being reviewed by R2's board. "We believe that a combination of Cardiotronics’ direct-sales
organization and R2 Medical’s market position would create a strong competitor in the stimulation-electrode market,
and would be in the best interests of the shareholders of each company," states Ronald Bromfield, president and CEO
of Cardiotronics. A Securities and Exchange Commission filing related to the proposal discloses Cardiotronics owns
175,500 R2 shares, or about 6.6% of outstanding common stock...
U.S. Bioscience: NIH is contemplating granting to the West Conshohocken, Pennsylvania-based company an
exclusive worldwide license to a U.S. patent for "Acid Stable Purine Dideoxynucleosides Active Against the
Cytopathic Effects of Human Immunodeficiency Virus" and a U.S. patent for "2 ’ -Fluorofuranosyl Derivatives and
Novel Method of Preparing 2 ’-Fluoropynmidine and 2’ -Fluoropurines" and corresponding foreign patents. The license
may be limited to the treatment for HTV infection using 2-fluoropurine dideoxynucleosides (F-ddA and F-ddI), which
have beeq shown to inhibit HIV reverse transcriptase and cytopathic effects of HIV in vitro, a May 10 Federal
Register notice explains..
Piagnostek: Contract with State of New Jersey will "generate revenues of approximately $75 mil. over the three-year
term," the pharmacy benefit management firm says. The contract calls for Diagnostek’ s HPI Health Care Services
subsidiary to service 19 public hospitals representing about 8,000 patients. "Although the incumbent provider indicated
an intention to protest this award, Diagnostek contemplates it will begin service under this contract this summer or
in the early fall".
Progenics/American Cyanamid: Research agreement will focus on the development of anti-HIV compounds
employing Tarrytown, New York-based Progenies’ Universal Neutralizing Antibody technology and Cyanamid’ s
conjugation technology, the firms announce May 10..
m si
Collagen: Firm receives FDA approval of an Investigational Device Exemption to begin human studies with its
Collagraft bone graft strip for the treatment of scoliosis, President and CEO Howard Palefsky announces at a May
10 session of the Alex. Brown 19th Annual Health Care Seminar in Baltimore, Maryland. The study, which will be
conducted by Collagraft marketing/clinical partner Zimmer, will include 224 subjects at 10 sites. The paste form of
Collagraft was approved in May 1993 for use in long-bone fractures and traumatic osseous defects; the secondgeneration solid strip form received approval in January..
Genelabs: Company announces May 10 that it has begun Phase II/III study of GL701-DHEA (dehydroepiandrosterone) for the treatment of mild-to-moderate systemic lupus erythematosus in women who require steroids for
treatment. GL-701 "may improve the quality of life for lupus patients through reduction of prednisone usage,'
Redwood City, California-based Genelabs maintains...
Chiron: Firm is discontinuing the development of the t-88 monoclonal antibody for the treatment of gram negative
sepsis after the agent "did not demonstrate a reduction in mortality" in an 826-patient Phase III trial. "This was a well
conducted clinical trial by a group of experts in critical care," Chiron explains. "The data are high quality, of which
we are proud. Unfortunately, the results were not positive." Chiron "remains committed to the development of
products that have potential application in critical care therapeutics, which includes the development of products that
have potential application in treating sepsis or septic shock," the company adds. Chiron and Miles are collaborating
on the development of an anti-TNF monoclonal antibody for the indication, which showed a "trend" in favor of drug
in a first Phase III trial..
© F-D-C Reports, Inc., 1994. Photocopying w ithout permission is strictly prohibited. See Page One.
Multiple copy rate: $495 when mailed in the same envelope w ith $1,195 subscription.
141
P A T T O N , B O G G S & B L O W , L . L. P
2 5 5 0
M
W A S H IN G T O N .
S T R E E T ,
D.C.
N.W.
2 0 0 3 7 - 1 3 5 0
( 202) 4 5 7 -6 0 0 0
Facsimile. (202) 457-6315
W RITER S D IRECT DIAL
(202) 457-6424
May 13, 1994
The Honorable Donald Riegle
Chairman, Committee on Banking,
Housing and Urban Affairs
United States Senate
Room SD-105
Dirksen Senate Office Building
Washington, D.C
Re: May 11, 1994 Hearings on Homeowners Insurance
Discrimination______________________________________
Dear Mr. Chairman:
On behalf of American Family Mutual Insurance Company, I would like to address one
aspect of the testimony presented on May 11, 1994, by Assistant Attorney General Patrick That
testimony stated that American Family had not cooperated in an investigation initiated by the
Civil Rights Division in 1986. American Family respectfully disagrees and would like to correct,
the record.
First, when the Civil Rights Division requested materials from the company in 1986, the
controlling law was the Fourth Circuit's decision in Mackev v. Nationwide Insurance Companies.
724 F.2d 419 (1984), that the Fair Housing Act did not cover home insurance. Rather than
undergo an extensive intrusion into its business operations, American Family respectfully
advised the Division that it did not have jurisdiction. The company wants to urge the Committee
not to confuse the legal position it took , based on that opinion, with rejection or resistance to the
principles underlying the Fair Housing Act.
This issue arose again in NAACPv. American Family Mutual Insurance Company. The
United States District Court for the Eastern District of Wisconsin agreed with the company’s
view that the Act did not reach insurance companies. As a consequence, the Division did not
pursue its negotiations with the company to obtain the information it had requested. Instead, it
presented the Government's views before the Court of Appeals for the Seventh Circuit. On
October 20, 1992. the Court of Appeals reversed the decision by the Eastern District and held
that, as a result of HUD's promulgation of regulations in 1989, the Act does apply to home
insurance. Within a month of that decision, the company offered to voluntarily provide the
142
The Honorable Donald Riegle
May 13, 1994
Page 2
information the Department substantial volumes of data and records and made its employees
available for interviews by Division investigators. It did so even though it sought further judicial
review of the matter. Mr. Shriner's letter of November 12, 1992 to the Assistant Attorney
General (attached) documents the company’s offer.
I would appreciate your making this letter a part of the hearing record.
Sincerely,
John L. Oberdörfer
JLO/mm
cc:
Paul Hancock, Esq.
143
F
o l e y
&
l a r d n e r
riO S T WISCONSIN CCNTCB
tTtCAflTWISCONSINAVCNUC
w ii.w A U K lt . WISCONSIN 9 3 2 0 2 -6 3 «?
rCLC PH O Ne « ♦ ' * ' * 7 i.* 4 o o
MAO'OON, W ISC ON SIN
Crt.CAGO. H.U.NOIS
rcucx im
iFOLtrlaAO MiU
raCSIMILC(*u)ta»%9?»i
* ahminotON, DC.
ALkA A ^O KIA. VlOGINlA
AiMNAPOUli*. MAHVLANO
.. A C XSON VILLC , WL C *IO A
(ju tA N O O , rv.O «iO A
*A (.'.A M A S i£ C . ruUlxiOA
LONDON. CNG U M O
*»A »I* . rR A N C e
8CWLIN. OC^MANY
5TUTTGABT. OCPMANY
OWCSOCN.OCRHANV
wttiTCR'3 Oi»CCT UNC
a. ruonriOA
a»N0A#»0<*c
414-289-3549
WCST PALM OCACM, ►LOHIOA
TAIPCI. TAIWAN
November 12, 1992
BY Fedixaz. Express
John R* Dunne, Esq.
Assistant Attorney General
Civil Rights Division
United States Department of Justice
Washington, D.c.
20530
Re:
American .Family Mutual Insurance Company
Dear Mr. Dunne:
We are counsel to American Family in the NAACP case now
pending in the United states District Court for the Eastern
District of Wisconsin and the United State* Court of Appeals for
the Seventh Circuit, to which you adverted in your letter of
November 3, 1992 to Mr. Dale Mathwich, chairman and CEO of
American Family. Accordingly, Mr. Mathwich has asked that I
respond to that letter on American Family's behalf and represent
the company with respect to the Department’s investigation.
The short answer to the question asked in your letter
is that American Family is willing voluntarily to provide the
Information which the Department needs to complete its
investigation.
Because the Court of Appeals has determined that
the Fair Housing Act applies to the business of homeowners'
insurance (although American Family may seek further review of
the question), the company considers that the previous ground for
its declining to provide information (which, of course, went to
the Department's jurisdiction to investigate at all) is no longer
extant,
Since the last contact between American Family and the
Department, nearly four years ago, the NAACP action has been
filed and there has been considerable discovery and other trial
preparation.
Thus, we are in a position to discuss specifically
which information is available, in what form, and with what
NOU 1 3
’92
0 0 :5 2
4142893791
PAGE. 0 0 2
144
John R. Dunna, Esq.
November 12, 1992
Page 2
degree of expense and difficulty in retrieval.
I had hoped to
discuss these matters with Mr. Conrad before writing to you, but
I understand that he will ba out of his office until next
Tuesday.
I will call him then to discuss these details of
production.
American Family notes with pleasure and relies upon
your statement that the Department has not mads a final decision
on the merits of this matter and that your purpose is to avoid
litigation if it is not meritorious.
We believe that we can
persuade the Department, based upon what ve have developed in
defending the NAACP case, that there is in fact no basis for
Justice Department involvement.
X looJc forward to discussing
these matters further with Mr. Conrad next week.
Very truly yours
Thomas L. Shriner, Jr
145
HOMEOWNERS’ INSURANCE:
AN INVESTIGATION INTO POSSIBLE ILLEGAL DISCRIMINATION
Co n d u c t e d
by th e
M in n e s o t a D e p a r t m e n t
M arch
op
Co m m e r c e
8,1994
Executive Summary
The Department of Commerce is the State agency responsible for regulating in
surance industiy practices within Minnesota. The Department initiated an inves
tigation into potential “redlining” problems in the Twin Cities Metropolitan Area fol
lowing a nationally released report produced by the Association of Community Orga
nizations for Reform Now (ACORN). The report identified Minneapolis/St. Paul as
one of fourteen (14) metro areas nationwide where “redlining” was occurring.
The term “redlining” is frequently used to describe a practice whereby insurance
companies refuse to offer or cancel homeowners’ insurance solely because of the lo
cation of a home within a city. Minnesota Statutes do not use nor define redlining.
However, Minnesota Statute {72A.20 subdivision 13 (1992) does provide protection
for homeowners from such discriminatory practices. Specifically, the statute pro
hibits an insurer from refusing to offer or cancel coverage based on: (1) The location
of a home within a city, (2) the age of the home, (3) a prior declination by another
insurer, or (4) because the home was previously insured under the FAIR Plan.
Minnesota Statutes require insurers who offer to provide coverage on homes with
in any area within a city to offer the same policies/programs to homes in all areas
within the same city. However, insurers are entitled to charge rates which reflect
the age of mechanical systems within the dwellings and other factors such as prox
imity to fire protection etc.
In February 1993, the Department began accumulating information from a vari
ety of sources in an attempt to determine if redlining was occurring within the Twin
Cities Metropolitan Area. During the course of the investigation, the Department re
viewed the records and data provided by the (23) companies having at least a 1 per
cent market share of the Minnesota homeowners’ insurance market. To determine
whether any patterns of abuse or violations had occurred, Department investigators
reviewed thousands of pages of underwriting guidelines and over 4,000 cancellation,
declination, and non-renewal notices to homeowners.
Our investigation found that 97 percent (3,917) of the cancellations, declinations,
and non-renewals reviewed were m compliance with the applicable statutes and
rules. In addition, of the 3 percent (125) cancellation, declinations, and non-renewals
identified as not complying with the statute only twelve involved activities that
might fall within the commonly used definition of redlining. Based on our findings,
we have concluded that although isolated violations of statute may have occurred,
the available evidence does not substantiate the allegation that insurers are engaging
in any pattern or general practice o f redlining within the Twin City Metropolitan
Area.
Statutory/Regulatory History
Minnesota Statute §72A.20 subd. 13 (1992) prohibits insurers who write home
owners’ coverage within a city from refusing to offer, write, or renew a policy or
charge differential rates for homeowners’ insurance solely because of the location of
a home in that city; the age of the home; declination by another insurer, or because
of prior coverage issued under the FAIR (Minnesota Property Placement Facility)
plan.
The law allows companies to underwrite and set premium rates to reflect extraor
dinary hazards, availability of lire protection, and concentration of the insurer's
risks. The law also allows for rating, but not declining a homeowner policy based
upon the age of the electrical, plumbing, heating/cooling system, wiring, or other
structural items affected by age.
The Commerce Department enforces the statutory protections of the law and
places special attention on homeowner policy cancellation, nonrenewal, and declina
tion actions of insurance companies. Department records show that there were 124
cancellation/nonrenewal investigations conducted in 1991. As a result of these inves
tigations, seven (7) company actions were rescinded by the insurer or reversed by
the Department, and five (5) administrative actions (i.e. Consent Orders) were
taken. In 1992, 166 cancellation/nonrenewal investigations were conducted, fourteen
(14) voluntary reinstatements or Department reversals occurred, and four (4) admin
istrative actions taken. 1993 statistics available for this study indicate that the De
partment has conducted 100 such investigations, leading to five (5) voluntary rein
statements or Department reversals, and one (1) administrative action.
146
Most, if not all, of these corrective actions were based on deficiencies in form, con
tent, or notice requirements. Neither Department complaint records nor investiga
tion results indicate a pattern of illegal discrimination by insurance companies in
the area.
Investigation Chronology
In February, 1993, the Association of Community Organizations for Reform Now
(ACORN) provided the Department with a copy of its report stating that “redlining”
was occurring in 14 cites throughout the Nation including the Minneapolis/St. Paul
Metropolitan Area.
The report said that companies were refusing to write homeowners’ coverage in
certain areas of the Twin Cities. The ACORN report specifically cited: (1) more fre
quently required home inspections in the inner-dties than in ¿he suburbs; (2) a
higher percentage of uninsured homes in the inner-cities; (3) higher costs of cov; and (4) fewer agents in the inner-cities as evidence that
s based primarily on the review of zip code reports which
list insurance company cancellation, decimation, and nonrenewal actions and, test
calling conducted by members of ACORN. ACORN’S findings were based upon 48
calls made in each city to different insurance agents. Because of the limited scope
of ACORN’S survey, one call could effect a difference in survey findings of between
4-9 percent.
In February 1993, then-Commissioner Bert McKasy wrote to ACORN to inform
them that our Department would conduct an investigation, per their request, into
the sale of homeowners’ coverage by a number of the major insurance carriers. Our
investigation attempted to verily the accuracy of ACORN’S assertions and to deter
mine whether such factors, in and of themselves, evidence the existence of redlining
in the Twin Cities.
The Department obtained a computer printout from the National Association of
Insurance Commissioners (NAIC) which identified the maiket share of business for
homeowner insurers in Minnesota. In order to determine insurance company compli
ance with Minnesota statute, an extraordinarily large group of companies was in
cluded in our investigation. The Department of Commerce investigated companies
with a market share of 1 percent or greater (23 companies, with a total market
share o f 77.08 percent per the 1991 annual statements; see attached listing). Of this
group, (6) companies were affiliated with at least one other company in the sample
group (standard/preferred), and share joint underwriting, claims, and policy issu
ance departments.
In addition to the zip codes used by ACORN, we added the communities of West
St. Paul and South St. Paul. These communities were added because of the potential
for having a higher than average minority population. It was also determined that
the investigation would thoroughly review each company’s: 1) underwriting stand
ards, directives, manuals, and instructions; 2) agent locations; and 3) declinations,
cancellations, and nonrenewals of homeowners’ insurance for 1991 and 1992. During
the week of April 15, orders requiring the production of extensive amounts of docu
mentation were sent to each of the 23 companies. Most insurers needed up to 60
days to assemble the information and documentation required.
A review of the thousands of pages of underwriting material was completed by
the week of August 9. Upon completion of the review, companies were sent a letter
advising them of the provisions of their underwriting standards which were of con
cern to the Department. After these notifications were sent, a total sampling of
4,042 declinations, cancellations, and nonrenewals was drawn from the policyholder/
applicant lists provided by each insurer. During the week of August 20, each insurer
was sent a letter recjuinng the production of the corresponding notices that were
sent to the insureds indicated in our sample. The responses were received, in their
entirety, by early November, 1993. A thorough review of each notice was conducted
to assure compliance with regulations and statutes concerning the form, method,
and basis for tne company’s action.
This review was completed and companies were notified of the results the week
of November 18. Each company was given the opportunity to review actions which
appeared to be in violation o f Minnesota statute. The companies were also asked
to provide the Department with any additional information, documentation, or re
buttal. This information was received, in its entirety, by December 16, 1993.
In December, the Department determined that a cost analysis should be done by
comparing each company’s rates between the inner-cities and suburbs, and between
different geographic locations within both Minneapolis and St. Paul.
On December 29, 1993, rate surveys were sent to each of the 23 companies and
the Minnesota Property Placement Facility (FAIR Plan). The rate surveys required
147
companies to provide premium quotes for all available company programs/policies
for a HO-3 (homeowner) policy (or their closest equivalent) on a 1940 frame con
struction home with a replacement value of $75,000. The age of the roof was estab
lished at 10 years old, and the home had a one car detached garage. The value of
the personal property was established at $35,000. Companies were also required to
provide rate quotes for an HO-4 (tenants) policy, with personal property valued at
$35,000. Each company was provided with a list of zip codes for the cities of St.
Paul, Minneapolis and surrounding suburbs.
This expanded investigation required Enforcement Division employees, who live
in the inner-city, to call agents and obtain quotes from two different agents of each
of the six (6) companies with the largest market share. In total, the employees made
contact with thirty-four agents. Two of the employees represented minority cultures,
and all used actual homes for purposes of obtaining quotes and inspections (if re
quired).
Investigative Findings
U n d e r w r it in g (C h a r t # 1 )
Based upon the review of the underwriting material, there appeared to be no
basis to conclude that any of the twenty-three (23) insurance companies were inten
tionally targeting inner-city homeowner applicants or policyholders in an attempt to
illegally discriminate in the issuance of homeowner coverage.
Actuarial examinations conducted by this Department in coqjunction with the in
vestigation strongly suggest that insurance carriers’ loss experience closely cor
relates with premium rates. We examined two carriers with a combined market
share of 33.6 percent. One carrier was found to actually be charging less in the
inner-cities than their loss experience would dictate. The other carrier was found
to closely align its premium rates with loss experience.
While the actuarial data did show there to he higher costs for inner-city coverage
than for suburban coverage, the costs were reflected in the loss experience of the
insurance companies. On average, the difference was $42.43 per year. The investiga
tion concluded that homeowners insurance is available to inner-city homeowners
and that loss experience on inner-city policies is a reason for the different premium
prices. Of particular note is that some rural areas actually have higher premium
expenses than either the inner-cities or the suburbs. The primary reason for the
higher premium rates is lack of fire protection, water sources, etc.
The review of the underwriting standards and guidelines of each company found
that eighteen (18) insurers had some portion of their underwriting guidelines which
needed to be changed to conform with all current statutory requirements. The De>artment did identify a few instances where company guidelines provided for decinations based solely on the age of the applicant’s dwelling. Fourteen (14) of the
companies provide rating discounts based upon the age of the home (mostly in the
form of a ‘rlew Home” discount). The Department of Commerce only allows such
discounts if the company also offers comparable coverage and discounts for older
homes which have updated electrical, plumbing, heating/cooling systems, and wir
ing. Our investigation revealed that the vast majority of the companies do offer com
parable premium discounts and/or coverage options for older homes. Those that did
not were notified of the need to amend their underwriting policies.
It should be pointed out that prior to June 1992, one company did have under
writing standards which specifically mentioned the geographic location of the home
(e.g. “St. Paul-Minneapolis Area”). However, this criteria was eliminated by the com
pany prior to our investigation.
i
C a n c ell a t io n /N o n r en ew a l /D e c l in a t io n (C h a r t # 2 )
A review of each of the 4,042 declination cancellation, and nonrenewal actions
shows that 125 actions were found to be in violation of the statutes/rules. Most of
the violations were a result of deficiencies in the notice form and length of prior no
tification (107 actions). These violations occurred because of the failure by some of
the insurers to include specific information in the notice (e.g. right to appeal, right
to have a new agent assigned, specific loss information) or in the time notification
requirements (e.g. exceeding the 59 day underwriting standard, less than 60 days
notice, *etc.)
Twelve of the 125 violations identified appear to relate to potential redlining. In
these cases, the declination appeared to be based solely on the age of the home. The
remaining six actions were based on reasons other than those allowed by statute
or rule (e.g. requested by agent without documentation as to basis, requested by
mortgage company, etc.). Based upon this information and documentation the inves
tigation concluded that 99.7 percent of the 4,042 met the requirements o f Minnesota's
“redlining" statute.
148
S um m ary
of
A g e n t L o c a t io n s (C h a r t # 3 )
Although there are no regulatory or statutory requirements that establish a mini
mum agent-to-policyholder ratio, the office locations for each company^ agents were
requested and charted. The Department does not believe the charting should be in
terpreted as an accurate depiction of a company's agent representation in any spe
cific geographic location because these agencies could have more than one agent at
each location/agency.
In a society that is effectively linked by telecommunications, companies can con
duct most of their business through direct mail solicitation and provide application
processing, underwriting, claims, and customer service by phone. Also, independent
agents can legally broker business for other non-appointed agents. This would allow
a company to market in a geographic location without officially establishing an
agent/agency presence.
While quantifiable results were not reached, the Department does not believe that
agency location significantly inhibited access to insurance. The Department does rec
ognize, however, that agency location is an important convenience issue.
S um m ary
op
R a t e S u r v e y (C h a r t # 4 )
For purposes of the rating survey, South St. Paul and West St. Paul were consid
ered suburban quotes. This did not adversely affect the comparison, since these com
munities were usually <moted at a lower rate than St. Paul and Minneapolis. The
rate surveys were compiled in a chart which allows the comparison of each compa
ny’s rate within the city limits of St. Paul and Minneapolis as well as urban to sub
urban comparisons. The chart does not provide company-to-company rate compari
sons. Each column represents the dollar amount difference between the highest and
the lowest rate quoted in that category for urban vs. suburban locations.
The maximum annual difference of any company rates between equivalent subur
ban and urban risk was $95.38 for the HO-3 (homeowners) policies and $26.50 for
the HO-4 (renters) policies. The average difference between the suburban and
urban rates for the 23 companies and the FAIR Plan was $42.43 for the HO-3 risks,
and $7.49 for the HO-4 risks. There were three (3) companies that had no difference
between suburban and urban rates for their HO-3 policies, and seven (7) companies
that had no rate difference for their HO-4 policies. Une company quoted some uiban
rates lower than some of their suburban rates for their HO-4 policies.
Seven (7) companies appear to have charged different rates within the city limits
of Minneapolis/St. Paul lor policies on identical homes, a practice in violation of the
current statute. The differences ranged from $7-$73 and the Department is requir
ing these companies to take corrective action. These were isolated instances ana did
not appear to show a pattern of intentional wrongdoing.
Q u o t e s O b t a in e d F r o m A g e n t s
There were 34 actual calls completed by Department personnel. The callers found
only one instance where the agent seemed reluctant to provide a quote (but it was
provided) and one instance where the agent did not return the phone call (contact
was not made with the agent personally, only a name and phone number was left
with his office). In all other instances the agents were described by the callers as
polite, willing, and sometimes eager to provide quotes. Very few agents advised that
an inspection would be necessaiy prior to writing the coverage, and all provided
written confirmation of their quotes to the callers when requested. The callers have
also received several follow-up phone calls from the quoting agents anxious to write
the insurance coverage.
Conclusions and Corrective Actions Taken
Although patterns of illegal discrimination were not substantiated, the Depart
ment of Commerce would like to thank ACORN for requesting this investigation.
Several corrective actions resulted from the investigation.
The Department strongly supports inner-city home ownership as one wav to sta
bilize neighborhoods ana available homeowners’ insurance is critical to this goal.
Continuing oversight and the monitoring of insurance industry practices by the De
partment of Commerce combined with an on-going and constructive dialog between
insurers, agents, community organizations, and inner-city homeowners is the best
method of insuring compliance with Minnesota’s statutes.
Conclusions
U n d e r w r it in g
Underwriting policies regarding new home construction discounts were the pri
mary area in need of updating to conform with the Minnesota statute prohibiting
premium discounts based exclusively on the age of a dwelling. However, because in
149
surance companies can adjust rates based on the level of risk (age of mechanical
systems within a dwelling) the Department does not believe that the updating will
significantly change the affordability of homeowners’ insurance.
C a n c e l la t i o n / N o n r e n e w a l/ D e c li n a t i o n
Ninety-seven percent of the 4,042 actions reviewed by the Department were found
to be in compliance with Minn. Stat. 72A.20 subd. 13. Most of the violations in the
3 percent were technical in nature. The Department considers this low frequency
of violations a sign that illegal cancellation, nonrenewal, or declination is not preva
lent in the Twin Cities.
A g e n t L o c a tio n s
There is no regulatory or statutory requirement for locations of insurance agent
offices. Although there are fewer agents located in the dty than in the suburbs, this
does not constitute illegal discrimination. With insurance available by both mail and
phone, agent location aoes not seem to limit access to coverage.
Ora l
and
W r it t e n I n s t r u c t io n s
to
Ag e n t s
to
A v o id I n n e r C it y A r e a s :
Three agents recommended by ACORN were interviewed in the belief that they
knew of illegal discrimination practices on the part of insurance companies. No evi
dence of illegal discrimination was uncovered as a result of these interviews.
R ate S urvey
The Department found that some companies have higher insurance premium
rates in the inner-cities than in the suburbs. The average difference was $42/year
for homeowners and $7 for renters. The urban-suburban differences were relatively
small and related to the loss experience of the insurance company as confirmed by
an actuarial study performed in correlation with the investigation.
Hie investigation identified companies which charged different rates within dif
ferent areas of the city. This is not allowed and corrective action is being taken.
Qu o t e s
fr o m
Ag e n t s
The Department conducted tests to see if agents were willing to sell in the innercity and concluded that agents were reasonably eager to sell and quote insurance
to inner-city homeowners.
Corrective Action
Departmental action which has been taken as a resolution to the several findings
of this investigation:
• The Department is issuing a bulletin to all insurers authorized to write home
owners’ insurance^which details the requirements and prohibitions of Minn. Stat.
72A.20 subd. 13. This bulletin specifically addresses the prohibition of discounts
based on the age of the home; underwriting eligibility restrictions based on the
age of the home; and the proper form ana filing requirements of cancellation/
nonrenewal notices.
• The Department required each of the companies found to have underwriting and
rating standards in violation of State statute, to immediately correct these stand
ards, materials, guidelines, and instructions, and provide confirmation and evi
dence of these corrections.
• The Department required each company, where applicable, to file corrected copies
of their cancellation, nonrenewal, and declination forms.
Market Share (91)
UNDERWRITING
Restricted Binding
or Decline Due to
Age of Structure *
Rating based on Age
of Structure +
Allstate
5.9%
X
A M CO
1.47%
X
American Family
13.22%
X
1 Austin Mutual
1.57%
X
X
HOwners Insurance Company
1.12%
1.64%
X
X
X
X
I Auto Owners
Declined Based on
Location
Citizens Insurance Company
1.20%
X
Farmers Home Mutual
Gopher State Mutual
1.05%
1.09%
N /A
n/ a
N/A
N/A
N/A
N/A
Horace Mann
1.07%
N /A
N/A
N/A
Illinois Farmers
9.24%
X
X
Milbank Insurance Company
1.63%
X
X
Minnesota Mutual
1.27%
X
Mutual Service Insurance Company
1.74%
X
X
Northstar Mutual
2.68%
X
X
1 Pacific Indemnity
1.54%
1Prudential
1.35%
X
X
•
+
1
Property & Casualty
Dcclinntion may only pertain to preferred programs or policies.
Most companies have filed a *new home" discount.
Underwriting rule change 6/92.
X
150
CHART #1
I Company
CHART #1 - UNDERWRITING (CONTINUED)
| Company
Market Share (91).
Restricted Binding
or Decline Due to
Age of Structure *
Rating based on Age
of Structure +
| St. Paul Guardian
2.97%
RState Farm Fire & Casualty
| State Farm General Insurance
20.16%
1.22%
X
X
X
X
| USAA
1.28%
X
X
I Waseca Mutual
1.21%
X
X
1.46
X
] Western National
I
*
+
1
Declination may only pertain to preferred programs or policies.
Most companies have filed a "new home" discount.
Underwriting rule change 6/92.
Declined Based on
Location
X
X1
CHART #2 - CANCELLATIONS/NONRENEWALS
Company
Number of Improper
Cancels/NonRenewals (#/samp!e
size/percentage)
Notice/Form/
Notification
Age of Dwelling
(Declined/Canceled)
2
Allstate
10/217/4.6%
8
AMCO
4/113/3.5%
4
American Family
Austin Mutual
Auto Owners/Owners
Citizens Insurance
6/279/2.1%
4
0/54/0%
N/A
4/46/8.6%
3
Farmers Home Mutual
29/714/4.06%
29
25/330/7.57%
25
1/15/6.6%
1
Illinois Farmers
7/201/3.48%
7
Milbank Insurance Company
6/172/3.48%
6
Minnesota Mutual
Mutual Service Insurance
Northstar Mutual
1/86/1.1%
2
0/153/0%
N /A
l/120/.8%
1
Prudential
2/76/2.6%
1
St. Paul Guardian
14/275/5.09%
11
State Farm Insurance
1/329/0.03%
1
Waseca Mutual
Western National
Other
H
2
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1
1
2/253/0.7%
Pacific Indemnity
IJSAA
Property
Location
(Declined/
Canceled)
4
4/434/9%
Gopher State Mutual
Horace Mann
Improper Basis
for Cancellation/
Nonrenewal
N /A
N /A
1
3
1/1/100%
1
0/5/0%
N /A
N/A
7/169/4.14%
3
4
N/A
J
153
CHART # 3
NUMBER OF AGENTS
(D oesNoe Indicate The Number O f Office Locations)
Co m p a n y
St. Paul,
S « St. Paul,
W. St.Paul
Allstate Ins. Co............
.4...
Surrounding
7 Countv Area
..8 ............ - . 1 2 1
American Family Mut........ ..17 „ „ „
. 1 . 8 . . . .................... - .252
Auto Owners/Owners. ,.... .... .71...
, , 7 . , . . ................ . . . . 3 1 4
AMCO • <1«» <i« o h o h mi « « « • ilo ilil..io » •9 • « •■
<
0 . 4 . .............................. 9 9
Austin M u t . ...... ............ . 8 ...
. . . , 4 . 1 . . ......................... 5 7
Citizens Ins. Co« ..««««. o.•<, •12 •• hi
. . 3 ...................i, . . . . „ 6 6
Farmers Home Mut......... ... 11...
I I . 6 . . . . . I I . . I . . . I . i . . 89
Gopher State -Mut....... ..... »7 ....
. . 3 ............................ . . 8 7
Horace Mans. ......«••.«• « «>. » <►•4 ....
. .0 .............. . .21
Illinois Farmers............. 34..,,.
. 1 0 ................................352
Milbanfc Ins. Co ............„..2....
. . 0 ............................. . . 2 4
Minnesota Mut ............... 1 4 1 ...,.
. 3 2 .......................... .. . 5 2 8
Mutual Service Ins..... ...... 5....
•.. 0 ........................ ....... . 5 9
North star Mut....... ........ 4....
. . 0 ............................. . . 5 3
Pacific Indemnity............. 5....
. . 5 .................................. 31
Prudential Prop. 6 Cas........0 ....
•.. 0 ................................260
state ?arm (fire & General). .26. ....
. 1 9 ................................182
St. Paul Guardian............ 22....
.22 .................................. 122
ÜS A A ...................... ... .O.....
. .0 ........ ...... .
Waseca Mut.............. ...... 3......
. . 0 .................................. 55
Western National......... .... 12....
. . 6 ................ . . , , . , 9 4
Totals:
Percentages
397
13.9%
137
4.8%
.0
2/840
81.8%
CHART # 4 -
Company
HO-3
St. Paul
HO -4
S t Paul
HO-3
Mpls.
HO-4
Mpis.
rate su rvey
HO-3
St. Paul
v. Suburbs
HO-4
St. Paul
v. Suburbs
HO -3
Mpls. v.
Suburbs
H O -4
Mpls. v.
Suburbs
Max Diff. H O O
Suburbs v. Cities
Max Diff. HO -4
Suburbs, v. Cities
1 Allstate Insurance Company
S61
S12
S61
$12
S77
$12
$61
$12
$77
A M C O Insurance Company
SO
SO
SO
so
S57
$8
$59
$8
$59
S8
Am erican Family Mutual Insurance Com pany
SO
SO
SO
SO
S54
$14
$0
$0
$54
S14
Austin Mutual Insurance Company # 1
SO
$0
SO
so
S40.91
■$o
$40.91
*$0
$40.91
•so
Austin Mutual Insurance Company # 2
SO
.
SO
S39.91
.
S39.91
-
$39.91
Auto-Ow ners Insurance Company
SO
SO
SO
so
SO
so
SO
$0
$0
SO
Citizens Security Mutual Ins. Co. # 1
SO
so
SO
so
S87.16
S2638
S70.16
$23.88
$87.16
S26.38
Citizens Security Mutual Ins. Co. # 2
SO
Farmers H om e Mutual Insurance Company
SO
G opher State Mutual Insurance Company
SO
H orace Mann Insurance Company
S57
SO
Illinois Farmers Insurance Company
SO
so
$9538
$7638
so
so
S22
so
-
$23
S10
S57
$10
S57
S10
so
so
so
S59
S10
S10
$12
$9538
$22
$10
$22
$23
-
$23
$57
$10
$57
$10
$59
$10
$59
$10
$10
so
so
so
so
S55
S9
$55
$9
$55
$9
Minnesota Mutual Fire & Casualty Co. # 1
$72
S7
so
so
$72
S7
$72
S7
$72
$7
Minnesota Mutual Fire & Casualty Co. # 2
S71
-
so
Mutual Service Casualty Ins. Co. # 1
S73
S10
so
so
S73
Mutual Service Casualty Ins. Co. # 2
S56
so
-
S56
so
so
S24
Milbank Insurance Company
North Star Mutual Insurance Company #1
SO
North Star Mutual Insurance Company # 2
SO
so
so
$71
$71
S71
S10
$73
S10
S5
S16
Ncte: For purposes o f pricing So. St. Paul and W est St. Paul were considered suburbs.
1 Indicates some suburban rates higher than St. P a u l/M p ls . rates.
T h e F air Plan does not issue H O -3 coverage (H O -8 ); H O -3 and H O -4 are not replacem ent policies ( A C V )
S24
S16
$73
$10
$56
$56
S5
S24
$16
S5
CHART # 4
Company
HO-3
St. Paul
HO-4
St. Paul
HO-3
Mpls.
RATE SURVEY
HO-4
Mpls.
HO-3
St. Paul
v. Suburbs
Continued
HO-4
St. Paul
v. Suburbs
HO-4
HO-3
Mpls. v. - Mpls. v.
Suburbs
Suburbs
Max Diff. HO-3
Suburbs v. Cities
Max Diff. HO-4
Suburbs, v. Cities
Owners Insurance Com pany
SO
SO
SO
SO
so
SO
SO
SO
SO
Pacific Indemnity Com pany
$0
$0
SO
SO
$42
SO
$42
SO
$42
SO
Prudential Property & Casualty Ins.
SO
SO
SO
SO
S57
. S12
S57
S12
$57
$12
St. Paul Guardian Insurance Com pany # 1
SO
SO
S45
SO
SO
SO
S7
SO
S7
S7
St. Paul Guardian Insurance Com pany # 2
SO
SO
SO
SO
so
SO
SO
SO
SO
SO
| State Farm Fire and Casualty Com pany
SO
SO
$60
SO
S60
SS
S60
S8
S60
S8
| State Farm General Insurance Com pany
SO
S64
-
S64
1 United Services A utom obile Assn.
SO
SO
SO
SO
S50.81
SO
SO
SO
$50.81
| Waseca Mutual Insurance Com pany
SO
SO
SO
SO
S78
S4
SO
so
S78
S4
| Western National Mutual Insurance Company
SO
SO
SO
S 17.87
$10.76
S 17.87
S10.76
$17.87
$10.76
| Minnesota Property Placem ent Facility1
SO
SO
SO
S3830
- $26.50
$35.28
S26.50
$3830
$26.50
S64
S64
so
so
Noie: For purposes of p r ic in g So. St. Paul ;uul West St. Paul were considered suburbs.
’ Indicates some suburban rates higher than St. I’a u l/M p ls . rales.
Th e Fair IMan does not issue I IO-.» coverage (1 1 0 *8 ); 110-3 and 110 -4 are not replacem ent policies (A C V ).
SO
SO
156
RESPONSE TO WRITTEN QUESTIONS OP SENATOR BOND
FROM ROBERTA ACHTENBERG
Q .l. Condition Federal Assistance on a Housing D iscrim ina
tion Plan. Ms. Achtenberg, I understand that HUD is developing
a regulation which would provide for the Department to withhold
Federal assistance, including public housing, CDBG and HOME as
sistance, if a jurisdiction fails to develop a HUD-approved plan to
combat nousing discrimination. This sounds like another way for
HUD to micromanage the decisionmaking of jurisdictions. How do
you envision implementation of this rule? Won’t the withholding of
HUD funding, such as CDBG funding, primarily adversely impact
the low-income families that need the benefit of the funding most?
A.1. HUD is developing a regulation to implement the requirement
which Congress placed in Title I in 1983 requiring every recipient
of CDBG assistance to certify that it will affirmatively further fair
housing. Similar statutory requirements appear in the HOME pro
gram and the legislation creating the CHAS. The certification is
not limited to these programs and includes both publicly assisted
and private housing within a jurisdiction.
The proposed rule will require each Entitlement community (and
State) to develop an analysis of impediments to fair housing and
develop a plan to address these impediments. Communities which
have not previously developed an analysis (more than 100 commu
nities have done so since the concept was introduced as a “safe har
bor” in HUD’s 1989 CDBG regulations) would have 1 year from the
date of HUD’s final regulation to do so. The fair housing plan
(analysis plus action plan) would NOT be submitted to HUD for ad
vance approval. Instead, a summary of the plan would be submit
ted, together with the fair housing actions taken the preceding year
and those planned for the forthcoming year.
HUD would raise questions about an applicant’s certification
only where there was evidence of a problem which the applicant
failed to address. We do not believe that this would constitute
“micromanaging the decisionmaking of jurisdictions.” In fact, it is
the present system that often requires us to second guess a juris
diction’s actions under its certification because no document exists
summarizing the city’s fair housing “temperature.”
We would expect that any differences of opinion about an appli
cant’s identification of impediments or actions to address them
could be readily resolved through negotiation. If not, HUD could re
quire special assurances or condition the grant on the community’s
taking certain actions by a prescribed time. HUD would seek to im
pose sanctions only as a last resort, recognizing that any interrup
tion of the flow of funds would be harmful to the low- and moderate-income residents whom the programs are designed to benefit.
This is precisely the same enforcement mechanism which HUD
currently has with respect to CDBG and HOME funds. The pro
posed fair housing planning regulation does not change enforce
ment procedures in any way.
Q.2. Property Insurance Redlining. Ms. Achtenberg, I under
stand that President Clinton signed recently an Executive Order
directing HUD to issue guidelines on property insurance discrimi
nation. I am concerned that HUD intends to regulate the insurance
157
industry under the Fair Housing Act without any legal authority.
How does HUD intend to implement this order?
A.2. Several administrations, beginning with a HUD General
Counsel opinion in 1978, have concluded that Title VIII of the Civil
Rights Act of 1968, as amended (the Federal Fair Housing Act),
prohibits discrimination in the área of property or hazard insur
ance. Because HUD is the primary Title VIII law enforcement
agency, and the only agency with authority to promulgate regula
tions under that Act, the Department has the responsibility to
issue rules applying the Act to property insurance.
In regulations implementing the Fair Housing Amendments Act
of 1988, HUD determined that the Act prohibits “refusing to pro
vide
property or hazard insurance .
or providing such .
insurance differently because of race, color, religion, sex, handicap,
familial status, or national origin (24 C.F.R. Section 100.70(a)(4)).
Courts that have considered the issue have concluded that insur
ance is covered by the Act. Dunn v. Midwestern Indemnity Mid
American Fire & Casualty Co., 472 F. Supp. 1106 (S.D. Ohio 1979),
McDiarmid v. Economy Fire & Casualty Co., 604 F. Supp. 105
(S.D. Ohio 1984), NAACP v. American Family Mutual Insurance
Co., 978 F. 2d 287 (7th Cir. 1992), cert, denied, 113 S. Ct. 2335
(1993), Nationwide Mutual Insurance Co. v. Cisneros, No. C3-9252 (S.D. Ohio Feb. 24, 1994). But see Mackey v. Nationwide Insur
ance Co., 724 F. 2d 419 (4th Cir. 1984).
The key sections of the Fair Housing Act are 3604(a) which
makes it unlawful to “otherwise make unavailable or deny, a dwell
ing” and 3604(b) prohibiting discrimination “in the provision of
services or facilities in connection therewith.” Because property in
surance is required to secure a mortgage loan, which generally is
required to purchase a home, denying insurance makes that home
unavailable. As the Court explained in the American Family case,
“no insurance, no loan; no loan, no house.” When the Supreme
Court was presented with the opportunity to modify this ruling it
declined to do so.
It is important to understand that HUD has no intention of regu
lating the insurance industry. The responsibility of the Department
is to interpret and enforce the Fair Housing Act as it applies to in
surance.
. In order to clarify the application of the Act to insurance, HUD
has begun a collaborative process that will involve close consulta
tion with the insurance industry, regulators, civil rights organiza
tions, and other community groups.
An advanced notice of proposed rulemaking will be issued to so
licit written comments from all interested parties. Several informal
meetings will be held with representatives of insurance companies,
agents, State insurance commissioners, and various community ad
vocates. At least four public hearings will be held around the Na
tion. A proposed rule applying the Fair Housing Act to insurers
will be drafted and widely distributed. Comments will be obtained
and the proposal will be revised prior to issuance of the final rule.
Throughout this process HUD will retain the assistance of outside
consultants with recognized credentials and experience in fair
housing and the business of insurance.
158
Q.3. Fair H ousing Litigation Policy. Ms. Achtenberg, I under
stand that HUD may have a policy to settle as many fair housing
claims [as] possible. Are you aware of any policy of HUD directing
attorneys to settle fair housing cases? What is HUD’s policy as to
settling fair housing cases?
A.3. I am not aware of any departmental policy on settling fair
housing litigation. However, the Secretary and I agree that HUD
should not be in the business of conducting protracted legal maneu
vers for the purpose of merely delaying a finding ana remedial
order against the Department. HUD has settled, or is engaging in
settlement negotiations, in several law suits where HUD’s liability
has already been established or where the Department has deter
mined that the plaintiffs have raised legitimate claims of violation
of law.
Q.4. O ccupancy Standard. Ms. Achtenberg, I understand that
HUD may be developing a new polity on occupancy standards. For
example, is HUD considering developing a new occupancy standard
that is broader than a 2-person per bedroom standard? Many
States establish a 2-person per bedroom standard as a legitimate
occupancy standard under the Fair Housing Act? If not, why not?
A.4. As you know, the Fair Housing Act prohibits discrimination
based on familial status—that is, based on the presence of children
under the age of 18 in housing. Any policy limiting the number of
persons who reside in housing may operate to disqualify or other
wise adversely affect families with children.
The guidance issued by former General Counsel Keating in 1991
permits consideration of a variety of factors, including the size and
configuration of bedrooms, to be considered in the establishment of
occupancy standards. Although that guidance has sometimes been
misinterpreted to allow housing providers to set two person per
bedroom occupancy standards in every situation, it does not in fact
authorize such action. There are a number of circumstances where
the availability of particularly large bedrooms, use of space other
than that denominated as “bedroom” space (such as dens or living
rooms), or other factors could result in upward revision of a two
person per bedroom guideline.
HUD is considering developing a new occupancy standard policy,
which will provide clearer standards regarding the number of per
sons who can occupy housing.
RESPONSE TO WRITTEN QUESTIONS OF SENATOR RIEGLE
FROM J. ROBERT HUNTER
Q .l. Most studies have been based on data grouped according to
zip code. Would it be helpful for the data to be grouped in smaller
geographic units, like census tracts, or nine-digit zip codes?
A.1. Yes. Five-digit zip codes are much smaller than the typical
rating territory for homeowners’ insurance and are therefore more
useful in evaluating insurance availability within a city or county.
However, five-digit zip codes are still large enough to capture two
or more neighborhoods with differing characteristics, as interpreted
by some insurers. The preferred method for geographic coding of
premium and loss experience is with a census tract, census block
group, or census block number. The census tract is typically small
159
er than a zip code and consists of a number of census block groups
which, in turn, consist of the smallest unit—census blocks. In addi
tion to a more precise mapping of particular neighborhoods or
neighborhood characteristics, the use of census mapping for insur
ance premium and losses allows for direct comparison with other
census information, including economic, demographic, and housing
statistics. Since zip codes do not exactly match census groupings,
census data by zip codes represent an estimate, albeit a statis
tically accurate estimate, of various economic, demographic, and
housing characteristics. The availability of nine-digit zip code in
surance premium and loss experience would allow for a more accu
rate mapping of zip code to census data.
Q.2. The American Insurance Association’s study on homeowners’
insurance found that blacks are three times more likely than
whites to purchase their insurance through State plans, the insur
ance pool of last resort for those unable to get insurance through
the regular market. State plan policies cost more and provide less
coverage than conventional homeowners’ policies. Is this evidence
of insurance discrimination, or did the study provide some alter
native explanation for the disparity? What are your general views
of the ALA study?
A £ . The AIA report purported to be a study of homeowners’ insur
ance availability. In tact, the study did not measure availability as
you or I would understand the term—the ability of people to pur
chase the insurance they want and/or need at affordable prices.
The AIA study surveyed homeowners in selected cities and found
that the vast majority of these homeowners had homeowners’ in
surance. This finding simply confirms that if you need a mortgage
to purchase a home, then you need to purchase insurance to protect
the mortgage lender. The study was essentially a tautology by lim
iting the sample to those residents who, unless they owned their
homes outright, had to have insurance. This is not a random sam
ple of insurance consumers. The study shed no light on the real is
sues of availability. Were people priced out of the home buying
market because of the price of insurance? Were some neighbor
hoods denied mortgage lending because insurers would not sell in
surance to protect the lenders? Were some consumers forced to ac
cept inferior policies because insurers would not offer full home
owners’ coverage in certain areas? The AIA study was silent on
these availability issues.
The AIA study does report a higher incidence of blacks purchas
ing insurance thorough State plans than whites. Thus, tor those
consumers who were able to obtain homeowners’ insurance, the
study suggests differential insurance availability by race.
Q.3. Would loss data have been helpful in analyzing the data you
collected for your study?
A & Yes. While we can evaluate insurance availability using only
premium and exposure data by zip code, it is only with the analysis
of loss data that we can evaluate whether the charged rates are
fair. Loss data by zip code, or by census block in the future, is es
sential for determining whether the rates charged in two different
geographic rating territories actually reflect tne expected losses
and are fair for those areas. With loss data by zip coae, we can de-
160
termine if the rates are fair within a rating territory, i.e., are the
existing geographic territorial groupings fair? In addition, loss data
by type of loss—theft, fire, wind & hail, freezing water, etc.—en
able us to evaluate insurers’ explanations for their underwriting
practices in different geographic areas.
Q.4. Ms. Schubert’s written testimony states repeatedly that phys
ical location alone should not be the basis of underwriting deci
sions. To what extent should physical location be significant in un
derwriting decisions?
A.4. Location can be important for exposure to catastrophic risk,
such as flood, earthquake, and wind damage. (Damage from wind
is the only one of these causes of loss covered in the typical home
owners’ policy.) Otherwise, location should not, in my view, be a
significant underwriting factor. Of course, the decision about how
to group risks geographically—the determination of rating terri
tories—is a fundamental public policy decision.
Q.5. Insurers frequently say that certain urban neighborhoods pay
more for insurance because of a higher incidence of crime and van
dalism. Is this accurate? How do insurance companies determine
the incidence of crime and vandalism? How do they draw the lines
between neighborhoods and decide which areas should be charged
more?
A.5. Insurance losses are caused by a variety of factors in addition
to theft and vandalism, including fire, wind & hail, freezing water,
and liability. All other factors equal, higher theft losses in one area
than another should result in higher rates in one area. However,
the other factors are rarely equal. For example, the amount of in
sured property, and thus the insurer’s exposure to loss, is typically
greater in neighborhoods with greater insurance availability than
those neighborhoods which insurers label as “high crime.”
Theoretically, insurers would determine the incidence of crime in
a given neighborhood by looking at loss data by cause of loss. For
all but the very largest insurers, looking at statistically valid loss
data by type of loss by zip code has been impossible because loss
data by zip code has not been available. Moreover, even the largest
insurers will not have loss data for those zip codes in which they
do not actively write business. Another source of crime data comes
from the FBI and/or local police departments. These data are typi
cally not available by small geographic areas within a city. Finally,
insurers may use the services of a third-party vendor who inter
prets existing sources of crime data into smaller geographic units.
An example of this type of product is the Geographic Underwriting
Service sold by the Insurance Services Office (ISO).
In some States, like Texas, the Department of Insurance promul
gates rating territories for all insurers. In most States, insurers ei
ther create their own rating territories or rely on those developed
by an advisory organization like ISO. In establishing rating terri
tories, insurers may create areas ranging in size from a single zip
code to several counties. We have found that the basis for insurers’
determination of rating territories has as much or more to do with
subjective factors as with objective characteristics of an area.
o
8 4-051
(1 6 4 )
@FedFRASER