Full text of Economic Perspectives (Federal Reserve Bank of Chicago) : July/August 1981, Volume V, Issue 4

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

Improving housing finance in
an inflationary environment:
alternative residential mortgage
instruments

CONTENTS

R ecen t increases in the leve l and
volatility o f m arket interest rates have
made fixed-rate, fixed-paym ent m o rt­
gage contracts less d esirable than they
used to be. A lternative m ortgage
instrum ents have b een d esig n ed to
m eet the cu rren t needs o f residential
m ortgage lend ers and b o rro w ers. This
article d escrib es the alternative m ort­
gage plans and analyzes their advan­
tages and disadvantages.

E C O N O M IC PERSPECTIVES
July/August 1981, Volum e V , Issue 4
Economic Perspectives is published bim onthly by the Research D epartm ent of the Federal
Reserve Bank of C h icag o . The p ub licatio n is produced under the d irectio n of H arvey
R osen blu m , Vice P resident, and is edited by Larry R. M ote, Assistant V ice President, w ith the
assistance of Sandra C o w e n (e d ito ria l), Roger Thryseliu s (artw o rk and graph ics), and Nancy
Ahlstrom (typesetting). The view s expressed in Economic Perspectives are the au th ors’ and do
not necessarily reflect the view s of the m anagem ent of the Federal Reserve Bank of Chicago or
the Federal R eserve System.
Single-copy sub scrip tion s of Economic Perspectives are available free of charg e. Please
send requests fo r single- and m ultip le-co p y sub scriptions, back issues, and address changes to
Public Info rm atio n C e n te r, Federal R eserve Bank of C h icago , P .O . Box 834, C h icag o ,
Illinois 60690, or telep h o ne (312) 322-5112.
A rticles may be rep rinted provided source is credited and Public Inform ation C e n ter is
provided w ith a copy of the published m aterial.
C o ntrolled circu latio n postage
paid at C h icago , Illino is.




Improving housing finance in an
inflationary environment: alternative
residential mortgage instruments
G eo rg e G. Kaufman and Eleanor Erdevig
A lternative mortgage instrum ents are m ort­
gage plans designed to accom m odate better
than traditional mortgages the current needs
of residential mortgage borrow ers, mortgage
lenders, or both. The long-term fixed-rate,
fixed-paym ent mortgage becam e the preval­
ent type in the U nited States in the 1930s and
served both borrow ers and lenders w ell as
long as p rice and interest rate movem ents
w ere relatively sm all. But recent increases in
the level and vo latility of m arket interest rates
have made this mortgage contract less desir­
able to lender and borrow er alike.
A hom e is the largest single purchase that
most A m ericans m ake and the largest single
item in their w ealth portfolio. Because of the
m agnitude of the expense— generally some
tw o to t h r e e times th e p u rc h a s e r ’s a n n u a l
incom e— a large portion of the purchase
price of most homes is borrow ed. The ow ner's
dow np aym ent, w hich represents the initial
e q u ity, is generally only 10 to 30 percent of
the purchase price. As a result, the cost,
term s, and availability of mortgage credit are
an im portant part of the hom e purchase
decision.
In recent years the cost of obtaining
mortgage cred it has increased sharply. For
exam p le, in 1971 the average interest rate on a
30-year single-fam ily fixed -coup on, fixedpaym ent mortgage with a loan-to-value ratio
of 90 p ercent was about 7Va percent. In 1975
the rate on this mortgage had clim bed to 9
p ercen t, and in 1980 it reached 121/2 percent.
O n a 90 p ercent fixed -rate, fixed-paym ent
mortgage on a m edian price hom e of $24,800
in 1971, this represented m onthly payments
of $160. The paym ents on a new mortgage on
a hom e of the same price would have in ­
creased to $182 in 1975 and $238 in 1980. But

Federal Reserve Bank of Chicago




the median price of existing homes also
increased, from $24,800 in 1971 to $62,200 in
1980. Thus, a rate of 12V2 percent on a 90
percent fixed-rate, fixed-paym ent mortgage
of $55,980 translates into m onthly payments
of $597.
M onthly payments have increased con­
siderably faster than household incom e. In
1971 the annual sum of m onthly payments
was equal to 19 percent of the median family
incom e of $10,285. In 1980 this percentage
had almost doubled to 34 percent of an esti­
mated m edian fam ily incom e of $21,652.1
Some households found this percentage too
high to pay and, as a result, decided not to
purchase a house. Because most hom ebuyers
are already housed and w ill sell their homes
when buying ano th er, the higher mortgage
rates reduce the turno ver of existing homes
and thus labor m obility, but have relatively
little effect on the overall stock of housing.
The dem and for new ly constructed housing is
m ore severely im pacted, but it accounts for
only a small proportion of total housing. In
the 1970s the average 1.7 m illion new units
co n stru cte d a n n u a lly acco u n ted for only
about 2.2 percent of the total housing stock of
some 79 m illion units.
The heaviest burden of the increase in
mortgage payments falls on first-tim e home
b uyers, w ho did not share in the rapid appre­
ciation in hom e prices and the perm issible
tax-free transfer of gains from one home to
another, and, in particular, on younger house­
holds, whose cu rren t incom es tend to be
below the average for all households. In addi1Assumes an increase of 10 percent in 1980 over the
reported median family income of $19,684 in 1979. Of
course, the sizes and amenities of homes purchased in
different years may be different.

3

tion, for these as w ell as other households,
the burden of the mortgage payments may be
expected to be greater at the beginning of the
loan than later w hen the expected rate of
inflation incorporated in the mortgage rate
and in the m onthly paym ent m aterializes and
increases incom es. This paym ent pattern,
w hich is generally referred to as the “ tilt,” will
be analyzed m ore closely in later sections of
the article. Because the tilt is com m on to
many mortgage plans, a large num ber of bor­
rowers w ould benefit from mortgage plans
that w o u ld re d u c e th e in itia l m ortgage
burden.
At the same tim e traditional mortgage
lenders, such as savings and loan associations,
com m ercial banks, and mutual savings banks,
have becom e increasingly reluctant to make
long-term, fixed-rate mortgage loans. Because
of the operating characteristics of these insti­
tutions, many of the fixed-rate loans they
have made in recent years have proven unpro­
fitable. O n the w hole these institutions raise
funds through short-term deposits. Through
tim e, a series of successive short-term depos­
its, each at a fixed interest rate, is equivalent
to a single long-term deposit at a variable
interest rate. In order to charge an interest
rate sufficient to cover the cost of the funds,
other operating costs, and a com petitive
profit when they relend the funds as a long­
term loan with the coupon or contract rate
fixed for the m aturity of the loan, the institu­
tions must predict their future cost of funds
over the life of the loan. If they predict their
future deposit costs in co rrectly, the loans
may be unp ro fitab le. Borrow ing (or lending)
on a fixed-interest rate basis for one maturity
w hile lending (or borrow ing) on a variablerate basis for the same m aturity is term ed
interest rate interm ediation.
Interest rate intermediation
Econom ic theory suggests a relationship
between short- and long-term interest rates.
In eq u ilib riu m and assuming certainty of
forecasts investors w ould be ind ifferent be­
tween buying two fixed-coupon rate secur­

4




ities differing only in term to m aturity if the
two securities w ere expected to yield the
same interest return over the investors' ex­
pected holding period. If one fixed-rate se­
curity had a m aturity less than that holding
period, the proceeds w ould have to be rein ­
vested in one or m ore successive securities. It
follow s that the investor must predict the
rates on these successive securities in order to
m ake a fully inform ed investm ent d ecision. If
one alternative prom ised a higher return than
the other, the funds w ould be invested in the
one prom ising the higher return. By bidding
up the price of the securities, this w ould serve
to low er the expected interest rates on this
alternative until the two alternatives promised
the same average expected return.
In m athem atical terms the yield on the
long-term , fixed-coupon security is equal to
an average of the current yield on the short­
term security and the yields predicted on suc­
cessive short-term securities over the life of
the longer-term security. For a depository
institution this relationship im plies that the
rate it charges on a long-term , fixed-coupon
mortgage loan should be equal to the average
of the rate it cu rren tly pays on short-term
deposits and the rates it expects to pay on
these deposits in the fu tu re plus operating
costs and allowances for both the risk of
default on the mortgage and a com petitive
profit.
This relationship is illustrated in figure 1.
Interest rates are plotted on the vertical axis
and tim e on the horizontal axis.2 Assume that
in the current p eriod, N, the institution pays
com petitive interest rate A for one-period
deposits and that this rate includes an allow ­
ance for operating costs and a com petitive
profit. The institution wants to make a fixedrate loan for q periods to period M . W hat rate
should it charge? The institution first needs to
predict its one-period deposit rate in each
period from N + 1 to M . If it expects the costs
2Because the interest on most securities is com­
pound interest, it is necessary to plot the logarithm of
(1 + i) on the vertical axis. But this may be interpreted as
simply the interest rate. For the sake of simplicity, the
diagram also assumes a nonamortized loan.

Econom ic Perspectives

Figure 1. Determination of interest
rate on fixed-rate mortgages depends
on projection of future deposit rates
interest rate*

•Technically, interest rate on the vertical axis is measured
in the form of ln(1+i).

of deposits to rise steadily to C along the
straight line A C , the average deposit cost, B, is
the rate it should charge on a long-term loan
in order to m ake a com petitive profit.
The expected total cost of the deposits is
the area N A C M . The expected revenue on
the loan is the area in the rectangle N BDM .
The two areas are eq ual, so that total ex­
pected revenues are equal to expected total
deposit costs. This equality may also be seen
by exam ining the d ifferences between the
interest rate earned on the loan and that
expected to be paid on the deposits. In
period N the profit is B - A. This profit, w hich
is in addition to the com petitive profit in ­
cluded in the deposit cost, is reduced steadily
as deposit rates are expected to rise until the
two rates are equal at E at tim e P. Th ereafter,
the deposit rate is expected to rise above the
loan rate up to C - D at period M . Although
the institution w ill exp erien ce losses from
time P through tim e M , it breaks even over
the en tire period N - M as the loss triangle
C D E is exactly equal to the earlier gain tria n ­
gle ABE. These losses must be charged against
the previous extra profits, w hich would be
classified m ore accurately as reserves against
future expected losses than as profits.

Federal Reserve Bank o f Chicago




Note that if deposit rates had been e x­
pected to d e clin e , the interest rate on the
long-term , fixed-coupon loan would have
been low er than the initial deposit rate. The
institution w ould have experienced losses at
the beginning of the p erio d , but would have
expected to recoup them later as deposit
rates d eclin ed below the loan rate. During
the loss period the institution is said to be
exp erien cing a liq u id ity p ro b le m as its cash
inflows are insufficient to satisfy its cash out­
flow s. But this is expected to be only a tem ­
porary problem and may be accom m odated
by using the reserves accum ulated in past
periods of greater than com petitive account­
ing profits.
Now assume that future deposit rates
w ere predicted in co rrectly. After a loan was
m ade, the cost of deposits actually increased
along line A G rather than A C , so that the cost
of deposits was underestim ated. The loan rate
charged, B, is now insufficient to cover the
actual cost of the deposits. The gain triangle
ABH is sm aller than the loss triangle DH G.
This is a lasting loss and, if not offset quickly
by unexpected gains experienced in periods
in w hich the increase in deposit rates was
overestim ated and the loan rate charged was
higher than necessary, w ill result in a solvency
p ro b le m . In retrospect, the institution should
have charged a loan rate of J.
T h u s, the im p o rta n ce fo r lon g-term ,
fixed -rate len d in g of co rre ctly predicting
future deposit rates is clear. In making fixedrate loans, the lender assumes all the risk of
unfavorable interest rate changes over the
life of the loan. It is effectively selling interest
rate in su ra n ce to th e b o rro w e r. Likeany insur­
ance co m pany, it may be expected to charge
a prem ium for this insu ran ce, the size of
w hich is dependent on the estimated degree
of risk in cu rred . This prem ium is sim ply
included in the interest rate charged the
borrow er.
M arket interest rates also incorporate an
inflation prem ium to com pensate lenders for
the expected loss in the value of their p rin ci­
pal due to inflation over the period that the
credit is outstanding. In recent years higher

5

and more volatile rates of inflation have
caused m arket interest rates to be higher than
most market participants— including most
depository institutions— expected. As a re­
sult, many long-term loans made at fixed
interest rates expected to be profitable at the
time of origination have turned out to be
unprofitable. That is, in retrospect the insur­
ance prem ium charged was insufficient to
com pensate the len d er for the loss incurred .
In addition, the increased volatility of both
inflation and interest rates has im paired the
co nfid ence of depository institutions in their
ability to predict future deposit rates and
therefo re to estim ate accurately the insu r­
ance prem ium to be charged.
Not only have depository institutions
been faced with higher and m ore volatile
market interest rates, but regulatory ceilings
have lim ited the rates that they are permitted
to pay on deposits. W hen m arket interest
rates rise above the m axim um interest rates
permitted on deposits, savers w ithdraw their
deposits to purchase unregulated financial
instrum ents in the m arkets. To enable com ­
m ercial banks and thrift institutions to co m ­
pete more effectively for funds in the open
market and to provide a m ore even flow of
funds for home mortgage lending, the regu­
latory agencies authorized m oney market
certificates (M M Cs) in 1978 and small savers
certificates (SSCs) in 1979, the rates on w hich
are tied to market rates on Treasury securities.
These moves w ere carried to their logical
conclusion by the D epository Institutions
Deregulation and M onetary Control Act of
1980 (D ID M C A ), w hich provides for phas­
ing out interest rate ceilin g s on all tim e
and savings deposits at depository institutions
by M arch 31, 1986.3 C o n seq u en tly, the aver­
age cost of funds of depository institutions
will increasingly reflect changes in market
rates of interest and therefore be even more
unpredictable in the future.
Because of the increased volatility in
3See “The Depository Institutions Deregulation and
Monetary Control Act of 1980,” Econom ic Perspectives,
Federal Reserve Bank of Chicago (September/October
1980).

6



deposit rates, many institutions have become
increasingly reluctant to engage in interest
rate interm ediation and to make long-term ,
fixed -rate loans. These institutions w ould
benefit from long-term loans whose interest
rates could vary more closely with their cost
of funds. U nexpected increases in the cost of
funds w ould be passed through to the b or­
rower and leave the institution u naffected.4
Alternative mortgage instrum ents are de­
signed to accom m odate the new needs of
both mortgage borrow ers and lenders. Be­
cause the problem s of borrow ers and lenders
d iffer, the mortgage plan best suited to one
may not be the one best suited to the other.
Risk averse mortgage lenders would benefit
most from mortgages whose rates fluctuate
closely with their cost of deposits, w hile
younger mortgage borrow ers stand to gain
from mortgages whose initial m onthly pay­
ments are low relative to their incom es.
U n fo rtu n ately, these two objectives are not
always m utually consistent. M ortgage plans
with increased rate volatility may not reduce
early m onthly payments and w ill transfer
interest risk from the lending institution to
the b o rro w er, w ho is often less w illing and
able to assume it. M ortgage plans with low er
initial m onthly payments need not increase
rate flexib ility and could result in monthly
payments to the depository institution smaller
than necessary to cover the m onthly interest
cost of the loan. Any shortfall would be added
to the p rincipal of the loan in the form of
negative am ortization and, by increasing the
unpaid balance of the loan, w ould increase
the risk of default. The increase in default risk
may offset part or all of the benefit to the
lender of the d eclin e in interest rate risk from
rate flexib ility. The fo llow ing sections des­
cribe a num ber of “ alternative” mortgage
plans, either already in use or proposed, and
their advantages and disadvantages to lend­
ers and borrow ers.
“•Regulations adopted by the FHLBB, effective July
10,1981, giving S&Ls broader latitude to engage in inter­
est rate futures transactions, are also expected to reduce
their net interest rate exposure. See Am erican Banker,
“ Broad Powers Given S&Ls to Use Futures,’’ Vol. 146, No.
130, Monday, July 6, 1981, page 1.

Econom ic Perspectives

Adjustable-rate mortgages

Adjustable-rate mortgages (A R M s), as
their name im plies, are mortgage plans whose
contract or coupon interest rates change
p erio d ically after origination according to
some agreed upon conditions. Pure A R M s,
that is, A R M s w ithout any restrictions on rate
changes, w ould elim in ate th e lending institu­
tio n ’s interest rate exposure com pletely if the
loan rate changed im m ediately every tim e
the deposit rate changed and by exactly the
same am o u n t.5 The change in the cost of
funds effectively w ould be passed through to
the mortgage b orro w er and not im pact the
institution at all. All the risk of unfavorable
interest rate changes that was borne by the
lender under fixed-rate mortgages (FRM s)
w o uld now be borne by the b orro w er. In
term s of figure 1, the mortgage loan rate on
an A R M w ould be superim posed on the d ep­
osit rate lin e. The institution would be freed
of any need to forecast interest rates in order
to price its mortgages co rrectly.6
But the A R M is not costless to either
le n d e r or b o rro w e r. A lth o u g h m ortgage
lending institutions incurred risk in making
long-term , fixed-rate mortgages, they charged
a prem ium for this service w hich until recent
years was su fficien tly high to make this activ­
ity profitable. If the institutions elim inate this
risk, they w ill also elim inate a line of business
that could again be potentially profitable.
Financial institutions, w hich are in the busi­
ness of dealing with interest rate changes
every day and are by nature fam iliar with
finance and eco n o m ics, may be expected to
be better situated than individual households
to m ake m eaningful interest rate predictions
and assume interest rate risk. Although house­
hold mortgage b orrow ers would benefit from
alternatively, the institution could eliminate inter­
est rate risk by altering its deposit structure to match the
characteristics of its fixed- and/or variable-rate
mortgages.
6The appropriate mortgage rate would be the rate on
deposits with a maturity equal to the period between
permitted changes in interest rates on the mortgage, plus
a premium for default risk and other costs associated with
originating and servicing mortgage loans.

Federal Reserve Bank o f Chicago




A R M s if interest rates w ere to fall, many tend
to be risk averse, putting greater weight on
interest rate increases than decreases of the
same m agnitude. They are generally w illing
to pay some prem ium to insure themselves
against the possibility of paying unexpectedly
higher rates during the life of the mortgage.
This is so even though the average borrower's
incom e rises w ith the interest rate as might be
expected if market rates incorporate a pre­
mium for expected inflation.
To achieve a com prom ise between these
two positions, the cum ulative change in in ­
terest rates on A R M s can be limited to a
predeterm ined band, e.g ., 2Vi or 5 percent­
age points or 30 or 50 percent above and
below the initial contract rate. M oreover,
individual rate changes also can be restricted
to some m axim um am ount— e.g ., Vi or 1 per­
centage point—and lim ited as to frequency—
e .g ., once every six m onths, every year, or
every five years. Thus, the interest rate risk is
shared by the b orro w er and the lender. The
cost of changes in market interest rates within
the overall band is borne by the b orro w er;
outside the band by the lender. Because the
width of the band affects the p re m iu m the
borrow er pays for “ interest rate insurance’’—
the low er the p rem iu m , the w ider the band—
the band acts like a deductability clause in an
accident or fire insurance policy.
A R M S do not mitigate a particular form
of the “ tilt” problem that arises when market
interest rates increase in response to upward
revisions of inflationary expectations. These
rate increases are translated im m ediately into
increases in m onthly paym ents, w hile the
b o rro w er’s incom e increases only slowly with
the realized rate of inflation. Thus, the bur­
den of the mortgage increases im m ediately.
Through tim e, as the b o rro w e r’s incom e rises
and m onthly payments remain unchanged,
the burden w ill declin e. N evertheless, the tilt
of the burden to the early months of the
mortgage may price some potential home
buyers out of the market.
A variety of A R M plans have been used
or proposed. In January 1979 the Federal
Hom e Loan Bank Board (FH LBB) permitted

7

federally chartered savings and loan associa­
tions to make variable-rate mortgages (VRM s)
on w hich the interest rate could change after
origination with the cost of funds to all fed e r­
ally insured savings and loan associations, but
by no m ore than Vi percentage point once a
year, and by no more than 2Vi percentage
points over the life of the m ortgage.7 M onthly
payments would change accordingly.
In 1980 the FH LBB perm itted a variation
on this them e term ed the renegotiable-rate
mortgage (R R M ) in w h ich the interest rate
could again change by V2 percentage point
annually but the m axim um change over the
life of the loan increased to 5 percentage
points. M o reo ver, the changes in the rate
could be translated into changed m onthly
payments only once every three to five years
and w ere tied to changes in the national aver­
age rate on mortgages as m easured by the
FHLBB contract mortgage rate index for co n ­
ventional mortgages for the purchase of exist­
ing hom es, rather than to the cost of funds to
the lenders. Because the new mortgage rate
was based on loans by all lenders, not just
savings and loan associations, and because it
is highly publicized, the causes of rate changes
on loans could be better understood by bor­
rowers. H ow ever, the inability of the new
mortgage rate to change in perfect synchron­
ization with the cost of funds means that the
lender is not fully protected from interest rate
risk. In ad d ition, as the rem aining life of the
mortgage declines through tim e, the appli­
cable rate on the mortgage may be expected
to differ m ore and m ore from the prevailing
rate on new fixed-rate m ortgages, w hich c u r­
rently dom inate the new mortgage index.
O n M arch 23, 1981, the C o m p tro ller of
the C u rre n cy (C O C ) authorized national
banks to offer, w ithin specified guidelines,
adjustable-rate mortgage loans for the pur­
chase of one- to four-fam ily ow ner-o ccup ied
homes. U nder the regulation the interest rate
on a mortgage loan may change in accor­
dance with any one of three specified refer7Federally chartered associations in California were
granted this authorization in 1978.

8



Federal regulation ot mortgage lending
Some have questioned w h e th er the C o m p ­
tro lle r of the C u rre n c y has the authority to
prom ulgate adjustable-rate mortgage regula­
tions for national banks and to preem pt state
laws that restrict such mortgages. Tw o recent
federal court decisions support Federal Hom e
Loan Bank Board preem ptive regulations estab­
lishing uniform standards for real estate lending
by fe d e ra lly c h a rte re d saving s and loan
associations.
In C o n fe r e n c e o f Federal Savings and Loan
Associations v. Stein, 604 F.2d 1256 (9th C ir.
1979), a ffirm ed 445 U.S. 921 (1980), the state of
C alifo rn ia had req uired federal savings and
loan associations to abide by the provisions of
the state's an ti-red linin g act. The court co n ­
cluded that w here federal regulation such as
the regulatory control of the FH LBB over fed ­
eral S&Ls is so pervasive as to leave no room for
state regulatory co n tro l, im plicit preem ption
can be fo und.
In G len dale Federal Savings and Loan A sso ­
ciation v. Fox, 459 F. Supp. 903 (C .D . C al. 1978),
appeal p e n d in g , the co urt reasoned that the
Congress had given the FHLBB com plete author­
ity to charter and regulate federal S&Ls so as to
proh ibitstatesfro m reg ulatingthem . (A num ber
of other circu it courts have reached sim ilar
co nclusio ns.) T h e re fo re , w ith respect to m ort­
gage loans extended by federal S&Ls, federal
law can preem pt state regulation of the validity
and e xercisab ility of “ due-o n-sale” clauses re ­
qu iring com plete repaym ent of the mortgage
w hen the hom e is sold.
Federal law governing m ortgage lending
by national banks is contained in the general
and sp ecific rule-m aking authority granted by
Title 12 of the United States C ode on banks and
banking, 12 U .S .C . §1, et seq. (especially §371 (g),
w hich states that loans are subject to conditions
and lim itations prescribed by the C o m p tro ller
of the C u rre n c y by rule or reg u latio n ), the
National Bank A ct, other federal banking laws,
and the Housing and C o m m u n ity D evelopm ent
Act of 1974 (w hich liberalized the powers of
national banks to m ake real estate loans). Taken
to gether, these provisions suggest that it was
the intent of the Congress that the C o m p tro ller
of the C u rre n cy regulate real estate lending by
n a tio n a l b a n k s , th u s p r e e m p tin g sta te
regulations.

Econom ic Perspectives

ence rates by a m axim um of 1 percentage
point every six m onths with no limit on the
cum ulative change over the life of the m ort­
gage. At the option of the bank, monthly
paym ents may be m aintained at a fixed dollar
am ount for a specified tim e up to five years,
regardless of changes in the interest rate. If
the bank chooses this o p tio n , increases or
decreases in the interest rate w ill result in
changes in the proportions of the m onthly
paym ent credited to interest and repaym ent
of prin cipal. If the required interest payment
is greater than the am ount of the fixed
m onthly paym ent, the differen ce may be
added to the outstanding principal in the
form of negative am ortization within a speci­
fied lim it. Upon expiration of a period of
fixed m onthly paym ents, the m onthly pay­
ment is adjusted up or dow n for the next
period to provide full am ortization of the
outstanding p rincipal balance w ithin the re­
maining tim e to m aturity of the loan.
The FHLBB adopted regulations A pril 30,
1981, that granted federally chartered thrift
institutions broader flexib ility in thedesign of
adjustable-rate mortgages. Interest rate ad­
justm ents on the loans may be tied to any
referen ce rate provided that it can be verified
easily by the b o rro w er and it is not controlled
by the len d er. The interest rate may be
adjusted through changes in the m onthly
paym ent and /or in the loan term , subject
only to the conditions that the loan term from
the date of closing be lim ited to 40 years and
that the paym ent am ount be adjusted at least
every five years to a level sufficient at the
existing interest rate to am ortize the loan fully
over its rem aining life. (The restrictions im ­
posed by the two agencies are sum m arized
on page 22.) In allow ing such a broad range of
options, the FH LBB and the C O C are clearly
assuming the developm ent of a large variety
of individual mortgage plans with varying
degrees of risk sharing. Because many bor­
row ers may not wish to assume the risk of a
fully adjustable mortgage rate and many m ort­
gage lenders may not wish to spin off their
interest rate risk insurance business alto­
gether, this forecast is likely to be correct.

Federal Reserve Bank o f Chicago



Developm ent of mortgage plans

The new regulations of both the C O C
and the FHLBB perm it lenders to develop
mortgage plans designed to meet both their
own needs and those of borrow ers. The var­
iety of mortgage plans that can be designed
and offered under the regulations of both the
C O C and the FH LBB is virtually infinite. In
p ractice, h o w ever, variations in the design
are likely to depend prim arily on differences
in a relatively few key characteristics. These
in clu d e : (1) the choice of a reference rate;
(2) the frequency and amount of adjustment
of the interest rate; (3) the frequency and
am ount of change in the m onthly paym ent;
(4) the lim itations on additions to the out­
standing principal b alance, i.e ., negative
am ortization; and (5) perm itted extensions
of the m aturity of the original loan.
Using a co m puter, it is possible to sim u­
late the behavior of several critical variables
such as the frequency and amount of interest
rate adjustm ents of different mortgage plans.
Such sim ulations are useful in evaluating the
impact of each mortgage plan on both the
borrow er and the lender. For exam ple, the
sim ulations allow the borrow er to compare
the b e h av io r o ver tim e of the exp ected
m onthly paym ent and the outstanding p rin ­
cipal balance of each plan. In addition, they
enable the lending institution to project its
cash flow and interest incom e over the life of
the loan.
Sim ulations can be run using either data
from some past p erio d — in w hich case they
show how particular types of mortgages would
have perform ed if they had been made at the
beginning of that period— or hypothetical
data believed useful in illum inating the likely
future p erform ance of mortgages with d iffer­
ing characteristics. The relative desirability of
the various types of mortgages depends heav­
ily on the actual behavior of inflation and
m arket interest rates in the fu tu re, and that
may not resem ble either the past or any other
assumed behavior. N evertheless, given one's
expectations about the fu tu re, simulations
provide a useful way of exam ining the impli-

9

cations of those expected conditions for the
perform ance of different mortgages.
Choice of reference rate
Both the C O C and the FHLBB require
that adjustm ents in the mortgage interest rate
be tied to a sp ecific referen ce rate. National
banks are authorized to use the six-month
Treasury bill auction rate, the three-year co n ­

stant m aturity Treasury note rate, and the
FHLBB national average contract mortgage
rate on conventional mortgages for the p ur­
chase of existing hom es. A federal S&L may
use any agreed-upon published rate, but not
its own mortgage rate or its own cost of funds.
Sim ulations of mortgage plans with the
interest rate tied to each of the three re­
ference rates authorized by the C O C for the
period from January 1970 through Decem ber

Table 1
Sim ulations of A R M plans with alternative reference rates
1970-1980
($20,000 m ortgage)1
FHLBB contract rate*2
Six-month
period

Monthly
payment

Ending
balance

Six-month T-bill3
Monthly
payment

Three-year T-note3

Ending
balance

Monthly
payment

S&L cost of funds3

Ending
balance

Monthly
payment

Ending
balance

19,920
19,839
19,739
19,605
19,476
19,336
19,201
19,078
18,961
18,847
18,737
18,604
18,480
18,335
18,171
17,999
17,835
17,684
17,543
17,406
17,276
17,160

148.57
152.76
154.43
154.70
154.57
154.84
155.24
155.78
159.23
162.94
166.64
167.01
167.43
167.55
168.19
168.07
169.18
169.92
172.97
178.31
184.12
197.03

19,920
19,842
19,762
19,679
19,593
19,503
19,409
19,312
19,215
19,119
19,022
18,922
18,817
18,707
18,592
18,472
18,348
18,218
18,087
17,956
17,827
17,708

(dollars)
1970.1
.2
1971.1
.2
1972.1
.2
1973.1
.2
1974.1
.2
1975.1
.2
1976.1
.2
1977.1
.2
1978.1
.2
1979.1
.2
1980.1
.2

148.57*
149.68
145.97
140.06
139.40
138.36
140.42
142.60
152.54
157.09
163.61
158.05
160.34
157.58
157.46
157.34
158.78
164.29
172.35
179.47
192.96
198.47

Average

157.97

19,920
19,838
19,749
19,647
19,541
19,430
19,317
19,203
19,000
18,997
18,899
18,789
18,676
18,555
18,428
18,295
18,159
18,025
17,897
17,772
17,658
17,544

148.59
147.07
132.75
109.56
117.80
107.64
118.93
135.98
156.75
155.43
156.39
130.45
137.12
125.81
122.87
120.89
132.77
141.21
157.87
170.45
184.86
196.45
141.26

19,920
19,835
19,727
19,572
19,430
19,261
19,113
18,993
18,898
18,797
18,692
18,541
18,397
18,224
18,040
17,844
17,671
17,508
17,369
17,243
17,129
17,023

148.57
150.03
138.19
120.16
125.27
120.83
126.20
135.47
142.12
146.77
152.71
141.68
149.48
140.95
133.32
132.87
139.35
149.41
159.03
166.19
175.13
189.93
144.71

164.98

'Assumes 30-year $20,000 adjustable-rate mortgage at 8.13 percent closed in January 1970 and tied to indicated
reference rate. Mortgage interest rate and monthly payment are adjusted every six months without restriction on either.
-Mortgage loan interest rate for each six-month period is the actual FHLBB mortgage loan contract rate on loans
closed in January and July.
'Mortgage loan interest rate for each six-month period after the first is the average rate during the prior six months
plus the spread between 8.13 percent and the average rate during the last six months of 1969.
■'Monthly payment for comparable fixed-rate mortgage; ending balance in 1980.2 would be $17,228.

70



Econom ic Perspectives

1980 are shown in table 1. For the sake of
sim p licity, all of the sim ulations in the table
assumed that a 30-year mortgage loan for
$20,000 was closed in January 1970 at 8.13 p er­
cen t, the actual FH LBB national average co n ­
tract rate on fixed-rate mortgage loans for the
purchase of existing homes at that tim e. The
mortgage loan interest rate was adjusted
every six m onths and the m onthly payment
changed accordingly w ithout restriction. The
am ount of the adjustm ent in the interest rate
was equal to the change in the reference
rate.8
The sim ulations indicate that an adjust­
able-rate mortgage closed in January 1970
and tied to the six-m onth Treasury bill rate
would have required the lowest average
m onthly paym ent during the 11-year period
and w ould have resulted in the lowest ending
principal balance of the three adjustable
mortgage plans. The average m onthly pay­
ments and the ending principal balances for
the mortgage plans with interest rates tied to
the six-m onth Treasury bill rate and to the
three-year Treasury note were both lower
than those for the fixed-rate mortgage. The
results of a sim ulation of a mortgage plan with
the interest rate tied to the FHLBB mortgage
contract rate w ere very sim ilar to those of a
mortgage plan tied to the S&L’s average cost
of funds. S&Ls' cost of funds has increased
steadily since 1970, and mortgage interest
rates, although declining over short periods
of tim e, have also been in a generally upward
trend. Both of these plans w ould have had
paym ents higher than those on a fixed-rate
mortgage.
These results are specific to the period
used, h o w ever. Interest rates w ere at a rela­
tively high level in 1970 and the subsequent
d eclin e in rates w ould have required a co r­
responding reduction of the interest rate on
8The monthly rates were averaged over the sixmonth periods in the above simulations except when the
reference rate was the FHLBB mortgage contract rate.
The use of averaging generally reduces the amount of
fluctuation in the reference rate and may also prevent
tying a mortgage loan interest rate to a reference rate,
particularly a sh'ort-term interest rate, that is temporarily
distorted.

Federal Reserve Bank o f Chicago




an adjustable-rate mortgage. The w ider flu c­
tuations of the six-m onth Treasury bill rate
require greater changes in the mortgage loan
interest rate. Sim ulations of mortgage plans
beginning w hen interest rates w ere relatively
low and for shorter periods of time w ill show
different results. In addition, all mortgage
plans w ould not be expected to have the
same initial interest rate.
Adjustment of the interest rate
Although the above sim ulations adjusted
the mortgage loan interest rate every six
months by the full am ount of the change in
the referen ce rate, the interest rate may be
adjusted more or less frequently and the
am ount of the adjustm ent may be limited
both at each adjustm ent period and over the
term of the loan. M ore frequent adjustment
translates changes in market interest rates
into changes in mortgage loan rates more
quickly. Restrictions on the amount of the
adjustm ent w ill reduce the correlation of
mortgage loan rates with market interest
rates.
The regulations of the C O C limit the
adjustm ent of the mortgage interest rate in
either d irectio n to 1 percentage point every
six months with no limit on the cum ulative
change over the life of the mortgage. In the
above sim ulations this restriction would have
had little effect on the use of the FHLBB co n ­
tract mortgage rate and the three-year Trea­
sury note rate. It w o uld , how ever, have re­
duced the am ount of perm itted change if the
six-m onth Treasury bill had been used as the
referen ce rate. A com parison of mortgage
plans with the six-m onth Treasury bill rate as
the referen ce rate with and w ithout the lim ­
itation to a 1 percentage point change every
six months is shown in table 2. During the
period from 1970 to 1980, the limitation on
the adjustm ent to the interest rate in this
instance resulted in a slightly lower average
m onthly payment and ending balance out­
standing because the higher market interest
rates w ere not fully translated into higher
mortgage loan interest rates.
The FHLBB regulations impose no spe-

11

Table 2
Simulation of ARM plan with C O C limitation
on adjustment of interest rate1
1970-1980
No limitation
Six-month
period

Monthly
payment

C O C limitation

Ending
balance

Monthly
payment

Ending
balance

(d o lla r s )

1970.1
.2
1971.1
.2
1972.1
.2
1973.1
.2
1974.1
.2
1975.1
.2
1976.1
.2
1977.1
.2
1978.1
.2
1979.1
.2
1980.1
.2

148.57
147.07
132.75
109.56
117.80
107.64
118.93
135.98
156.75
155.43
156.39
130.45
137.12
125.81
122.87
120.89
132.77
141.21
157.87
170.45
184.86
196.45

Average

141.26

19,920
19,835
19,727
19,572
19,430
19,261
19,113
18,993
18,898
18,797
18,692
18,541
18,397
18,224
18,040
17,844
17,671
17,508
17,369
17,243
17,129
17,023

148.57
147.07
133.61
120.81
117.95
107.77
119.07
131.59
144.06
155.42
156.38
143.90
137.28
125.96
123.02
121.04
131.78
141.36
153.96
165.35
176.91
188.60

19,920
19,835
19,729
19,596
19,454
19,285
19,137
19,009
18,896
18,795
18,691
18,563
18,418
18,246
18,062
17,866
17,690
17,527
17,381
17,247
17,123
17,008

140.52

'Assumes 30-year $20,000 adjustable-rate mortgage at 8.13 per­
cent closed in january 1970 and tied to six-month Treasury bill rate
as referencerate. Mortgage interest rate and monthly payment are
adjusted every six months.

cific lim itations on the freq uency or am ount
of the adjustm ent of the mortgage loan in te r­
est rate. The frequency and am ount of change
w ill thus depend on the reference rate used
and how often it is available and on any lim ita­
tions and term s that are incorporated in the
mortgage contract by the lending institution.
Changes in the monthly payment
Changes in the m onthly payment may be
made w h en ever the interest rate is changed
or at some other agreed upon tim e. In the
latter instance the m onthly paym ent may be
held constant for a given period during w hich
the interest rate is perm itted to fluctuate in
accordance with the specified reference rate
and any sp ecific lim its in the contract. If the
required m onthly interest payment exceeds

12



the total m onthly paym ent, the m onthly pay­
ment must be increased to cover the interest,
unless negative am ortization is perm itted.
W hen the m onthly paym ent is changed, it is
generally increased or decreased, as neces­
sary, to am ortize the loan fully over the re­
maining term of the mortgage.
The earlier FHLBB regulations auth o riz­
ing VRM s and RRM s required a co rresp o nd ­
ing change in the m onthly payments w h en ­
ever the interest rate on the mortgage was
increased or decreased. The dollar amounts
of the changes in the m onthly paym ent were
limited by the restrictions on the periodic and
overall increases in the mortgage loan inter­
est rate. H ow ever, no such interest rate lim ita­
tions are included in the FHLBB regulations
issued in A pril 1981, and the C O C regulations
lim it the change in the mortgage interest rate
to 1 percentage point every six m onths. Thus,
unless restrictions are imposed on changes in
m onthly paym ents, interest rate changes will
be translated im m ediately into changes in
monthly payments.
Restrictions on changes in the m onthly
paym ent—so-called payment caps— are an
alternative to or may be used in conjunction
with lim itations on changes in the interest
rate. For exam ple, the m onthly payment may
be held constant for a given period, say three
years, but the interest rate on the loan may
change every six m onths. O r the monthly
payment may change sim ultaneously with
changes in the interest rate, but any increase
may be lim ited to some dollar or percentage
increase over the payment in the prior period.
Specific restrictions on payment caps
w ere not included in the recent regulations
of either the C O C or the FH LBB. H ow ever,
several com m ercial banks and thrift institu­
tions have been offering 30-year A R M s since
the fall of 1980 with lim itations on payment
changes.
Negative amortization
Any upward adjustm ent in the mortgage
interest rate that is not accom panied by a
change in the monthly payment sufficient to

Econom ic Perspectives

am ortize the loan over the rem aining life of
the loan may req u ire negative am ortization,
i.e ., an addition to the outstanding loan bal­
ance. As noted above, this may be necessary
w hen the m onthly payment is held constant
fo ra given period of tim e but the interest rate
increases or w hen the increase in the monthly
paym ent is lim ited to a specified percentage
or am ount.
Although the regulations of both the
C O C and the FH LBB perm it negative am orti­
zation in A R M s, the two differ som ewhat. The
FHLBB regulation requires that the m onthly
payments be adjusted at least once every five
years to am ortize fu lly, w ithin 40 years from
the date of closing, the outstanding principal
at the interest rate indicated by the reference
rate. The C O C regulation limits negative
am o rtizatio n , for periods during w hich the
m onthly paym ents are fixed , to no more than
1 percent of the principal outstanding at the
beginning of the fixed-paym ent period times
the num ber of six-m onth intervals w ithin the
fixed-paym ent period. M onthly payments
must be adjusted at least every five years to an
am ount su fficien t to am ortize the outstand­
ing p rincipal over the rem aining term . Thus,
the m axim um negative am ortization perm it­
ted over a five-year period would be 10 per­
cent of the p rincipal outstanding at the be­
ginning of the period.
D uring the period of negative am ortiza­
tion the ratio of the loan to the value of the
house increases and may becom e greater
than 1.00. W hen this o ccu rs, the lending insti­
tution incurs significant risk of default be­
cause the b orro w er may “ walk away” from
the loan. Because housing prices tended to
rise sharply in recent years, greater than 100
percent loan-to-current-value ratios would
have been u n like ly. N evertheless, all housing
prices did not rise and even the average
house price may not rise as rapidly in future
years. As a result, the initial loan-to-value
ratio takes on additional significance.
Sim ulations of mortgage plans that incorp oratethe lim itationson interest ratechanges
and negative am ortization prescribed by the
C O C regulations indicate that, with an inter­

Federal Reserve Bank o f Chicago




est rate of 14 p ercen t, the lim it on negative
am ortization is not breached, even when the
interest rate changes by the maximum allow ­
able 1 percentage point every six months,
unless the m onthly payment remains con­
stant for at least two years. In table 3 the
interest rate on a $1,000 mortgage loan with
an original rate of 14 percent was increased 1
percentage point every six m onths, the max­
imum perm itted. W hen the monthly pay­
ment was held constant for at least one year,
negative am ortization always occurred in the
second and subsequent six-m onth periods
until the m onthly payment was adjusted.
H o w ever, the perm itted lim it on negative
am ortization was not reached unless the
m onthly paym ent was unchanged for at least
two years and then not until the second half
of the sixth year of the mortgage term. In all
sim ulations, when the m onthly payment was
held constant for a year or m ore, the out­
standing balance of the loan was above $1,000,
the original am ount of the loan, at the end of
12 years.
Extension of loan maturity
In order to lessen the impact of an
increase in the interest rate on the required
m onthly paym ent, the FHLBB regulations
perm it extension of the m aturity of the loan
up to a m axim um of 40 years from the date of
closing. H ow ever, when interest rates are
high, an increase in the mortgage interest
rate, by even a small am ount, quickly results
in the extension of the loan term to the full 40
years, as shown in figure 2. This is especially
true in the early years when interest rep re­
sents substantially all of the monthly pay­
ment. The im m ediate benefit to the borrower
is, th erefo re, lim ited , and the monthly pay­
ments continue for 40 years instead of the
original 30.
Graduated-payment mortgages
Graduated-paym ent mortgages (GPMs)
are adjustable-paym ent mortgages on which
the m onthly mortgage payments increase

13

Simulations of ARM plans with C O C negative amortization limit and selected constant payment periods
b. end-of-period outstanding balance
($1,000 mortgage)

a. average monthly payment
($1,000 mortgage)

Six-month
period

1.1
.2
2.1
.2
3.1
.2
4.1
.2
5.1
.2
6.1
.2
7.1
.2
8.1
.2
9.1
.2
10.1
.2
11.1
.2
12.1
.2

Interest
rate

Constant payment period
1 year
2 years
3 years
6 mos.

(percent)

(dollars)

14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37

11.85
12.64
13.44
14.24
15.05
15.85
16.67
17.48
18.29
19.11
19.92
20.74
21.56
22.37
23.19
24.01
24.83
25.65
26.47
27.29
28.11
28.93
29.75
30.57

11.85
11.85
13.51
13.51
15.19
15.19
16.92
16.92
18.66
18.66
20.43
20.43
22.22
22.22
24.03
24.03
25.86
25.86
27.72
27.72
29.59
29.59
31.48
31.48

11.85
11.85
11.85
11.85
15.52
15.52
15.52
15.52
19.49
19.49
19.49
19.73*
23.71
23.71
23.71
24.21 ‘
28.19
28.19
28.19
28.89*
32.93
32.93
32.93
33.82*

11.85
11.85
11.85
11.85
13.32*
14.93*
17.60
17.60
17.60
17.71*
20.06*
21.20*
23.98
23.98
23.98
24.51*
26.82*
28.08*
31.00
31.00
31.00
31.81*
34.22*
35.60*

Six-month
period

1.1
.2
2.1
.2
3.1
.2
4.1
.2
5.1
.2
6.1
.2
7.1
.2
8.1
.2
9.1
.2
10.1
.2
11.1
.2
12.1
.2

Interest
rate

Constant payment period
1 year
2 years
3 years
6 mos.

(percent)

(dollars)

14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37

999
998
997
996
996
995
995
994
994
993
993
993
993
992
992
992
992
991
991
991
991
990
990
990

999
1003
1002
1006
1006
1010
1010
1014
1014
1019
1018
1023
1023
1028
1028
1033
1033
1038
1038
1043
1043
1048
1048
1053

999
1003
1012
1028
1027
1032
1042
1059
1059
1064
1075
1092
1092
1097
1109
1126
1126
1131
1144
1161
1161
1166
1180
1197

999
1003
1012
1028
1041
1051
1050
1055
1066
1084
1094
1105
1104
1110
1122
1139
1150
1161
1161
1167
1180
1197
1209
1220

‘ Limit on negative amortization requires increase in monthly
payment.

according to a p red eterm ined sched ule.
These plans are designed prim arily for young­
er fam ilies whose incom es are cu rren tly low
but may be expected to rise faster than aver­
age as they enter the m ore productive years
of their lives. G raduated-paym ent mortgages
were originally authorized by the Housing
and C o m m unity D evelopm ent Act of 1974,
and five plans w ere subsequently offered
by H U D -FH A . The plan in w hich monthly
paym ents could increase by IV.2 percent
annually for five years proved to be the most
popular. Since th en , a num ber of private
lending institutions have offered co nventio n­
al graduated-paym ent mortgage plans sim ilar
to this plan.
U nder many of these plans, the low early
m onthly paym ents may be sm aller than the
monthly interest on the outstanding balance.

14



If so, the d ifferen ce is added to the principal
of the mortgage loan. As the monthly pay­
ment increases, it eventually rises above the
am ount necessary to pay the interest on the
outstanding principal and the principal be­
gins to am ortize. The later payments are suf­
ficien tly higher than the required interest
in o rd e r to o ffse t th e e a rlie r n e g a tive
am ortization.
Default risk is potentially a serious prob­
lem with graduated-paym ent mortgages. The
higher the initial ratio of loan to house value,
the low er must be the graduation in order to
avoid having the loan value rise above the
house value. The risk of default may also be
greater on GPM s than on other mortgages
because the borrow er's incom e may fail to
increase as rapidly as the m onthly payments.
A lth o u g h the d o lla r am o u n t of the

Econom ic Perspectives

Figure 2. Increase in mortgage
interest rate necessary to extend
loan term from 30 to 40 years
percent

m onthly paym ent varies on all G PM s, the
interest rate may or may not. Alm ost all GPM s
that have been offered are fixed-rate m ort­
gages. Th u s, these plans do not assist lending
institutions in reducing their interest rate risk.
G rad u ated -p aym en t adjustable-rate m ort­
gages (G PAM s) have recently been proposed
by the FH LBB. Although G PAM s would offer
institutions better protection from interest
rate risk, rapid increases in interest rates in
the early years w ould raise sharply the e ffe c­
tive rate of graduation, thereby increasing
default risk. In addition, the variability in
m onthly paym ents w ould be increased to a
greater extent than with regular ARM s under
like circum stances. N evertheless, GPAM s may
be a w o rkab le com prom ise instrum ent that
reduces both the interest rate risk exposure
of thrift institutions and the early monthly
payments burden of econom ically promising
younger hom e buyers.
Price level-adjusted mortgages
In periods of rapid inflation, the m ort­
gage plans discussed so far (except GPM s)
req u ire, at the tim e of their origination, large
m onthly paym ents relative to the mortgage
borrow er's incom e. As the borro w er’s incom e
increases through tim e in line with the rate of
in flatio n , the burden of the m onthly pay­

Federal Reserve Bank of Chicago




ments declines and offsets the previous in ­
crease so that, for the en tire life of the m ort­
gage, the burden to the borrow er is no
higher. As discussed e a rlie r, the mortgage
paym ent-to-incom e ratio is said to be tilted
dow nw ard. This troublesom e “ tilt” problem
can be reduced or elim inated by a price leveladjusted mortgage (PLAM ) w hich ties the
m onthly paym ents on the mortgage to the
price level rather than to the interest rate.
Thus, accelerations in the rate of inflation are
translated into higher mortgage payments
only as they occu r rather than all at once
when they are first anticipated and become
em bodied in interest rates as an inflation
prem ium .
In addition, m onthly payments under a
PLAM reflect actual rates of inflation, rather
than expected rates of inflation. As a result,
neither mortgage lenders nor mortgage bor­
rowers are injured (or rew arded) financially if
the expected rate of inflation im pounded in
the interest rate is not realized. Like almost
everyone else, participants in mortgage mar­
kets badly underestim ated the future rate of
inflation throughout most of the 1960s and
1970s. In retrospect, the rates on fixed-rate
mortgages in these years were lower than
necessary to m aintain the purchasing power
of the p rincipal. The resulting loss to the lend­
ing institutions is the major cause of the
financial difficulties these institutions are now
exp erien cing . In contrast, mortgage borrow ­
ers enjoyed w indfall gains.
Because PLAM s both protect lenders
from unexpected increases in the rate of
inflation and reduce the initial burden to
mortgage b orrow ers, it might appear that
they are sup erio r to A R M s, w hich do only the
first. But this is not necessarily so. PLAM s pro­
tect lenders from unexpected interest rate
increases only to the extent that those in ­
creases are attributable to unexpected in­
creases in the rate of inflation. ARM s protect
lenders from u n expected interest rate in ­
creases regardless of the cause of those in­
creases. Thus, A R M s provide greater interest
rate risk protection to lenders than PLAMs
but less protection to borrow ers.

15

T h e o p e ra tio n of th e
Table 4
price level-adjusted mortgage
Simulations of price-level adjusted mortgages
($60,000 mortgage)
plan is shown in table 4. As­
sume that in the absence of
inflation the rate on a 30-year
V* expected inflation3
V i expected inflation2
No tilt 1
fix e d - ra t e , fix e d - p a y m e n t
Ending
M onthly
M onthly
Ending
M onthly
Ending
mortgage would be 5 percent.
balance
payment
balance
Year
payment
balance
payment
The m onthly payments on a
(d o lla r s )
loan of $60,000 on an $80,000
59,730
571.39
59,391
399.18
1
59,115
322.09
hom e w ould be $322. If the
60,617
347.86
582.82
62,261
62,839
423.13
2
61,480
594.48
borrow er's incom e rem ained
65,209
448.52
375.69
66,725
3
606.37
62,311
68,227
4
405.74
475.43
70,768
at, say, $30,000 per year, the
63,104
618.49
503.96
71,303
438.20
74,959
5
m onthly mortgage payments
63,851
630.86
74,425
79,287
534.19
6
473.26
64,541
643.48
77,576
511.12
566.25
7
83,735
would be equal to 12.9 p er­
65,164
80,737
656.35
88,282
600.22
8
552.01
cent of pretax incom e. Ignor­
65,708
669.48
83,883
636.23
596.17
92,903
9
ing taxes, if the rate of infla­
66,157
682.87
86,987
674.41
643.87
97,562
10
66,496
696.52
714.87
90,012
695.38
11
102,219
tion w ere suddenly expected
66,706
710.45
106,824
757.76
92,920
751.01
12
to accelerate to 8 percent
66,766
724.66
95,661
811.09
111,313
803.23
13
annually, the mortgage rate
66,651
739.16
98,179
875.97
115,612
851.42
14
66.333
753.94
100,409
902.50
946.05
119,633
15
on new loans might jum p to
769.02
65,781
102,272
123,268
956.65
16
1021.73
13 percent in order to m ain­
64,959
784.40
103,679
1103.47
126,392
1014.05
17
63,825
800.08
104,523
1074.90
1191.75
128,853
tain the purchasing pow er of
18
62.333
104,684
816.09
130,477
1287.09
1139.39
19
the loan and the m onthly pay­
832.41
60,429
104,019
131,057
1207.75
1390.06
20
ment on new loans would
58,052
849.05
102,366
130,349
1280.22
21
1501.26
55,132
866.03
99,535
1621.37
128,071
1357.03
22
rise to $664. Assum ing that
51,590
883.36
95,308
1751.07
123,892
1438.45
23
the new borrow er's incom e
47,337
901.02
89,434
117,427
1524.75
24
1891.16
does not increase im m ediate­
919.04
42.270
1616.24
81,623
2042.45
108,231
25
36.270
937.42
71,544
1713.21
2205.85
95,785
26
ly, the payments would rep re­
29,206
956.17
58,814
1816.00
2382.32
79,488
27
sent 26.5 percent of incom e.
20,926
975.29
42,994
1924.96
2572.90
58,646
28
994.80
11,256
A sim ilar pattern exists for
23,582
2788.73
2040.46
32,459
29
0
1014.70
3001.04
0
2162.90
0
30
ARMS.
The P L A M , h o w e v e r,
'Assum es 5 percent real interest rate and 8 percent inflation.
would increase the m onthly
^Assumes 5 percent real interest rate and 8 percent inflation but incorporates
2 percent expected inflation in mortgage rate.
paym ent by only the actual
'Assum es 5 percent real interest rate and 8 percent expected inflation but in co rp o­
inflation realized in the p eri­
rates 6 percent inflation in mortgage rate.
od. If this w ere also 8 percent,
the payment would increase
by 8 percent from the contract am ount of
pal rises sharply through the early and m iddle
$322to $348 after 12 m onths. At the same time
years of the mortgage and declines sharply in
the principal on the loan also would be
the later years. This occurs because a large
adjusted by the 8 percent rate of inflation. If
part of the early payments represents interest.
the borrow er's incom e increased at the aver­
Thus, little of the principal is paid back and
age rate of inflatio n , the paym ent would
the principal increases by almost the full
rem ain at 12.9 percent of incom e. In the next
am ount of the rate of inflation. Near the end,
period both the mortgage paym ent and in ­
almost all of the m onthly paym ent represents
com e would again increase by equal p ercent­
principal reduction and the outstanding p rin­
ages so that the mortgage burden remains
cipal declines sharply.
constant.
PLAM s have three potential problem s.
As indicated by figure 3, the loan princiFirst, the ratio of the m onthly payment to the

76



Econom ic Perspectives

Figure 3. Comparisons of monthly
payments and outstanding balances
of FRM* and PLAM**
monthly payments (dollars)

ending balance (dollars)

years
•Assumes $60,000 mortgage loan with loan term of 30
years and 13 percent mortgage interest rate.
••Assumes $60,000 original mortgage loan with 5 percent
real interest rate and 8 percent rate of inflation.

incom e of a given borro w er w ill rem ain co n ­
stant only if that b o rro w er’s incom e changes
m ore or less in line w ith the particular price
index used. If increases in the b o rro w er’s
incom e lag substantially behind the index,
the mortgage burden w ill increase through
tim e and might increase the risk that the bor­
row er w ill default. The possibility of default
might be reduced by choosing an appro­
priate price index and, possibly, by modifying
the paym ents sched ule to reintrod uce some
but not all of the tilt.
The p rice index to w hich the mortgage
paym ents are tied should reflect the ability of
households to service their mortgage debt.
Th u s, it w o uld be appropriate to use a wage
index. But, as noted, everyo ne's incom e does
not change by the average am ount. Further­
m ore, under most circum stances, some in ­
creases in labor productivity might be e x­
pected through tim e. If so, prices will increase
m ore slow ly than wages. Because incom e is
the desirable characteristic to track, it would
be m ore desirable to use a price index asso­
ciated with inco m e, such as the GN P deflator,

Federal Reserve Bank o f Chicago




than a price index associated with a fixed
market basket of goods, such as the consum er
price index.
The m aintenance of some tilt in the
mortgage paym ent-to-incom e ratio may be
desirable as a means of reducing the risk of
default. This can be achieved by indexing
only part of the d ifferen ce between the nom ­
inal market and real mortgage interest rates
rather than all of the d ifferen ce. The real
mortgage rate may be defined as the constant
purchasing pow er mortgage rate or the rate
that w ould exist if prices at the end of the loan
period w ere expected to be the same as at the
beginning. Assum e, as before, that the annual
rate of inflation over the next 30 years is
expected to be 8 p ercent, so that in the
absence of incom e taxes the market rate
might be 13 p ercent if the real rate is 5 per­
cent. The mortgage plan might add to the real
rate one-quarter of the interest rate com po­
nent attributable to the expected rate of infla­
tion. This w ould increase the initial m onthly
payment and change the subsequent monthly
paym ent and p rincipal amounts by only
three-quarters of the actual rate of inflation.
For the exam ple above, the initial interest rate
w ould be 5 plus 2, or 7 percent. The monthly
payments on this “ m o d ified ” PLAM would be
higher initially than on the “ p u re” PLAM , but
would increase m ore slow ly through time
and eventually becom e sm aller. They would
also increase m ore slow ly than average in ­
com e, so that for most borrow ers the burden
of the mortgage payments would decline.
Because the initial m onthly payments are
higher, the principal outstanding on the loan
w ould not increase as rapidly. The amounts
for the $60,000 mortgage loan discussed ear­
lier aresh o w n in fig u re 4 . Note that, unlike for
the pure PLA M , the b o rro w er’s payment-toincom e ratio d eclin es, but by less than for the
fixed- or adjustable-rate mortgage plans.
The second difficulty concerns the cash
flow to mortgage lenders. If the len d er’s
prim ary source of funds is regular interestbearing deposits, then a liquidity problem
may arise. In the absence of deposit rate ceil­
ings, an acceleration in the expected rate of

17

Figure 4. Comparisons of monthly
payments and outstanding balances
of PLAM* and modified PLAMs**

2

4 6 8 10 12 14 16 18 20 22 24 26 28 30
years

♦Assumes $60,000 original mortgage loan with loan term
of 30 years and 5 percent real interest rate and 8 percent rate of
inflation.
♦♦Assumes $60,000 original mortgage loan with loan term
of 30 years and 5 percent real interest rate and 8 percent rate of
inflation, but PLAM i incorporates 2 percent expected rate of
inflation and PLAM2 incorporates 6 percent expected rate of
inflation in mortgage interest rate.

inflation w ill im m ediately increase the cost of
deposits by the full amount of any accom pany­
ing increase in interest rates. But revenues
from PLAM s w ill increase m ore slow ly in line
with the actual rate of inflation. O ver the life
of the m ortgage, of course, the increases in
incom e inflow and exp en d itu re outflow will
be equal. Thus, the “ tilt” risk has been shifted
from the b orrow er to the lending institution.
The institution may protect itself by offering
similarly indexed deposits— price level-adjust­
ed deposits or PLADs— or by tem porarily
dipping into reserves to finance the d iffe r­
ence. Because the institutions are larger, bet­
ter positioned in the financial markets to b or­
row funds, and m ore know ledgeable in the
area of fin an ce , they may be expected to be
better equipped to manage the tilt problem
than most mortgage borrow ers.
The third d ifficulty arises because most
mortgage lenders and borrow ers are subject
to incom e taxes. Lenders must share part of
their interest incom e with the governm ent,

18



w h ile borrow ers may deduct their interest
costs from their taxable incom e. Thus, as
mortgage paym ents increase w ith the rate of
inflation , the aftertax incom e of taxable len d ­
ers from the mortgage loan w ill increase
m ore slow ly and be less than is required to
maintain the purchasing pow er value of the
payments. At the same time the aftertax cost
of the mortgage paym ents to taxable b orro w ­
ers w ill increase by less than is required to
maintain the purchasing pow er value un­
changed. That is, on an aftertax basis, the
lenders are not fu lly protected, w h ile the bor­
rowers benefit. This asymmetry may be co r­
rected by indexing the mortgage payments
and principal payments to the price index by
a factor to correct for the tax effects.9
Although the incom e tax brackets of the
individual participants in the mortgage m ar­
ket may be expected to vary, a factor based on
the estimated average marginal tax rate for all
participants may be a w o rkab le solution in
practice. Sim ulations of m onthly payments
and principal amounts for a PLAM plan
assuming an average marginal tax rate of 25
percent are shown in table 5. In these sim ula­
tions the aftertax ratio of mortgage payments
to incom e rem ains constant w h ile the pretax
ratio increases.

Shared-appreciation mortgages
The recent period of high rates of in fla­
tion and interest has also been characterized *1
9The appropriate factor for bonds is 1/1-t where t is
the marginal tax rate; the factor for amortized mortgages
is more complex because the monthly payments are
divided between interest and principal. For the ratio of
aftertax mortgage payments to income to remain con­
stant for a PLAM plan, the factor, g', for increasing the
outstanding mortgage balance must be less than the
inflation factor and is:
,

1 - t + t (1+i)-n

a — ---------------- -------1----------

where

x

R

1 - t + t (1+i)-n+1
t= marginal tax rate
i = interest rate
n = number of payments
g = 1 + inflation rate.

The derivation of this factor has been provided by Henry
J. Cassidy, Office of Policy and Economic Research,
FHLBB.

Econom ic Perspectives

interest must be specified as a
percentage, say, 30 or 50 per­
cen t, of the amount of appre­
ciation accrued, either at the
Ratio of
tim e the hom e is sold or at
aftertax
payments
some designated earlier date.
to income
This interest is received only
at that tim e. Thus, shared ap­
10.5
preciation represents contin­
10.5
10.5
gent deferred interest. Until
10.5
th e s e t t le m e n t d a te th e
10.5
10.5
am ount of this interest is un­
10.5
10.5
certain. The other component
10.5
of the interest is determ ined
10.5
10.5
at the origination of the loan
10.5
and is paid regularly through
10.5
10.5
e
ith e r fix e d or graduated
10.5
10.5
m onthly payments.
10.5
The two com ponents of
10.5
10.5
the interest payments are re­
10.5
lated. The higher the amount
10.5
10.5
of
contingent interest expect­
10.5
10.5
ed at origination, w hich de­
10.42
pends both on the expected
10.4
10.4
am
ount of appreciation and
10.4
10.4
on the appreciation sharing
10.4
ratio, the lower the im m e­
diate interest rate and the
corresponding m onthly pay­
ments. Co nversely, the lower
the expected contingent interest, the higher
the im m ediate interest rate and monthly
paym ents. As eith er the am ount of apprecia­
tion expected or the sharing ratio declines,
the SAM approaches a fixed-rate, fixed- or
graduated-paym ent mortgage.
Potential hom e buyers may find a SAM
d esira b le if th ey can no t affo rd the high
m o n th ly paym ents associated w ith other
mortgage plans. They w ould be able to defer
a part of their interest payments until the
necessary liquidity could be obtained through
the sale of the house or through an im prove­
ment in their cash flow situation.
Although appealing, the SAM presents a
n u m b er of sig n ifican t d raw b acks for the
lender, particularly if it is a thrift institution,
for the b o rro w er, and for the relevant regula­
tory agency. W ith respect to the lender, SAMs

Table 5
Simulation of PLAM
with constant ratio of aftertax mortgage payments to income1
($60,000 mortgage)

Annual
income

Annual
pretax
payments

30,000
32,400
34,992
37,791
40,815
44,080
47,606
51,415
55,528
59,970
64,768
69,949
75,545
81,589
88,116
95,165
102,778
111,001
119,881
129,471
139,829
151,015
163,0%
176,144
190,235
205,454
221,891
239,642
258,813
279,518

3,903
4,200
4,518
4,859
5,226
5,619
6,040
6,491
6,975
7,493
8,047
8,641
9,276
9,954
10,680
11,455
12,283
13,167
14,110
15,115
16,187
17,329
18,545
19,838
21,214
22,676
24,230
25,879
27,628
29,483

Ending
balance

Annual
aftertax
payments

Ratio of
pretax
payments
to income

(dollars)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30

59,097
62,566
66,155
69,855
73,651
77,529
81,469
85,444
89,426
93,377
97,255
101,007
104,574
107,884
110,855
113,392
115,384
116,703
117,203
116,718
115,056
112,001
107,306
100,692
91,845
80,409
65,983
48,119
26,313
0

3,153
3,405
3,677
3,970
4,287
4,629
4,998
5,397
5,827
6,292
6,794
7,335
7,920
8,552
9,233
9,%9
10,763
11,621
12,547
13,546
14,625
15,789
17,046
18,403
19,868
21,449
23,156
24,998
26,986
29,132

13.0
13.0
12.9
12.9
12.8
12.7
12.7
12.6
12.6
12.5
12.4
12.4
12.3
12.2
12.1
12.0
12.0
11.9
11.8
11.7
11.6
11.5
11.4
11.3
11.2
11.0
10.9
10.8
10.7
10.5

’Assumes 5 percent real interest rate and 8 percent inflation.
2Change due to internal computer rounding.

by very high rates of appreciation in residen­
tial property values. To the extent that m ort­
gage lending institutions underestim ated the
rate of inflation and, in retrospect, charged a
low er fixed mortgage rate than w arranted, a
greater share of the appreciation in value
accrued to the hom e ow ner. In light of this
exp e rie n ce , some mortgage lenders have
begun to view appreciation in hom e values as
protection against the risk of unexpected
inflation and w ould prefer to index their
mortgage interest rates to this appreciation.
This may be achieved in a shared-appreciation
mortgage (SAM ).
The SAM is m ore com plex than the other
types of alternative mortgages. The interest
rate consists of two com ponents. Because the
am ount of the app reciation , if any, is un cer­
tain at the tim e a loan is extended, part of the

Federal Reserve Bank o f Chicago




19

do not alleviate the liq uid ity problem arising
from interest rate interm ed iatio n. Ind eed , to
the extent that the regular m onthly mortgage
payments are low er than on other mortgage
plans, higher deposit costs from inflation are
more d ifficu lt to pay out of cu rren t revenues,
even in com parison with fixed-rate m ort­
gages. The liquidity problem s of the institu­
tions are exacerbated.
SAM s are also likely to increase substan­
tially the degree of credit risk assumed. The
future value of residential real estate is not
known and the degree of co nfid ence with
w hich values can be predicted d eclines as the
time to m aturity of the mortgage increases.
Thus, at the tim e of o rig in atio n , the actual
return is u ncertain. M o reo ve r, the price of
real estate has not always increased rapidly
and, over some periods, it has not increased
at all. Between 1947 and 1978, the average
return realized on residential real estate was
about 7 percent per year. This was tw ice as
great as the average annual increase in co n ­
sum er p rices, but less than half the average
return on the stock m arket. By decades, the
return on residential real estate was 5 percent
in the 1950s, 6 percent in the 1960s, and 10
percent in the 1970s through 1978. Thus, the
double-digit rates of hom e appreciation in
recent years w ere not typical of past exp e­
rience. Most of us know of individual homes
or individual neighborhoods that have d e­
clined in value.
If the lend er's expected appreciation is
not realized , the im m ediate interest rate
charged on the loan w ill prove in retrospect
to have been too low and the lender's real­
ized return may well be lower than on other
mortgage plans. D eterm ination of future
housing values by geographical and neigh­
borhood area is also likely to require person­
nel with d ifferen t skills than those the institu­
tions currently em ploy.
Because of neighborhood effects, two
otherw ise identical homes owned by eco ­
nom ically com parable households may have
different expected rates of appreciation. Thus,
at origination, the same lending institution
would charge d ifferent im m ediate interest

20



rates on the two properties. This is likely to
lead to resentm ent on the part of the bor­
row er being charged the higher rate and to
charges of "re d lin in g ” if entire n eighbor­
hood areas are charged higher initial m ort­
gage rates.
Lenders can generally reduce the degree
of default risk they assume by diversifying
among a num ber of different loans whose
default probabilities may not be expected to
be closely related. But many thrift institutions
make loans in a lim ited geographical area. All
the houses in the area may reasonably be
expected to be subject to sim ilar underlying
changes in value. Thus, any one lender may
not be able to diversify this risk away. Real
estate is also relatively illiquid so that the
estimated market price may not be realized
q uickly. In terms of a hedge against u n ex­
pected inflation, SAM s may not be much less
risky than equities, w hich have generally
b e e n c o n s id e re d to o ris k y fo r th rift
institutions.
Because the lender's return is dependent
on the am ount of appreciation in the home's
value after the loan is originated, anything
that affects the potential for appreciation is of
concern to the institution. Thus, the lender is
likely to require m inim um m aintenance stan­
dards and to engage in p eriodic surveillance
to see that they are met. If the standards are
not m aintained, then costly en fo rcem ent and
legal actions may be necessary. Capital im ­
provem ents made by hom eow ners generally
add value to the house that is greater or
sm aller than the cost of the im provem ents
them selves. Separation of the appreciation in
the value of a hom e between the unim proved
hom e and the capital im provem ents is d iffi­
cult and may also result in controversy and
litigation.
Lastly, to reduce the liquidity strains,
SAM s generally provide for the lender to
receive instalm ents on the deferred co ntin ­
gent incom e from appreciation before the
final m aturity, say, every five or ten years.
(Regulations recently proposed by the Fed­
eral Hom e Loan Bank Board effectively re­
quire that the m axim um period for adjust-

Econom ic Perspectives

merit be no longer than ten years.) If the
hom e is not sold before this period ends,
interim settlem ent is based on the appraised
value at that tim e. The accrued interest may
be paid in cash as a lum p sum or be refi­
nanced by a mortgage not providing for
shared appreciation.
These features present two potential
problem s for the len d er. First, the am ount of
the appraised value may not be accepted by
the b o rro w er, causing ill feelings, loss of good
w ill, and possibly litigation. (O f course, the
appraised value at times may be less than the
institution w ould like.) Even sale prices may
be co ntro versial. A seller may sell the house at
an artificially low price to a friend for other
com pensation to avoid paying the lender its
share. The lender can protect itself against
this possibility by reserving a right of first
refusal.
Second , the refinancing of the accrued
interest delays further the cash inflows to the
lending institution, intensifying any liquidity
pressures. M o re o ve r, if the appreciation is
large so that the accrued interest is large, the
increase in m onthly payments may cause
finan cial strains on the b orrow er at the tim e
the mortgage is refinanced . This increases the
risk of default on top of any increase in liquid­
ity pressures. The Federal Hom e Loan Bank
Board proposes that the refinancing be guar­
anteed by the lending institution “ without
regard to the . . . b o rro w er’s incom e . . . for
a term of not less than thirty years.”
The disadvantages of SAM s to the lend ­
ing institutions are not necessarily advantages
to the b o rro w er. Because of the lender's
interest in the appreciation of the hom e, the
b orrow er loses partial control over the man­
agem ent of the hom e. Changes in the house,
capital im pro vem ents, and even color of
paint may need to be approved by the lender
in advance. Freedom to make decisions of
this nature ind epend ently is often view ed as
one of the m ajor advantages of hom e o w n e r­
ship. Disagreem ents with respect to m ainte­
nance, co ntribu tion of capital im provem ents,
and assessed value at interim adjustm ent
dates are lik ely to be, at m inim um , nuisances

Federal Reserve Bank o f Chicago




and, at w o rst, may involve the borrow er in
exp en sive, tim e consum ing, and unpleasant
litigation. Lastly, SAM b orrow ers e xp e ri­
encing a large appreciation in the value of
their homes and wishing to sell and move may
not be able to purchase another home of
equal value w ithout financial strain because
part of any appreciation must be paid to the
lender on sale.
Th e reg u la to ry agencies sup ervising
banks and thrift institutions are charged with
the responsibility for protecting the financial
solvency of the institutions. To do so, they
evaluate the quality of the institutions’ loans.
It is d ifficu lt, if not im possible, to judge accu­
rately the correctness of the institution’s pre­
dicted rate of appreciation for a large number
of w id ely d ifferen t hom es and, th erefo re, the
quality of the loans. M o reo ve r, even to try to
do so w ould requ ire a staff with both skills
and training different from those that exa­
m iners cu rren tly possess, as well as substantial
fam iliarity with the characteristics of the geo­
graphical areas served by each lending insti­
tution. This w ould increase significantly both
the cost and d ifficulty of the agency’s task.
In light of these problem s, the SAM is
probably not a viable alternative mortgage
instrum ent at this tim e for depository institu­
tions. Further w o rk is required to correct
some of these lim itations before it can make a
major positive co ntribu tion to the mortgage
market.
Conclusions
It is evident that the traditional fixedcoupon rate, fixed-paym ent mortgage is no
longer king of the hill. Prim arily because of
the dual im pact of rapid inflation and high
and volatile interest rates, it no longer serves
the needs of all b orrow ers and lenders. Some
borrow ers find the high initial m onthly pay­
ments required by this mortgage an undue
burden, w h ile many lenders view making
long-term loans at a fixed rate financed by
deposits at an adjustable rate as too great a
risk. Thus, fixed -rate, fixed-paym ent m ort­
gages are being supplem ented by a large

21

num ber of alternative mortgage plans tai­
lored to the sp ecific needs of individual
mortgage borrowers and lenders. These m ort­
gage plans include adjustable-rate mortgages,
graduated-paym ent m ortgages, price level-

adjusted m o rtgages, sh a re d -a p p re c ia tio n
mortgages, and a variety of com binations
thereof.
All of these mortgage plans differ in at
least one com m on way from the fixed-rate,

Major characteristics of recent federal regulations
governing adjustable-rate home mortgage lending
Federal savings and loans
and mutual savings banks

National banks

Requirem ent to offer fixed-rate
mortgage instrument to borrower

None

None

Limit to amount of A R M s that may
be held

None

None

Any interest rate index that is readily
verifiable by the borrower and not
under the control of the lender, in­
cluding national or regional cost-offunds indexes for S&Ls.

O n e of three national rate indexes—
a long-term mortgage rate, a Trea­
sury bill rate, or a three-year Trea­
sury bond rate.

M ajor characteristics

Indexes governing mortgage rate
adjustments

Limit on frequency of rate
adjustments

None

Limit on size of periodic rate

None

Not m ore often than every six
months.
1 percentage point for each sixmonth period between rate adjust­

adjustments

ments, and no single rate adjustment
may exceed 5 percentage points.
Limit on size of total rate adjustment

None

None

over life of mortgage
Allow able m ethods of adjustment to
rate changes

Limit on amount of negative am orti­
zation

Any com bination of changes in
monthly payment, loan term , or
principal balance.
No limit, but monthly payments must
be adjusted p e rio d ically to amortize
fully the loan over the remaining
term.

Advance notice of rate adjustments

Prepayment restrictions or charges

Disclosure requirem ents

22




Limits are set, and m onthly payments
must be adjusted periodically to
amortize fully the loan over the
rem aining term.
30 to 45 days prior to scheduled

30 to 45 days prior to scheduled
adjustments.

adjustments.
Prepayment without penalty per­
mitted after notification of first
scheduled rate adjustment.

None

Full disclosure of ARM character­
istics no later than time of loan
application.

S O U R C F : David F. Seiders, “ Changing Patterns of Housing Finance,”
the Federal Reserve System (June 1981), p. 468.

Changes in m onthly payment or rate
of amortization.

Full disclosure of ARM character­
istics no later than time of loan
application.

F e d e r a l R e s e r v e B u lle t in ,

Board of Governors of

Econom ic Perspectives

fixed-paym ent mortgage— they are consid­
erably m ore co m plex. This hampers both the
design and operation of efficient plans and
their acceptance by household borrow ers.
Partially in response to this problem , the
agencies w ith prim ary resp o n sib ilities for
regulating the m ajor residential mortgage
lending institutions— the Federal Home Loan
Bank Board for savings and loan associations
and the C o m p tro ller of the C u rre n cy for
national banks— have issued regulations pre­
scribing the m ajor features and restrictions of
these mortgage plans. In itially, the regula­
tions specified very carefu lly the perm issible
characteristics of each mortgage plan, in par­
ticu la r, the degree of flexib ility in changing
interest rates and/or monthly payments. M ore
recen tly, the agencies have issued more gen­
eral regulations with greatly reduced restric­
tions, in p articular, on rate and payment flex­
ibility for adjustable rate mortgages. These
changes reflect, in part, the current trend
tow ard deregulation of firm s and markets
and, in part, the b elief that com petition in the
m arket w ill foster the developm ent of m ort­
gage plans satisfactory to both borrow ers and
lenders.
The more liberal regulations place the
burden of developing m arketable mortgage
products on the lenders. For exam ple, to the
extent that b orrow ers are risk averse and do
not wish to assume the total interest rate risk,

Federal Reserve Bank o f Chicago




lenders w ill need to design plans that share
that risk. To the extent that borrowers' in ­
com es do not change as rapidly as interest
rates, lenders w ill need to design plans that
m aintain m onthly payments unchanged for
some time when interest rates change w ith ­
out incurrin g undue default risk from in ­
creases in the value of the loan. It is likely that
individual lending institutions w ill design a
num ber of differentiated mortgage plans, as
is the case with other products sold by a large
num ber of producers in a com petitive market.
The design of the mortgage plans will also be
greatly influenced by the policies established
by the Federal Hom e Loan Mortgage C o rp o­
ration and the Federal National Mortgage
Association for purchases in the secondary
m arket.10 The sim ulations presented in this
article illustrate the major characteristics of
the mortgage plansthat have been permitted
or proposed and may be useful to institutions
in designing their plans.

10The Federal National Mortgage Association an­
nounced on June 25, 1981, that it would make commit­
ments to purchase eight types of adjustable rate mort­
gages based on five different indices beginning in late
July. The program announced by the Federal Home Loan
Mortgage Corporation in late May was more limited. See
Am erican Banker, “ FNMA Unveils Adjustable Rate
Mortgage Plan," Vol. 146, No. 125, Friday, June 26,1981,
page 3, and “ FNMA’s Adjustable-Rate Mortgage Pro­
posal,” Vol. 146, No. 132, July 8, 1981, page 2.

23

Public Information Center
Federal Reserve Bank
of Chicago
P.O. Box 834
Chicago, Illinois 60690
Please attach address label to correspondence
regarding your subscription.