View original document

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

DOCUMENT

RESUME

00277 - [A0751125]
Highlights of a Study cf Federal Supervision of State and
National Banks rVol. I J
OCG-77-la; b-11831; 1.-118535;
B-168904. January 31, 1977. 56 pp. + appendices (102 pp.).
Report to the Congress;

by Eimer D.

Staats, Comptroller General.

Issue Area: Internal Auditing Systems (200); Federal Regulatory
Activities (3000).
Contact: Office of the Comptroller General.
Budget Function: General Government: Central Fiscal Operations

(8C3).
Organization Concerned: Federal Deposit Insurance Corp.; Federal
Reserve System; Office of tht Comptroller of the Currency.
Congressional Relevance: House Committee on Banking, Currency
and Housing; Senate Committee on Banking, Housing and Urban
Affairs; Congress.
Authority: National Banking Act. Edge Act. Financial
Institutions Supervisory Act of 1966. Federal Reserve Act of
1913. Banking Act of 1933.

Several congressional committees requested the
evaluation of the effectiveness of the supervisory efforts of
the three Federal agencies involved in monitoring banking
operations, because of the increasing instab'lity of banks. The

study objectives were to evaluate the agencies' efforts to (1)
identify unsound conditions and violations cf laws in banks, and
(2) cause bank management to take corrective actions.
Examination reports and correspondence files on sore than 900
banks supervised by FDIC, Office of the Comptroller of the
Currency, and the Federal Reserve Boards were examined,
including 30 of 42 banks that had failed, 294 of ?87 problem
banks, and a general sample of 600 of the banks in the United
States.
Findings/Conclusions: Adverse economic conditions
contributed to some bank failures, but generally embezzlement
and pocr management of loans were the cause. Prctlems were not
corrected because: (1) the regulatory agencies were reluctant to
use their legal authority to force the banks to change, (2) the
agencies did not consult with bank boards, (3) examinations were
set up on a time basis rather than a problem solving basis, and
(4) recommendations were not generally made as to how to solve
problems. Examiners have enforcement tools they may use, both
informal and formal: (1) informally request chat banks make the
changes, (2) formal written agreements to confirm correction
plans, (3) cease and desist orders, (4) removal cf management,
(5) financial assistance, (6) cancellation of deposit insurance,
(7) cancellation of Federal Reserve membership, and (8;
revocation or charter. Federal Reserve Board surveillance of
bank holding companies is not adequate. training of examiners is
not ad:quate. Major improvements of bank supervision include
organizational chenges, closer bank surveillance, self-dealing
and insider transaction monitoring, consumer protection law

enforcement, new examination procedures,
contact with
bank bcards, problem solving monitoring, closer
more use of formal
powers, experiments on relying on state
t raining
examinations, and better
of examiners. The agencies involved
are not working as
closely as they shotld. Recommendations:
The
agencies should
revise their eya&ination practices
and frequencies to better
identify problems. Examination reports
boards should follow all examinations. and meetings with bank
More aggressive policies
should be developed for the use of
formal
actions against
problem banks. Better training and
examiners should be implemented. Thescreening of iotential
three agencies, either
through their own initiative cr legislation,
should coordinate
their efforts more closely. More stringent
procedures
for
handling charter applicaticns should
be devised. (Ss)

REPORT TO THE CON(GREfSS
;/:

t

-,·

BBY THE (C.MPTROLLERtGENE''A!.
~OF
~ THE UNITED STATES

Highlights of a StudyA
of Federal Supervision of
State and National. Banks

OCG-77-lo a

A

C3MFTZROLLER cENERAL OF THE UNITED STATES
WASHINGTON. D.C.

20548

B-114831
B-118535
B-168904

To the President of the Senate and the
Speaker of the House of Representatives
This report highlights our unprecedented study of
the effectiveness of State and rational bank supervision
by the Federal Deposit Insurance Corporation; the Federal
Reserve System; and the Office of the Comptroller of the
Currency, Department of the Treasury.
This study was made at the request of several congressional committees concerned over large bank failures ir
recent years and public disclosure that supervisory agencies'
lists of "problem banks" had lengthened.
Our OLfice does not have legislative authority to

audit

the operations of the Federal Reserve System or the Comptroller
Also, our access to the bank examination
of the Currency.
reports of the Federal Deposit Insurance Corporation has long
been a matter of dispute.
In light of the heavy congressional interest in the
They agreed,
area, the agencies allowed us to make tl.e study.
bank
their
to
access
us
unlimited
give
in April 1976, to
provided
records,
related
other
and
examination reports
we would not disclose any information about specific banks,
bank officers, or bank custcmners.
The focus of the report is on evaluating the agencies'
bank examination functions and their efforts to get banks to
Several recommendations foL
correct problems identified.
improvements are made.

B-ll] 'o3l
B-118535
B-168904

The three agencies have reviewed and commented on a
Their comments are presented, in
draft of the report.
In view of the time
full, as appendixes to the highlights.
constraints placed on us for completing and releasing
the study, we have not been able to fully evaluate their
comments.
In the past we have supported proposals before the
Congress to give this Office continuing legislative
authority to review the operations of the bank regulatory
With such authority,
agencies and report to the Congress.
we could be more helpful to tne Congress in carrying
out its legislative and oversight responsibilities for
In view of the very important
bank insurance and regulation.
part that the three agencies play in the Nation's system
of money and credit, we feel that the Congress should
provide for GAO audits of the agencies.
We are sendirg copies of these highlights to the
Secretary of the Treasury; the Comptroller of the Currency;
the Chairman, Board of Governors of the Federal Reserve
System; and the Chairman, Board of Directors of the Federal
Deposit Insurance Corporation.

Comptroller General
of the United S.at'P

-

2

Contents

Page
Introduction
What lessons can be drawn
from recent bank failures?
Are Federal bank examinations
of adequate scope to identify
significant problems in banks?
Once problems are identified, are
they communicated effectively to
banks?
Have agencies been effective in
getting banks to take corrective
action?
What observations did GAO have
on FRS surveillance of bank
holding companies?
What did GAO find regarding examiners'
competence and objectivity?
What progress have the agencies made
in the past year in improving their
examination and supervision of banks?
How close is the working relationship
among the Federal bank regulatory
agencies?
Hr'w does GAO feel about the agencies'
need for additional powers?
Has the Comptroller of the Currency
considered applications for national
bank charters on a fair and consistent
basis?
Have some banks changed charter.ng
authorities to avoid supervisory
pressure?
What do bankers think of bank supervision?

1
6

10

19

22a

31
35

39

44
48

50

52
54

APPENDIX
I

II

III

Letter dated January 14, 1977, from the
Comptroller of the Currency, to the
General Accounting Office

I-1

Letter dated January 16, 1977, from the
Chairman, Federal Reserve Board, to the
General Accounting Office

II-1

Letter dated January 17, 1977, from the
Chairman, Federal Deposit Insurance Corporation, to the General Accounting Office

III-1

ABBREVIATIONS
CSC
EDP
FDIC
FW1
FRS
GAO
NBSS
OCC

Civil Service Commission
electronic data processing
Federal Deposit Insurance Corporation
Federal Reserve bank
Federal Reserve System
General Accounting Office
National Bank Surveillance System
Office of the Comptroller of the Currency

INTRODUCTION

Why GAO made this study
The Congress is concerned with the soundness of the
commercil. banking system. In the past 3 years, several
major banks have failed in the U.S. Tne public has become
aware that several major banks are on the supervisory
agenci.s' list of problem banks I/ and that the number
of problem banks is increasing.
In early 1976, several congressional committees asked
us to evaluate the effectiveness of the supervisory refforts
of the three Federal agencies involved: Federal Deposit
Insurance Corporation (FDIC); Federal Reserve S;stem (FRS);
and Office of the Comptroller of the Currency (GCr%. Department of the Treasury. Specifically, the study was requested
by

the Chairmen of
--the House

Committee on Banking, Currency and

Housing;

--the Domestic Monetary Policy Subcommittee, House
Committee on Banking, Currency and Housing;
-- the Financial Institutions Supervision, Regulation
and Insurance Subcommittee, House Committee on Banking, Currency and Housing;
-- the Commerce, Consumer, and Monetary Affairs Subcommittee, House Committee on Government Operations;
and
-- the Senate Committee on Banking,
Affairs.

Housing

and

Urban

1/In the context of this report we use the term "problem
banks" to refer to banks requiring special supervisory
attention. FDIC and FRS also commonly refer to them as
problem banks, but OCC considers problem banks as a portion of banks requiring special supervisory attention.

The objective of our study was to evaluate the agencies'
efforts to (1) identify unsound conditions and violations
)f laws in banks and (2) cause bank management to take
corrective actions.
Our study was directed to determining
whether:
-- Bank examinations are of sufficient scope to identify
banks which are likely to run into serious managerial
or financial difficulties.
-- Supervisory agencies' efforts to improve their operations are satisfactory.
-- Supervisory agencies can and do follow through on
their findings of problems in banks to see that
corrective acti -3 are taken by bank managers.
-- Examiners are qualified and trained to conduct reliable bank examinations.
-- OCC considers applications for national bank charters
on a fair and consistent basis.
We reviewed examination reports and correspondence files
on over 900 banks
punervised by the 3 agencies.
These included three sample groups:
-- 30 of the 42 banks which failed from 1971 to mid-1976.
--234 of 787 problem banks as of December 31,
and December 31, 1975.

1970,

-- A general sample of 600 of the over 14,000 banks in
the United States.
The Federal bank regulatory agencies
Government involvement in the American banking industry
has consisted of recurring attempts to balance the need foi
healthy competition among banks with the need for a sound
banking system.
As history shows, these objectives are not
easily reconcilable.
Attempts to balance them nfve led
o a banking system which is unique in the cont'.~porary
world; Government involvement in tne Nation's 14,700
commercial banks is dispersed among 50 States and 3 Federal
agencies.
2

-- OCC was established in 1863 by the National Currency
Art which was superseded by the National Bank Act in
1864.
The Comptroller was authorized to charter and
supervise national banks.
-- FRS was created by the Federal Reserve Act of 1913.
The act established the Federal Reserve banks (FRBs)
to carry out monetary policy and to improve tne supervision of banking in the United States.
--FDIC was created by the Banking Act of 1933 as an
independent Government corporation, to insure small
depositors against losses resulting from bank
failures.
The Federal agencies, as well as agencies in 50 States,
all have some responsibility for Dank regulation (the process of interpreting banking legislation and issuing rules
and regulations for the banks) and bank supervision (ti.e
process of monitoring, examining, and advising individual
banks).
The Federal agencies do not examine the sante banks.
OCC examines national banks, the FRBs examine State banks
which are members of the System, and FDIC examines insured
State banks which are not members of FRS.
State banking
authorities also examine State banks.
FDIC examines about 60 percent of all commercial banks,
but these banks account for less than 23 percen. of total
deposits and the majority are smal' or medium sized.
OCC
and the FRBs examine nmost of the large bnks as well as many
small and medium sized banks.
The agencies receive no congressional appropriations,
but rely essentially on the banks they supervise and their
investments irn U.S. government securities for operating
funds.
Some appreciation for the relative size of the
agencies' operations at December 31, 1975, can be derived
from the following comparisons.
FDIC
Number of commercial banks
supervised
Number of bank examiners
1975 costs of examinations
(millions)

3

FRS

OCC

8,594
1,700

1,046
700

4,744
2,000

?C8

$22

$69

Recent trends in the banking industry
About 14,700 commercial banks are chartered to do business in the United States and its possessions.
Of these,
about two-thirds are chartered by the 50 States and onethird by OCC.
During the period 1971-75, the commercial banking industry underwent significant change. Several trends have become
apparent:
-- The number of banks with assets over $100 million
increased 44 percent, while banks with under $10
million in assets decreased 23 percent.
-- The number of banks controlled by holding companies
increased 52 percent; assets of these banks increased
83 percent.
-- Assets of foreign t anches of FRS member banks
increased 189 percent.
(Nonmember banks hold
less than 1 percent of foreign branch assets.)
-- Foreign loans of domestic banks and branches increased 122 percent.
other

From our analysis of banking industry financial data,
observations can also be made:
-- Total assets of the commercial banking industry grew
by over 50 percent.
The number of banks increased
only 6 percent.
-- Less than 19 percent of the asset growth was financed
by demand deposits (checking accounts) wl.ile 74
percent came from n3re costly time deposits (savings
accounts) and borrowings from other banks and sources
of credit.
(The remaining 7 percent came from an
increase in capital.)
-- Reserves for potential losses on loans and securities
increased 41 percent, 7 percentage points less than
the overall growth in loans and securities.
--Total capital increased 46 percent.
Approximately
half of the growth was the result of profits being
retained; half, the result of new capital being added.
-- Net losses from loans and securities for 1975 were
approximately 200 percent greater than for 1971.
4

The same 5-year period saw unfavorable trends in the
Nation's economy. As a result, banks which had placed
portions of their resources into higher risk ventures
encountered financial difficulties. This was reflected
by an increase in the number and size of "problem" banks.
Of 352 problem banks at January 1, 1971, 13 had deposits of over $100 million. At December 31, 1975, there were
607 problem banks, 90 of which had over $100 million in
deposits.
The number and size of bank failures also increased
during this period. There were more failures in 1975 than
in any year since 1943, and the 3 largest failures in history occurred between 1971 and 1975.

5

WHAT LESSONS CAN BE DRAWN
FROM RECENT BANK FAILURES?

For the first time since the massive bank failures of
the 1930s, the public is concerned over the health of the
banking industry. One direct result was the creation of the
FDIC in 1933 to protect depositors and prevent mass withdrawals. Since 1933, banks have continued to fail, but until
1965 these failures involved relatively small banks. (The
largest had deposits of $48.8 million.)
The graphs on the following page show the number of
failures i.nd the deposits of banks that closed between
January 1960 and December 1976. Although the number of
failures is still small, several larger banks have failed
since 1965.
Placing the figures in perspective, the largest numinis
ber of failures in any year shown was 16 during 1976.
represents about 0.1 percent of the total number of banks.
Although it is apparent that the economy can tolerate the
number of failures that have occurred in recent years, we
cannot determine at what point an intolerable situation
would develop. For example, if the number of large bank
failures were to increase, the economy could be seriously
affected.
We selected for detailed review 30 of the 42 banks that
were closed between January 1971 and June 19.6. An additional 27 banks merged with others to avert probable failure.
Causes of the bank failures
Adverse economic conditions in the 1970s contributed
to some of the bank failures. Banking entails risk, and
these risks become greater in periods of general economic
decline.
However, the bank examination reports show that the
primary cause of each failure was the practices followed by
the bank's managers. These practices left the banks more
vulnerable to economic fluctuations.
Among the 30 cases we reviewed, 14 banks' problems
were related to self-serving loan practices. Illegal acts
such as embezzlements caused eight of the failures. The
remaining eight banks failed because of general loan mismanagement.

6

NUMBER OF BANKS CLOSED 1960-1976
Number Of Closings
16
15
14
13
12
11
10
9
8
7
6
5
4
3

I
1960

i

61

62

I

I
63

64

55

66

67

68

69

70

71

72

I

I

I

73

74

75

73

74

76

YEAR

TOTAL DEPOSITS OF CLOSED BANKS 1960-1976

Total Deposits (millions)

700
60

300

1960

61

62

63

64

65

66

67
68
YEAR

7

6°

70

71

72

75

76

Had the 2 roblems been
identified by bank examiners?
Tie examiners identified the underlying problems which
In 21 of the 30 cases we reled to most bank failures.
viewed, the agencies identified the banks' problems at
Moreover, the examiners
least 2 years before they closed.
usually commented to bank managers on the problems in
The difficulty confronting the agenreports or meetings.
cies was not in identifying the problems but in influencing
the banks to solve them.
What did the agencies do?
First the agencies attempted to influence bank managers
and owners with informal techniques. In some cases they
made visits to the banks in addition to regular examinaSome banks were required to report periodically
t:.ons.
on progress in solving their problems. OCC even placed
examiners in one bank full time.
Although agency personnel said informal persuasive
techniques are usually sufficient to convince a bank's
managers to solve its problems, persuasion obviously didn't
work with the banks that failed. This was usually because
the bank officials fellowed self-serving loan practices
or were incompetent, as stated in examination reports and
In addition, the banks' boards of direccorrespondence.
tors did not meet their responsibilities.
Faced with this situation, the agencies could have
turned to their legal powers. However, we noted a tendency
by each agency to delay legal action until the banks' problems had become so severe as to be difficult at best to
correct. The regulators kept waiting for the banKs to take
the actions they had promised, and bank managers broke
those promises several times before the agencies began
legal steps.
In 7 of the 30 cases reviewed, the agencies threatened
to either close the banks or terminate deposit insurance.
FDIC actually began to terminate the insurance of four banks
we reviewed.
However,
The agency removed a bank officer of one bank.
he owned a controlling interest in the bank and remained
influential.

Judging the appropriate time to take formal measures
against a bank's management is difficult. Nevertheless,
we believe the supervisory agencies did not use their cease
and desist authority as effectively as they might have.

9

ARE FEDERAL BANK EXAMINATIONS OF ADEQUATE SCOPE
TO IDENTIFY SIGNIFICANT PROBLEMS IN BANKS?
Basic approach
Bank examinations have placed great emphasis on analyzing the bank's condition at the time of the examination. This approach has been reasonably effective in
However, in many cases
identifying problems in banks.
examiners do not address the underlying causes such as the
bank's basic management practices, operations, and controls.
The examination procedures followed by the agencies
They looked at the same things and did
were much alike.
The major emthe same kinds of analyses and evaluations.
phasis of the agencies' examination efforts was on evaluating quality of assets, adequacy of capital, and quality of
management. The examination approaches have emphasized
financial ratios and comparisons. The agencies had
not established criteria or acceptable levels for these
Their views of the condition
ratios and comparisons.
of banks depended largely on individual judgment.
At each agency the scope of examination was established
by the examiner-in-charge within general guidelines provided
by agency manuals, standard report formats, and agency
training and tradition.
The manner in which a bank was operated and controlled-its policies, procedures. and practices; its internal control system: and its internal and external audit functions-was not normally a leading determinant of examination scope.
Thus, the same things were usually looked at from bank
to bank.
Examinations should focus more o-i bank operations that
Also,
have weaknesses which could cause serious problems.
have
if
they
procedures
certain
examiners should waive
been satisfactorily performed by internal auditors, outside
auditors, or State examiners.

10

New approach beingdeveloped
We believe the new e.amination approach, development
oL which was initiated by OC r' in the fall of 1975, will
help examiners to focus on and deal with potential causes
of problems before the financial soundness of the banks
are affected. Under the new approach, examiners will
continue to evaluate the financial condition of banks
and test for compliance with laws, rules, and regulations.
The major change is the depth to which examine:s will
probe into the workings of banks to determine how well
they are being managed on a day-to-day basis.
Many bank problems, such as bad loans and poorquality investments, are the result of weak policies,
poor procedures or practices, or lack of sound control
mechanisms. By concentrating on bank policies, procedures,
practices, and control mechanisms and by requiring banks
to strengthen thcse areas when weak, the agencies can
gain greater assurance that a bank's financial condition
will nst deteriorate significantly between examinations.
Our review of the new procedures and the results of
10 test examina.tions. made by OCC in 1976, lead us to conclude that the concepts underlying the new OCC approach
are sound and that the new approach offers substantial
benefits over the current examination approach. The new
OCC approach has not been tested on very small banks,
very large banks, or banks known to have significant
problems, nor have the costs of the new approach been
estimated. Additional testing of the new approach
is needed to make sure that it will apply to all classes
of banks. Improvements or adjustments in the approach
may be appropriate after further tests, particularly
if the costs will be much greater than under the old
approach.
In our opinion, the concepts could also be applied
by FDIC and FES. The three agencies should jointly
evaluate the approaches being developed.

11

Problems found

in examinations

Examiners found some type of problem in nearly
The most frequently
all of the banks in our samples.
found problems were similar among the banks in our
general sample an.d our problem bank sample. The degree
of severity between problem banks and ba.ks in general
was dramatically different for problems related to
loan concentrations, liquidity, loan policy, and capital
adequacy.
Banks on the problem list were often cited for inadequate liquidity, inadequate collateral documentation,
ineffective management, or excessive insider loans in
addition to other problems.
Banks of different sizes had different problems.
Large banks were more often criticized for the character
of their business (classified loans, inadequate capital)
whereas smaller banks were more often criticized for
problems related to procedures and operations (inadequate
credit files, poor collection procedures).
For banks in general, FDIC examiners clearly cited
banks for problems more often than either FRS or OCC
The same was true for problem banks, but
examiners.
differences among agencies were less pronounced.
The relationship between the frequency with which
banks were cited for problems related to management effectiveness--such as inadequate internal routines and controls and violations of laws and regulations--and the
frequency with which management effectiveness was criticized was not what might have been expected. Far fewer
banks were cited for ineffective management than for
the related problems with inadequate internal routines
and controls and violations of laws and regulations:
-- Four percent of the banks in our general sample were
cited for ineffective management; 55 percent for
violations of laws and regulations; and 44 percent,
for inadequate routines and controls.
--Sixteen percent of the banks in our problem bank
sample were cited for ineffective management;
81 percent, for violations of laws and regulations;
and 55 percent, for inadequate routines and
controls.
12

The agencies rarely criticized a bank's loan policies
until loan problems developed.
Insider and out-of-territory
lending were not frequently mentioned problems for banks
in our samples.
Frequency of examinations
The National Bank Act requires that each national bank
be examined twice each year, but allows the Comptroller
to waive one examination in each 2 -year period.
During
the 1974-75 cycle, OCC examined 75 percent of the national
banks the required 3 times.
FRS policy is to examine
each State member bank at least once a year.
In 1975
it examined 97.5 percent of its banks.
FDIC, in practice,
attempts to examine each of its banks once every 12 months.
In 1975 it examined 85 percent of its banks.
In our view, the number of times a bank is examined
should not be based upon a rigid frequency requirement.
Rather, the agencies, using the results of previous
examinations and information from reports by banks, should
schedule examinations ' -ed on an evaluation of bank's soundness, and the quality oL its policies, procedures, practices,
controls, audit, and management.
Under this approach, banks in poor condition would be
examined more often than those in good condition. Each
agency should have policies to allow it to consider such
factors and exercise discretion in determining whien to
examine banks.
Scheduling of examinations
Agency officials said they tried to preserve the element
of surprise in scheduling examinations so banks would not
hide adverse conditions or wrongdoing. Therefore, as a
matter of policy, they did not disclose examination schedules
to banks or outsiders (other than State examiners) and
they tried to avoid establishing predictable patterns.
However, in some cases the agencies had established
definite examination patterns.
FRS examined 70, FDIC
examined 56, and OCC examined 79 of the banks in our samples
in the same month of 2 or 3 consecutive years.
Because the
agencies view surprise as an important element of an examination, they should be scheduling their examinations to
avoid obvious patterns.

13

Relationshipof Federal

and State examinations
FDIC and rRS sometimes conducted their examinations
at the same time as the State banking agencies. Both
agencies have started very limited experimental programs
to rely more on the work of State examiners instead of
examining banks in those States.
In our opinion, thes9 approaches are reasonable
If
needless duplication of work.
attempts to elimindte
cdIbe expanded to enable
found acceptatle, they sFDIC and FRS to conceal

r

t'heir efforts more on banks

be relying more on State examinawith serious problems.
tions, FDIC and FRS could free their own examiners
from relatively routine examinations of "good" banks
to examine, reexamine, visit, or monitor banks with
major problems. The Federal agencies, of course, should
rely on the States' examinations only if they are of acceptable quality.

Consumer protection laws
The Congress has enacted several laws to protect
consumers. Certain of these laws affect banks' lending
practices, efforts to attract depositors, and billing
discrepancies. The agencies are responsible for enforcing
these provisions.
In the most recent examinations of banks in our general
sample, FDIC examiners cited violations of consumer
credit and truth in lending regulations more often
than FRS and OCC examiners.
Percent of banks in which
violations were found_ by
FDIC

Regulation
Consumer credit (regulation B)
Truth in f-nding (regulation Z)

14

3
29

FRS
1
17

OCC
14

The agencies were not devoting enough attention to
monitoring banks' compliance with consumer protection laws
and regulations.
Their procedures were not sufficiertly
comprehensive or detailed.
Thus, the agencies relied
heavily upon the individual examiners to find violations;
although the examiners were insufficiently trained.
While the agencies reported some violations of
consumer protection laws and regulations, they
acknowledged that they have not aggressively
monitored consumer protection law compliance, and they have
begun revising their approaches.
They have started new
programs to improve their approaches, including more comprehensive procedures, specialized training, and specialized
examination staffs.
International operations
FRS and OCC are the primary examiners of international
operations because few FDIC-examined banks are internationally involved.
International examinations are similar
to commercial eY¥.ninations, in that loan quality, controls,
and management are evaluated.
However, these examinations
are complicated becauis
special tisks are involved
in foreign loans and foreign currency trading and because
the operations are conducted in foreign countries.
FRS examines the international operations of State
member banks and all Edge Act corporations 1/,
even when the parent bank is a national or nonmember
bank.
OCC examines the international operations of national
banks.
FRS and OCC conduct international examinations at
the parent bank's home office, the Edge Act corporation's
home office, and the foreign branch or subsidiary.
The
examiners usually assessed the quality of foreign loans by
using information at the home offices.

1/An Edge Act corporation is a domestic business chartered
by the FRS solely for the purpose of conducting banking in
foreign countries.
15

Our review of examination reports for 18 national
banks and 12 State member banks with substantial international operations revealed 2 cases where State member
banks were experiencing problems, some of which were related
to subsidiaries of the bank's Edge Act corporations.
Before the problems were noticed in the banks, FRS
examiners had stated that the credit information on the
foreign activities available at the home office was
inadequate.
The subsidiaries were not examined onsite
until after the banks had begun experiencing problems.
Early onsite examinations of the subsidiaries might have
disclosed their problems before parent banks were injured.
Foreign loans are more complicated to evaluate than
donestic loans, because they are o-.cen to foreign governmen-s and in different currencies. A special risk (called
country risk) is taken with loans made in different currencies, because the borrower may not be able to obtain
the currency borrowed which is needed to repay the loan.
FRS and OCC took different approaches to evaluating
loans subject to country risk.
OCC and one FRB used a
committee approach to evaluating country risk; the other
FRBs relied upon individual examiners to evaluate this
risk.
This difference has caused some banks' loans to a
country or a foreign business to be classified differently
than other banks' loans to the same country or business.
The method used by most Federal Reserve banks requires
individual examiners to keep abreast of economic conditions
in many countries and to judge loans in many countries.
A
team of experts who evaluate economic conditions in each
country would produce more accurate and consistent results
than numerous individuals evaluating loans case by caseRecommendations
We recommend that:
--FDIC and FRS establish procedures to base the scope
of each examination on the examiner's evaluation
of the quality of the bank's controls, policies,
procedures, and audit.
(See FDIC comments in
app. III, p. III-19 and FRS comments in app. II,
p. II-12.)

16

--FDIC and FRS develop standards for preparing, maintaining, and using examination wDrkpapers.
(See
FDIC comments in app. III, p. III-20 and FRS comments
in app. II, pp. II-13.
-- OCC invite FDIC and FRS to jointly evaluate its
new examination approach and, in the event of a
favorable assessment of the new process, FDIC and
FRS revise their examination processes to incorporate the concepts of OCC's approach.
(See FDIC
comments in app. III, p. III-39, FRS comments in
app. II, p. II-21, and OCC comments in app. I,

p. I-8.)

--FDIC and FRS adopt flexible policies for examination
frequency, which would allow them to concentrate
their efforts on banks with known serious problems.
(See FDIC comments in app. III, p. III-16, FRS commants in app. II, p. II-11, and OCC comments in
app. I, p. I-6.)
--The Congress amend the National Bank Act to allow
OCC to examine national banks at its discretion.
We would be glad to assist the committees in drafting appropriate legislation.
(See OCC comments
in app. I, p. 1-6.)
·--FDIC, FRS, and OCC establish examination scheduling
policies and procedures to avoid setting patterns.
(See FDIC comments in app. III, p. III-15, FRS
comments in app. II, p. II-11, and OCC comments in
app. I, p. I-6.)
--FDIC and FRS develop minimum standards for examiner
training and examination procedures and use reports
of State examinations meeting those standards.
(See
FDIC comments in app. III, p. III-18 and FRS comments
in app. II, p. II-12.)
--FRS and OCC, using all -vailable information, develop
and use a single approach to classify loans subject
to country risk.
(See FRS comments in app. II,
p. II-13 and OCC comments in app. I, p. I-7.'
-

FRS and OCC implement procedures to examine major
toreign branches and subsidiaries, including subsidiaries of Edge Act corporations, periodically and
whenever adequate information about their activities
17

is unavailable at the home office.
(See FRS comments
in app. II, p. II-14 and OCC comments in app. I,
p. I-7.)
-- FRS and OCC utilize each others examiners to cut
expenses when conducting examinations in foreign
countries.
(See FRS comments in app. II, p. II-15
and OCC comments in app. I, p. I-7.)

18

ONCE PROBLEMS ARE IDENTIFIED,
ARE THEY COMMUNICATED EFFECTIVELY TO BANKS?
The agencies prepared reports of .xamination which
were sent to the examined banks' boarcs of directors.
Each agency organizes its report differently, but they all
contain the same basic information.
Examiners described the problems identified during
the examination in a summary which was given to the
bank.
Detailed schedules, analyses, and listings contained
in the "body" of the report ,ere also given to the bank.
This section
-- supported
-- documented

the examiners'
some of their

-- communicated some of
to agncy officials.

criticisms and conclusions.
work

and

the bank's financial data

Examination reports also had "confidential" sections
which were not usually given to the banks. Here the examiner expressed his or her opinions regarding the bank's management quality and financial condition. This section allows
examiners to comment freely to agency management about
a specific bank.
Commu,icating with banks' directors
The success of the supervisory process depends heavily
on how results are disclosed to those responsible for
cor ecting problems--the bank's board of directors.
Although specific duties, responsibilities, and liabilities vary from State to State, generally bank directors
are required to be fully aware of the bank's policies,
operations, and condition. They are supposed to apply
ordinary care and prudence in administering the bank's
affairs, and they may be liable for any resulting ')sses
if they do not.
Thus, the results of an examination
should be important to the board of directors, and
the supervisory agencies should be doing their utmost
to communicate the examination results to the directors.
The agencies did not require their examiners or
regional officials to meet with the banks' boards of
diretors after all examinations. As shown by the
19

following chart, the agencies rarely met with the boards of
directors of banks in our gei-.ral sample and very often did
not meet with those of banks with major problems.

Agenc_

Percent of banks in which
agencies met with directors
ProblemE-anis
Ban s in general

FDIC
FRS
OCC

1
9
6

30
53
54

We believe that the agencies should discuss the
results of their examinations with the boards of directors
or their audit or examining committees after each examination irrespective of the nature of their findings to
-- emphasize the imrportance of examinations,
-- insure that the direetors, who are ultimately
responsible for the bank's operations, are fully
aware of the examination results,
-- discuss findings,
-- establish closer working relationships with the
boards, and
-- enhance the stature of the examiners.
FDIC and FRS have a general policy of meeting with boards
Officials of one FRB,
of directors of all problem banks.
however, had a policy of meeting with the board of directors
of each examined bank and believed that this practice reduced
In January 1976, the
the incidence of serious problems.
OCC implemented a policy of meeting with the boards of directors of all national banks each year.
Communicating needed actions
The examiners generally did not recommend how the
For 63 percent of
banks could correct the problems.
the problems noted in our problem sample banks, the
examiners did lot recommend corrective actions. In some
instances, the required corrective action would have
been obvious to the bank.
20

Although the body section was given to the bank, it
contained information, such as the balance sheet and income
and expense data, which the bank had furnished to the examiners.
The confidential sections of problem bank reports
we reviewed contained criticisms which were not noted
The examiners criticized bank
elsewhere in the reports.
the comments section, as shown
in
than
more
here
management
table.
following
by the

Problem or criticism
Inadequate or incompetent
management
Inadequate capital
Insufficient liquidity
Earnings (excessive or
improper)

Percent -f reports
in which proble,is were noted in
Examiner
Confidential
comment section
section
FRS OCC
FDIC
FRS OCC
FDIC

63
43
19

45
63
35

60
68
54

26
61
37

10
60
30

10
38
50

11

25

26

17

18

8

We believe that the report of examination should present clearly and concisely the results of the examination,
the agency's recommendations for corrective action, and
information necessary to support the examiners' conclusions.
It need not go to great length to provide information which
If the agency needs additional inforthe bank already has.
purposes, the report sent
statistical
or
mation for review
by a detailed, strucaccompanied
be
should
to headquarters
collection forms.
data
standard
and
workpapers
tured set of
report containing
a
by
burdened
be
not
would
bank
the
Thus,
be better
would
agencies
the
and
information
superfluous
work.
examiners'
the
re~view
to
able
The examination report on electronic data processing
(EDP), like the commercial examination report, should
state only the deficiencies noted by the examiner and any
necessary supporting information.
OCC is now changing its examination report to communicate information more effectively both to OCC management and
to the banks.
21

Recommendations
We

recommend

that FDIC and

FRS:

-- Require examiners to meet with the bank's board
of directors or audit or examining committee
after each examination.
(See FDIC comments in
app. III, p. III-27 and FRS comments in app. II,
p. II-19.)
-- Develop and use reports of examination which
provide the banks with the results of the examination and any necessary supporting information.
(See FDIC comments in app. III, p. III-30 and
FRS comments in app. II, p. II-20.)
-- Develop reports of EDP examinations which present
the problems found, corrective action needed
and any necessary explanatory data in a clear
and concise manner.
(See FDIC comments in
app. III, p. III-21 and FRS comments in app. II,
p. II-16.)

22

HAVE THE AGENCIES BEEN EFFECTIVE IN
GETTING BANKS TO TAKE CORRECTIVE ACTION?
Examiners find some type of problem in virtually all
banks; however, some banks have more serious problems and
require more supervisory attention than others. The agencies
cannot correct the banks' problems themselves, but they can
use many enforcement tools to get banks to correct their
These tools include both informal (persuasive)
problems.
and formal (generally legal) enforcement actions.
Our analysis of enforcement actions taken by the superviscry agencies for almost 900 banks in our samples showed
that informal actions were used most of the time and formal

actions were seldom used.
Are informal actions effective?
Informal enforcement actions are nothing more than an
agency's attempt to persuade bank managers to take correcThe supervisory agencies' success in getting
tive action.
bank problems corrected depends heavily on cooperation from
bank officials in changing the practices and policies which
caused the problems.
To assess the supervisory agencies' effectiveness in
getting banks to solve their problems, we analyzed the examination reports on the agencies' problem banks for the
5-year period ending December 31, 1975.
OCC's criteria for identifying problem banks varied
during the period, and it could not fu'.ly identify for us
which banks had been considered problem banks. Its rating
system included an overall composite rating of the bank
ranging from "1" to "4" -- "1" being the best. Officials
said that until December 1974, problem banks were mostly
We considered
those with composite ratings of "3" or "4."
"3" or "4" as
of
ratings
composite
those national banks with
OCC problem banks.
During that period, 718 State nonmember banks, 128
State member banks, and 686 national banks had at some time
been in the agencies' problem bank category. Of those
banks, 414 (58 percent) of the State nonmember baniks, 38
(30 percent) of the State member banks, and 392 '5? percent)
of the national banks were removed from problem status by
December 31, 1975. Although most returned to nonproblem
status within 2 years, 19 percent were problem banks from
2 to 5 years and 5 percent were problem banks for over
5 years.
22a

We reviewed a sample of 149 banks which were on the
supervisory agencies' problem lists at December 31, 1970.
As of December 31, 1975, the status of the 149 banks was
as follows:
FDIC

FRS

Number Percent

OCC

Number Percent

Number Percent

Removed from
problem bank
list (105)
44
Converted to
a national or
State
charter (5)
Withdrawn from
FRS member
ship

(7)

80

15

38

46

84

-

1

3

4

7

-

-

7

18

-

-

6

15

1

2

2

4

1

3

1

2

9

16

9

23

3

5

39

100

55

100

-

Merged with
another
bank

(7)

Failed (4)
Remained on
problem
list (21)
Total

55

10o'

OCC and FDIC had the most success using informal action,
returning 84 and 80 percent respectively of their problem
banks to nonproblem status, while FRS was successful with
only 38 percent of its problem banks.
For the 105 banks which had been taken off the problem
lists we determined the length of time spent in problem
status:
FDIC

Years
Under
1 to
2 to
3 to
4 to
5

FRS

Number Percent

OCC

Number Percent

Number
Percent

1
2
3
4
5

6
13
9
7
4

14
30
20
16
9

1
1
2
4
1

7
7
13
26
7

to 10

12
18
5
2
3

5

26
39
11
4
7

11

_6

40

6

13

Total

44

100

15

'00

46

100

23

OCC got problems resolved most promptly.
About 65
percent of its banks which returned to nonproblem status
did so within 2 years.
Only 44 percent of the FDIC banks
and 14 percent of the FRS banks which returned to nonproblem status did so within 2 years.
We are concerned by banks which are on problem lists
for long periods of time.
We believe that the supervisory
agencies should have used formal enforcement actions more
frequently when dealing with these banks.
How often have the supervisoryagencies
used their formal enforcement powers?
Formal enforcement actions are threatened or initiated
by the supervisory agencies as a last resort for getting
problems corrected.
Some formal enforcement actions do not
help correct problems but result in closing the bank to proteut depositors.
Written agreements
The supervisory agencies use formal writtem agreements,
sometimes referred to as voluntary agreements or letter
agreements, to confirm a bank's plans to correct problems.
The acency and the bank both sign the agreement.
A violated
agreemrent can be the basis for issuing a cease and desist
order against the bank.

1971

The agencies use of written agreements for the period
through 1976, was as follows:

Year

FDIC

1971
1972
1973
1974
1975
1976
Total

FRS
(note a)

OCC

Total

1
1
1
-

J
3
2
2
1

3
4
6
17
18
23

5
5
10
19
20
24

3

9

71

83

a/Does not include 12 agreements against bank holding
companies.
24

Cease and desist orders
The supervisory agencies, under authority of the Financial Institutions Supervisory Act of 1966, can issue cease
and desist orders against banks to get problems corrected.
First, a notice of charges is served upon a bank
-- which has engaged or
unsound practices,

is engaging in unsafe or

-- which has violated or is violating a law, a rule,
a regulation or a written agreement with the
agencies, or any condition imposed in writing by
the agencies in connection with the granting of
any application or other request, or
-- which is about to do either.
The notice of charges presents a statement of facts constituting the alleged violations or unsound practices and
establishes a time and a place for a hearing to determine
whether a cease and desist order should be issued.
If the bank representatives do not appear at the hearing or if the hearing confirms the violation or the unsafe
or unsound practices, the agencies may issue the cease and
desist order.
A bank can consent to the cease and desist order,
obviating a hearing.
The order remains in effect until
stayed, modified, terminated, or set aside by the agency
or
a reviewing court.
The agencies use of cease and desist orders for the
period 1971 through 1976, was as follows:
Year

FDIC

FRS
(notea)

1971
1972
1973
1974
1975

7
10
9
4
8

1
3
1

1976

2
4
2
5

29

8
15
13
6
14

4

7

40

67

9

20

96

Total
a/Does not

OCC

Total

include 12 orders against bank holding companies.
25

Removal of management
Also under the Financial Institutions Supervisory Act
of 1966, FDIC and FRS may order the removal of a director
or officer of a State bank which they supervise and OCC may
recommend that FRS remove one from a national bank when
-- the director or officer has violated a law, a
rule, a regulation, or a final cease and desist
order; has participated in any unsafe or unsound
banking practice; or has committed or engaged in any
act, omission, or practice which constitutes a
breach of his fiduciary duty and
-- as a result, the bank has suffered or will probably
suffer substantial financial loss or other damage,
or the interests of its depositors could be
seriously prejudiced and
-- the violation, practice, or breach involved personal
dishonesty on the part of the director or officer.
The agency must first serve the director or officer
with a written notice of its intention to remove him/her
from office. The notice of intention states the grounds
for removal and establishes a time and a place for a hearing. As with a cease and desist order, the agency can
remove the director or officer if he fails to appear at
the hearing or if the charges specified in the notice of
intention are substantiated. The removal order, too,
remains in effect until stayed, modified, terminated, or
set aside by the agency or a reviewing court.
In addition, the agencies have the authority to suspend
any bank director or officer indicted for a felony involving dishonesty or breach of trust.
The statute orovides
that such a suspension can be enforced by written notice
and remains in effect until the charges are disposed of
or the suspension is terminated by the agency. On August
13, 1976, the district court of the District of Columbia
held that this statute violates the due process clause
of the Constitution. FDIC officials said they are working
with the other agencies to prepare and issue regulations
in an effort to comply with the due process requirements.
During 1971-76, the agencies took action to remove or
suspend management as follows:

26

Year

FDIC

1971
1972
1973
1974
1975
1976

FRS

3
3
4
6
3
Total

Note:

1
2
-3

19

Includes 15

OCC

4
FDIC, 2 FRS,

Total

3
8
3
9
3

7
11
7
17
7

26

49

and 25 OCC

suspensions.

Financial assistance
FDIC has the authority to provide funds to insured banks
in danger o.f closing which are essential for providing banking services to its community or to assist a merger, or sale
of assets and assumption of liabilities of a failing or failed
bank into or by another insured bank.
In providing financial
assistance, FDIC can require that bank managers correct their
problems.
Such assistance may
-- maKing deposits
-- purchasing

in

assets

include
the troubled
of the failing

bank,
or

failed

bank,

--grant.ing a loan secured by the assets of the failing
or failed bank, or
-- guaranteeing another insured bank against loss
in assuming the assets and liabilities of the
troubled bank.
FRS has the authority to loan funds to member banks
--to enable them to adjust their asset positions because
of developments such as a sudden withdrawal of deposits
or seasonal requirements for credit which
can not reasonably be supplied front the banks' own
resources or
---to assist them in meeting unusual situations which
may result from national, regional, or local
difficulties.
27

Cancellation of deposit insurance
FDiC has the authority to terminate a bank's deposit
insurance if:
-- its officers or directors are engaging
or unsound banking practices,
-- it

in unsafe

is in an unsafe or unsound condition, or

-- it has violated an applicable law, rule, regulation,
or order; a condition imposed in writing; or a written agreement with FDIC.
When FDIC initiates proceedings to terminate insurance,
it may give that bank a maximum of 120 days to correct its
problems.
If the bank corrects all or some of its problems
within the time allowed, FDIC may drop the termination proceedings altogether or take other action, for example a
cease and desist order.
If the bank does not correct its
problems, FDIC's Board of Directors can terminate its
insurance.
During 1971-76, FDIC
as follows:

initiated termination proceedings

Year

Number of
proceedings

1971
1972
1973
1974
1975
1976

5
5
1
3
5
8
Total

27

only 1 of the 27 proceedings, in 1976, resulted in termination. Before 1971, FDIC terminated the insurance of 13
banks. Canceling a bank's deposit insurance does not solve
its problems.
Cancellation of FRS membership
FRS has the authority to cancel a bank's membership in
the Federal Reserve. As far as we could determine, FRS has
used this authority only once as a corrective tool.
As with
terminating deposit insurance, this action does not solve a
ban,:'s problems.
28

Revocation of charter
OCC has the authority to revoke national bank charters
and States have the authority to revoke State bank charters,
In the last 2 decades,
although this too solves no problems.
as far as we could determine, OCC has not revoked a bank
charter for not correcting its problems.
Once formal actions were taken

did the banks correct their _roblems?
The supervisory agencies used cease and desist orders
and/or suspension of managers 18 times against 17 banks in
As of November 30, 1976, eight of these
our problem sample.
banks had been removed from the problem lists by the agenAlthough formal actions were taken against relatively
cies.
few of the banks in our samples, it appears that, on thes
average, the sooner the action was taken the sooner the
We recognize, howbank was removed fr3m the problem list.
eve., that some problems will take a long time to correct
and formal actions will not always solve a bank's problems
immediately.
Identifying banks for
special supervisory attention
After an examination, the agencies evaluate the effect
that the problems identified can have on the bank's soundIf the problems are serious, the bank is designated
ness.
Such banks are sometimes
for special supervisory attention.
referred to as problem banks.
Because the agencies use different criteria to identify
problem banks, they often do not agree on which banks reOf the 4,744 national 'anks
quire special supervision.
operating on December 31, 1975, OCC considered 85 as requirAmong the
ing special supervision, FRS 267, and FDIC 52.
1,046 State member banks, FRS identified 65 problem banks
FDIC's rating of national and State
and FDIC identified 17.
member banks, however, is based on financial. risk to the inWe believe there should be some consistency
surance fund.
among the supervisory agencies in determining whether or not
a bank is a problem bank.
Because the agencies' supervisory responsibilities for
banks overlap, and because their interest should intensify
as serious problems are identified, we believe they should
work towards a common definition of banks requiring close
supervisory attention.
29

Recommendations
We recommend that FDIC, FRS and OCC
-- establish more aggressive policies for using formal
actions including written guidelines to identify the
types and magnitude of problems that formal actions
could appropriately correct, (see FDIC comments in
app. III, p. III-48, FRS comments 'n app. II,
p. II-23, and OCC comments in app. I, p. I-10
and 11.) and
-- develop uniform criteria for identifying problem
banks.
(See FDIC comments in app. III, p. III-50,
FRS comments in app. II, p. II-25, and OCC comments
in app. I, p. I-11.)

30

WHAT OBSERVATIONS DID GAO
HAVE ON FRS SURVEILLANCE
OF BANK HOLDING COMPANIES?
Bank holding companies are those which own or control
They are a major element in the American
one or more banks.
banking system, owning or controlling one-fourth of all commercial banks in America which control two-thiuds of all
assets and deposits.
A holding company may be a source of financial and
managerial strength to its affiliated bank or banks, or
In 1956 the Congress passed
it may be a source of weakness.
tne Bank Holding Company Act to control the concentration
of financial resources, and preserve effective competition.
FRS %,as assigned responsibility for supervising and regulating bank holding companies. 1/
The agencies' examiners were expected to review banks'
relationships with their affiliates, including bank holding
companies and to criticize any relationship which could cause
or was causing problems for the bank.
Examiners said that 72 of the 344 banks in our samples
which were affiliated with holding companies had problems
According to the examinaresulting from that affiliation.
tion reports for 50 of these banks, holding company manageHowever,
ment was not the primary cause of the problems.
for the remaining 22 banks, 20 holding companies' actions
According to the examination
were causing the problems.
reports for these 22 banks, problems were caused by inept
and ineffective holding company management--particularly
overexpansion, unsound operations of nonbank subsidiaries,
and real estate loans which were unpaid.
Bank holding companies are supervised by FRBs, with the
Division of Banking Supervision and Regulation providing
general policy guidance and oversight from FRS headquarters.
In 1972, FRS developed a surveillance system to identify and monitor actual and potential problems by gathering
The system includes
and analyzing information.

1/As agreed with the Board of Governors of the FRS, we conFRS actions
fined our evaluation of holding companies to
with regard to holding companies affiliated with banks in
our samples.

31

-- reviews of examination reports on holding-companyaffiliated banks, whether national, State member, or
State nonmember,
-- reviews of holding companies' registration stateraents, annual reports, applications, and other
financial information, and
-- visits to holding cor.')anies to review records and
operations.
The aim is to insure that bank holding companies are operated
in a manner that does not jeopardize subsidiary banks.
FRS inspection guidelines state that the frequency and
scope of holding company inspections should depend not only
on the holding company's size and complexity but also on
information gained from other sources, such as registration statements, annual reports, and particularly examination reports on the company's subsidiary banks.
According to responsible FRB officials, of the 12
Reserve banks:
-- 9 have no written guidelines detailing the scope
of inspections.
--5 do not evaluate nonbank subsidiaries' assets and
3 perform limited evaluations.
--4 do not meet with holding company board of
directors to discuss findings.
--2 do not submit inspection reports to either holding
compa'iy managers or directors.
--7 restrict supervisory activities, including inspections, due to budgetary restraints which preclude
hiring additional personnel.
FRS did not detect weaknesses in 15 of the 20 holding
companies until after they had damaged subsidary banks.
Problems in the 20 holding companies were first identified
by
-- examinations of subsidiary banks of 15 companies,
32

-- the review of financial data of 1 company,
-- 2 simultaneous bank examinations and holding
company inspections, and
-- inspection of 2 holding companies.
Nine of the 20 holding companies had not been inspected before problems appeared in their banking subsidiaries. Seven
holding companies had been inspected before problems were
found in the banks, but these inspections did not discover
the potential for problems. Four of the seven had last been
inspected 1 to 2 years before the problems were identified
in the banks. In one case the inspection was confined to
a review of the holding company's financial data. The remaining two inspections occurred less than 3 months before the
bank examinations that identified the problems.
The Division of Banking Supervision and Regulation
received data from the FRBs on specific holding companies,
and Division personnel were in frequent contact with FRB
employees. However, Division employees did not completely
monitor FRB supervisory activities. For instance, they
had no system, such as status reports, to keep track of the
number of holding companies inspected and to insure that all
holding companies with closely monitored subsidiary banks or
leveraging nonbanking subsidiaries had been inspected.
The agencies exchanged information on holding company
matters at both headquarters ani regional levels.
FDIC
and OCC provided the FRBs with copies of their examination
reports on banks affiliated with holding companies.
The FRBs
gave their holding company inspection reports to FDIC and
OCC.
The agencies did not normally conduct simultaneous holding company inspections and bank examinations, even when the
main subsidiary of the holding company was a national or
State nonmember bank.

Recommendations
We recommend that FRS implement a system of supervision
based on onsite Inspections of holding companies and their
major nonbanking subsidiaries. We also recommend that the
Board of Governors strengthen its oversight of Reserve banks'
holding company supervision by establishing
33

-- a systemw;ie manual of inspection procedures,
--a standard inspection report, and
-- periodic onsite evaluations of Reserve bank
(FRS comments in app. II,
supervisory activities.
p. II-17.)

34

WHAT DID GAO FIND REGARDING EXAMINERS'
COMPETENCE AND OBJECTIVITY?
Bankers'_ofinions
In responding to our questionnaire (see p. 53),
commercial bank officials generally reported favorably
on the
competency of examiners.
Senior exam ners' understanding
of the specialized examination areas of trust and
international was rated adequate or more than adequate
by 89 percent
of the bankers while 11 percent thought it was
borderline
or less.
Toward examinations of electric data processing,
however, bankers were less favorable. Here, approximately
25 percent thought senior examiners' understanding
borderline or inadequate. Opinions concerning each was
of the
three agencies were similar.
Personnel policies
The agencies are not legally subject to Civil
Service
Commission (CSC) rules and regulations governing
Federal
personnel practices; however, FDIC follows them
in recruiting, compensating, and promoting examiners.
OCC
uses CSC's
General Schedule in paying its examiners.
For the most
part, Federal Reserve district banks set their
own personnel
policies.
Although the agencies' personnel policies are similar
in many respects, there are important differences.
FDIC
and OCC are more centralized than FRS; therefore,
they have
more uniform policies and practices.
Each Federal Reserve
district bank has primary responsibility for recruiting,
training, evaluating, and paying examiners, and
as might be
expected, policies and practices vary considerably.
Source of examiners
Most of the examiners hired by the three agencies
have
undergraduate degrees in business-related subjects,
and some
have worked in banks or as bank examiners.
During 1971-75,
FDIC, FRS, and OCC hired 912, 594, and 1,147
examiners,
respectively, from the following sources:
FDIC
FRS
OCC
------ (percent)------College
Commercial
Other

81
2
17

Banks

35

58
12
30

71
12
17

Examiner training
The agencies operate internal schools which instruct
examiners in various aspects of bank examination,
such as
commercial banking, trusts, international banking,
and
electronic data processing.
Since these schools cover
generally the same topics, the agencies could (1)
realize
economies by consolidating their schools and (2)
assure high
quality instruction by exchanging information and
standardizing curriculums.
Bank examiners we questioned generally rated the
internal courses as useful or very useful; however,
many thought
they needed additional training, particularly in
law, EDP,
and accounting.
In the specialized areas of EDP and international
banking operations, FRS has not provided much training
in recent
years.
Its EDP school was not held in 1975 or 1976 though
plans have been formulated for a school in 1977.
Its international school was held once in 1972, 1974 and
1976.
FDIC
offers three EDP schools which are available to
examiners at
various stages of their careers. It does not have
an international school; officials said that the banks
supervised
by FDIC tend to be small and are therefore unlikely
to be
engaged in international banking.
FDIC uses OCC's schools
or instructors to provide international training
examiners need it. OCC annually operates one EDP when its
and three
international schools for its examiners.
FDIC's training program seems to be providing
its examiners with most of the skills needed to assure
high-quality
supervision of banks.
OCC has recognized problems with
its program and has acted to improve it.
Although the
Federal Reserve Board has improved its program
as a result
of a recent FRS study, we do not believe that
training can
receive enough attention as a part-time responsibility
of
the Board's Division of Banking Supervision and
Regulation.
Additional training in subjects such as EDP, law,
and
accounting would be useful for examiners in the
three agencies.
Evaluation and

testing of examiners

All three agencies periodically evaluate the job
performance of their bank examiners.
FDIC and OCC require
employees to complete a formal evaluation process
before
36

they can take charge of bank examinations. The process
emphasizes the skills needed to analyze a bank's management, assets, and soundness. FRS does not have such a
process.
We believe that formally evaluating examiners is
a sound practice for assuring that they have received the
necessary training and experience to make appropriate
decisions and judgments in examining banks.
Safeguards against conflict of interest
All three agencies have policies to guard against
actual or potential conflicts of interest among their
The policies generally prohibit examiners
examiners.
from owning stock in banks or bank holding companies,
from having loans or credit cards with banks that they may
be asked to examine, and from examining banks where their
relatives work.
Each agency requires examiners to file statements of
financial and personal interests when they are hired, but
We have been requested
only FRS requires annual updates.
by the Chairman, Subcommittee on Commerce, Consumer, and
Monetary Affairs of the House Committee on Government
Operations to examine, in a separate study, the financial
disclosure practices of the three agencies.
Examiner turnover
During 1973-75, the turnover rate of examiners in
Most
each of the agencies was approximately 10 percent.
examiners who leave the three agencies do so either to take
jobs with commercial b;-rks, other private firms or organizations, or other government agencies or to continue their
education.
Examiners who left for commercial banks accounted for
29, 37, and 41 percent of total departures at FDIC, FRS,
We checked at the agencies to deterand OCC respectively.
mine how many full examiners--i.e., those who can be in
charge of examinations--were hired by banks which they had
examined shortly before resigning.
At FDIC during 1974 and 1975, 19 full examiners went
to work for banks which they had examined during the year
Eight had functioned as examiner-inpreceding resignation.
At two Federal Reserve district banks that we
charge.
checked, no full examiners left during these 2 years to join
During 1974 and 1975,
banks which they had examined.
37

24 full OCC examiners were hired by banks which they
had examined in the 3 years preceding their resignation.
Nine of these had been in charge of the examinations.
Since few examiners left to work for banks they
examined, we see no threat to their objectivity as
long as the agencies continue rotating examiners-incharge among banks examined and reviewing examination
reports at regional offices and district banks.
Recommendations
We recommend that
-- FDIC, FRS, and OCC, increase their training in EDP,
law, and accounting, as desired by their examiners,
and, where feasible, combine their examiner schools
and standardize their curriculums.
(See FDIC comments in app. III, pp. III-55 and 56, FRS comments
in app. II, p. II-26 and 27, and OCC comments in
app. I, p. 1-12.)
--FRS, (1) establish a full-time training office to
operate its examiner training program and (2) carry
out the revision of examiner school curriculums which
it has recognized as needed for some time.
(See FRS
comments in app. II, p. II-27.)
-- FRS establish a formal evaluation process to measure
the competence of persons seeking examiner status.
(See FRS comments in app. II, p. II-28.)

38

WHAT PROGRESS HAVE THE AGENCIES
MADE IN THE PAST YEAR IN IMPROVING
THEIR EXAMINATION AND SUPERVISION OF BANKS?
Major improvements in several
were going on during 1976.

areas of

bank

supervision

Organizational chaines
OCC altered its organizational structure to implement
the recommendations of a 1975 Haskins & Sells study of its
operations. The changes resulted in a more functionally
oriented organizational structure, clearer lines of authority and responsibility, and more effective lines of communication. An Operations Review Office was also created to
review bank examinations and reports for compliance with
established policies and procedures.
FRS combined its bank supervision and bank holding company groups to achieve better coordination.
Bank surveillance
All three agencies are developing, or have developed,
systems for monitoring the performance of individual banks
using data regularly reported to the agencies by the banks.
OCC's National Bank Surveillance System (NBSS) maintains
data on each bank for the last 5 years and compares the
bank's performance with its peer group.
FDIC has developed a new system to monitor bank performance using ratios in certain critical areas relating to

In addition, it has
balance sheet and income-expense items.
other operational systems and is experimenting with several
other computer-based analytical techniques.
FRS has developed systems to monitor banks and bank
holding companies. In addition, several Reserve banks have
been developing their own monitoring systems.
Monitoring self-dealing and insider transactions
In late 1975, OCC issued regulations requiring national
banks to adopt certain procedures when lending to directors,
officers, employees, and their interests. Boards of directors must approve such loans and record them for examiners
to review. In 1976, FDIC issue' - - lar regulations which
applied to insured State nonmemoer banks.
39

Reviewin9_compliance with

consumer protection laws
FDIC adopted separate compliance reports and consumer
credit reviews in September 1974.
Currently, it is designating a consumer compliance examination specialist within each
region.
Compliance reports are prepared by examiners as part
of each bank examination.
FRS is Beginning to conduct separate reviews of compliance with consumer protection laws and regulations. Specialists will be trained to work concurrently with the commercial examiners.
OCC has recently begun scheduling separate compliance
reviews of national banks.
Examiners who have expressed an
interest in specializing in such reviews are assigned to
6-month tours and given 2-week training courses.
New examination procedures
FDIC adopted a policy in December 1976 permitting
examinations to be abbreviated under certain conditions.
The policy also provides that banks having supervisory or
financial problems be examined more frequently than nonproblem banks.
Likewise, FRS adopted a policy permitting its examiners
to reduce the scope of examination when a limited review
of the bank reveals no sign of deterioration. In March
1976, the Reserve Board authorized an asset-management examination for optional use by FRBs.
This examination approach
could be used for banks which have historically been well
operated and untroubled.
The approach focuses on evaluating
assets and management.
The most significant changes in the examination process.
however, are being made by the Comptroller of the Currency.
OCC is implementing a new approach to its fact-finding phase
of bank supervision.
The scope and approach to their examinations have been revised to place more emphasis on bank
policies, practices, procedures, controls, internal audit,
and external audit.
(See p. 11.)

40

Modification of examination reports
OCC's new examination approach and procedures involve
The main part
major changes to the report of examination.
of the new report is the examiners' narrative evaluation
Any needed
of the bank and discussion of problems found.
support is contained in an appendix. Although there will
still be a confidential section for internal purposes only,
its use will be restricted to matters requiring prompt
attention of OCC senior staff.
The FRS report of its recently authorized assetmanagement examination would contain information relevant
to those areas.
Effective January 1, 1977, FDIC began using an abbreviated e¥ymination which results in a shorter report. Certain schedules and &nialyses contained in the body of the
present report have been eliminated.
Meeting with banks'

boards of directors

OCC established a policy, in January 1976, of meeting
with each national bank's board of directors annually, preOCC
ferably at the completion of the bank examination.
meets with the directors whether or not the bank has
serious problems.
FDIC does not believe meeting with a bank's board of
directors is productive if the bank has no problems. Consequently, effective January I, 1977, each FDIC regional
director may decide after an examination whether a bank's
problems are serious enough t-o warrant a meeting with its
board.
Monitoring bank Eroblem fo.lowup
As part of the new examination approach being implemented during 1976, OCC has developed a computerized system for
monitoring bank problems to insure that they are resolved.
Called the action control system, it now includes only those
problems that NBSS identifies by analyzing regularly reported
data, but headquarters officials said the system will be
expanded to monitor actions taken to correct deficiencies
disclosed in examination reports.

41

Use of formal enforcementpgwers
The supervisory agencies have not used their formal
powers soon enough or often enough in dealing with problem
banks.
However, they have begun to use them more in the
last year.
The agencies' new attitude was exemplified in a November 11, 1976, speech by FDIC Board Chairman Robert Barnett.
He said that FDIC has reached the conclusion that formal
methods must be used more.
Accordingly, FDIC issued 29
cease and desist orders in 1976.
Only seven orders were
issued in 1975.
OCC issued seven orders as compared with
five in 1975 and FRS issued four orders as compared with
one in 1975.
In addition, OCC increased its use of written
agreements in 1976, issuing 23 compared to 13 in 1975.
Experiments in rejlying on
State examinations
During 1974-6, FDIC conducted experimental programs
in Georgia, Iowa, and Washington to determine how much
FDIC could rely on State agencies' examinations in lieu
of their own.
As a result of FDIC's evaluation of these
programs, they will not be continued in their present form.
Iowa and Washington do not wish to continue and Gotorgia
favors a different approach.
The Georgia plan calls for
independent examination of all problem banks, barks requiring special supervisory attention, and banks with deposits
over $100 millio.; all other banks will be examined by
Georgia and FDIC in alternating years.
FDIC will make
this arrangement available to other States or will consider other plans suggested by them.
Since 1975, FRS has also been conducting an experimental program with the Indiana banking agency to reduce the
duplication of examination effort between FRS and State
examiners.
Under this program a FRS examiner is present
as an observer 'uring State examinations.
Afterward, he/she
prepares a separate examination report for FRS use only.
If the program is successful, FRS implies, it will be expanded to other States.

Training
FRS and OCC are improving their examiner
programs as a result of studies made in 1975.

42

training

In 1977, OCC will begin a uniform personnel development program of continuing education and career development.
The continuing education segment will consist of at
least 80 hours of technical training during each of the
first several years of employment to assuie that examiners
acquire needed skills at appropriate stages of their
Later career development will include technical
careers.
and managerial training.
FRS broadened the training available to its examiners
by starting two schools in 1976 to cover specialized examinations bf bank holding companies and compliance with
Officials said they will
various consumer protection laws.
establish an EDP school and a seminar to update senior
exaniners on new developments in the baxlking industry.

43

HOW CLOSE IS THE WORKING RELATIONSHIP
AMONG THE FEDERAL BANK REGULATORY AGENCIES?
The legislation establishing the three agencies created
several overlaps in authority.
The area with the greatest
potential for duplication is bank examination.
FDIC has
statutory authority to examine all insured banks (including
national banks and State-chartered FRS members); FRS has
statutory authority to examine all member banks (including
national banks and State-chartered member banks that are
insured by FDIC); and OCC has statutory authority to examine
all national banks.
The three agencies
responsibilities:

have other

interrelated

-- OCC is responsible for closing national banks
which have become insolvent and FDIC is responsible for liquidating these banks.
-- FRS has
holding
insured
company
sidiary

primary responsibility for inspectiny bank
companies, but FDIC may examine a nonmember
State bank subsidiary and its parent holding
and OCC may examine a national bank suband its parent holding company.

-- All banks are required, under the Securities Exchange
Act of 1934, as amended (15 U.S.C. 78a et seqg.), to
report to FRS on extensions of credit for the purchase

of stock.
If any bank fails to furnish such information, HitS may inspect such bank in order to obtain the
information.
There are other reasons why a close working relationship is needed among the three agencies. During our study
we noted that in some areas similar activities were being
carried out differently by the three agencies. From an
overall Federal viewpoint, this did not provide for efficient operation. Moreover, in some cases these differences
resulted in treating different classes of banks unequally
under similar conditions.
We recognize that each agency has been granted certain
authority by the Congress and enjoys considerable independence
of action. Nevertheless, the agencies must deal with certain
common problems, and in our view, could deal with them better
by working together.
44

We found some evidence o' cooperation and coordination among the agencies:
-- Since 1938 they have operated under a uniform
agreement for classifying bank assets and
certain securities during examinations.
-- They have agreed not to examine the same banks.
OCC examines national banks, the FRBs examine State
banks which are members of the System, and the FDIC
examines insured State banks which are not FRS
members.
--An interagency coordinating committee was established
in 1965 to resolve conflicting rules, regulations,
and policies.
-- The agencies exchange their examination reports.
Each agency can request any report of another agency.
In addition, as a matter of practice, FRS provides
FDIC with all reports on banks composite-rated "3" or
"4", and OCC provides FRS with all reports and FDIC
with all reports on problem banks.
--They also exchange information on holding company
matters both at the headquarters and regional levels.
FDIC and OCC provide the FRBs with copies of their
The FRBs
examination reports of subsidiary banks.
to
report_
give their holding company inspection
FDIC and OCC.
We could not ascertain the full extent of coordination
and cooperation among the three agencies because such
For example, no minutes are
efforts are mostly undocumented.
taken at the coordinating committee meetings and few records
are maintained of telephone conversations and informal discussions between the staffs of the three agencies.
The current framework for coordinating the activities
of three regulatory agencies, i.e., the coordinating commitis primarily a forum for exchanging information about
tee
possible conflicting rules, regulations, or policies which
It does not provide a mechmight exist among the agencies.
forces to improve
join
anism for the three agencies to
common problems.
resolve
to
or
the bank supervisory process

45

We identified several areas where, the agencies could
have benefited by sharing experiences about innovations
in bank supervision and undertaking activities jointly
or on a reciprocal basis. These areas include
-- developing and testing new examination approaches,
-- developing procedures for enforcing consumer protection laws,
-- developing bank monitoring systems,
-- training bank examiners,
-- developing a formal evaluation process for determining whether an examiner has the skill to be in
charge of a bank examination,
-- evaluating the soundness of loans to foreign governments and their agencies,
-- examining foreign branches and subsidiaries of
U.S. banks,
-- processing bank data reported periodically,
-- supervising bank holding companies,
-- evaluating the soundness of large loans that are
shared by two or more banks, and
-- developing criteria for identifying

problem banks.

Recommendations
We recommend that either (1) the FDIC, FRS, and OCC
jointly establish a more effective mechanism to combine
forces in taking significant initiatives to improve bank
supervision or in solving common problems or, (2) the
Congress enact legislation to establish a mechanism for
more effective coordination.
We would be glad to assist
the committees in drafting appropriate legislation.
(See
FDIC comments in app. III, p. III-60, FRS comments in
app. Iv , p. II-29, and OCC comments in app. I, p. 1-13.)

46

We

recommend also that FDIC,

FRS,

and OCC:

-- Joinfly staff a group to analyze shared national
credi.; at State and national lead banks under
Federal supervision and that the three agencies
use the uniform classification of these loans when
they examine the participating banks.
(See FDIC
comments in app. III, p. III-41, FRS comments in
app. II, p. II-22, and OCC comments in app. I,
p. I-9.)
-- Work together to refine their monitoring systems and
their approaches to examining for compliance with
consumer credit laws.
(See FDIC comments in app. III,
p. III-41, FRS comments in app. II, p. II-22, and OCC
comments in app. I, p. I-9.)

47

HOW DOES GAO FEEL ABOUT
THE AGENCIES' NEED FOR
ADDITIONAL POWERS?
The supervisory agencies have asked the Congress to
influencing banks to solve
legislate additional powers fo
their problems. Their experience has been that current legal
powers are not useful for addressing major causes of bank
problems exemplified by recent failures.
For example, the agencies contend that to remove
a bank official they now must prove he is dishonest.
This is sometimes difficult to do; furthermore, bank problems can be caused by persons who are incompetent, but not
dishonest. Therefore, the agencies have requested authority
to remove bank officers for gross negligence.
They also maintain that cease and desist orders,
although useful for some situations, are not always deterOrders require a bank either to
rent to bank mismanagement.
stop performing an act or to take affirmative action to
correct the conditions resulting from any such violation
or practice.
One basic contention is that individuals cause bank
problems, but available legal powers are not aimed at deterring individuals. Therefore, they have requested authority
to levy administrative fines against bank officials who
violate certain laws and regulations.
We found that the agencies have not used their available powers as aggressively as they might have, especially
(Last year the agencies increased
cease and desist orders.
Evidence from our sample
the number of orders issued.)
banks is inconclusive as to the effectiveness of the legal
However, most bank
enforcement ;tep, that have been taken.
failures in the last 5 years were caused by individual bank
managers who followed self-serving loan practices or were
incompetent as stated in examination reports and corresponFurther, of the banks on the problem list as of
dence.
December 31, 1975, 57 percent were cited by examiners for
ineffective management.

48

Therefore, notwithstanding the agencies' underused
their current powers, we believe additional powers could
enhance their ability to deal with bank problems. We would
support legislation empowering FDIC, FRS, and OCC to
-- remove bank officers for gross negligence and
--levy fines against banks, or against bank officers
or directors, for violations of certain laws and
regulations.
We would also support legislation to allow OCC to
present removal proceedings at adminsitrative hearings
conducted by FRS.
Under current statutes OCC provides
information to FRS, which presents the case at hearings.

49

HAS THE COMPTROLLER OF THE CURRENCY
CONSIDERED APPLICANTS FOR NATIONAL BANK

CHARTERS ON A FAIR AND CONSISTENT BASIS?
From January 1, 1970, to April 30, 1976, OCC considered
The Comptroller
865 applications for establishing new banks.
approved 57 percent of the applications.
In evaluating a charter application, OCC primarily
considers the proposed bank's capital structure, future
earnings, and management, and the convenience and n,.eds of
persons in the area to be served.
The Comptroller had considerable latitude in deciding
-or the
whether to approve or reject an application and,
most part, gave no reason for ruling a particular way.
According to our interpretations of written comments by
each of the five staff reviewers, approval of applications
appeared to have been mainly related to the "convenience and
needs" factor, broadly interpreted to include need for competition, for new or better services, or for service to a

special clientele.

Redctions included

in our

sample seemed

to be based largely on the lack of need for a new bank
or on expectations of newly approved State banks opening
in the community.
There was no practical way to determine whether OCC
had been fair and consistent in approving or disapproving
new banks because the agency lacked (1) definitive criteria
for its staff to use in evaluating applications and (2)
an adequately documented decisionmaking process.
The differing opinions of the staff reviewers
suggests that more definitive criteria are needed to provide for uniformity in the application review process and
insure that all factors are considered and resolved either
favorably or unfavorably. Although definitive criteria
that would apply to every application may be difficult to
develop, we believe the matter warrants further study by
OCC.

On November 1, 1976, the Comptroller took several actions to improve the processing of applications and to make
Even with these changes,
charter decisions more consistent.
however, we believe that more definitive criteria and documentation are still needed.

50

Recommendations
Accordingly, we recommend that OCC (1) develop more
definitive criteria for evaluating charter applications
and (2) thoroughly document the decisionmaking process,
including an identification by reviewers of each factor
as either favorable or unfavorable.
(See OCC comments in
app. I, p. I-5.)

51

HAVE SOME BANKS CHANGED CHARTERING
AUTHORITIES TO AVOID SUPERVISORY PRESSURE?
With the Comptroller of the Currency's approval, Statechartered banks are allowed to convert to national banks.
By changing charters in this way, banks become subject to
supervision by the Comptroller, instead of by a State
authority and either by the Federal Deposit Insurance Corporation or by the Federal Reserve System.
Because such banks are already in operation and usually
subject to Federal supervision, decisions on requests to
change charters are not as important to the banking iiidustry as initial charter decisions.
From OCC's viewpoint,
however, approval means entry into the national banking
system; therefore, decisions to change charters are equally
as important to it as initial charter decisions.
In November 1976, OCC began requiring an applicant to
give reasons for wanting to convert to a national bank.
It
also

established policy stsa-~ments whic. say that it will
ordinarily approve conversions that are consistent with
maintaining a sound national banking system but that conversions should not be motivated by supervisory pressures from
other

bank

regulators.

We reviewed the 71 State-to-national conversion appJications OCC acted on trorm January 1972 through April 1976.
Sixty-four were approved
four were rejected, and three
were withdrawn.
Before deciding on the conversion applications, OCC
either examined the bank itself or reviewed earlier Federal
or State bank examination reports.
Most banks converting
to national banks were judged by OCC or their previous supervisors as sound in every respect.
Only one bank was receiv-

ing special supervisory attention when it converted.

Before OCC had policies governing conversion requests.
several banks appear to have converted to national charters
to avoid another agency's supervisory action.
Supervision
was usually consistent because OCC addressed the problems
identified by the previous regulators.
Other banks converted to obtain more favorable consideration of requests for branches, mergers, or other
structural changes. OCC approved many of these requests
52

ail-r

separately considering

their

merits.

State banks also converted for reasons unrelated to
supervisory disagreements, such as to have the same type
of charter as affiliated banks or to obtain the prestige
and Federal Reserve-related banking powers of national
banks.
OCC's recently established policies and its requirements that State banks explain their reasons for wanting
a national charter should help it make better informed
decisions about whether a bank should be allowed to
change supervisors.

53

WHAT DO BANKERS THINK OF BANK SUPERVISION?
Historically banks in the United States have always
been regulated by some level of government.
At a minimum,
establishing a bank has depended on a government "blessing"
in the form of a charter.
Inherent in the power to grant

a charter are the power to revoke it and an interest in
the soundness of the chartered entity.
Protecting this
interest has led, historically, to bank supervision.
Charters, examinations, and followup actions are central
aspects of the Federal Government's relationship to commercial banks.
Regarding this relationship, bankers surveyed 1/ endorsed government intervention in the banking industry.
Almost 90 percent indicated that "elimination of bank
regulation entirely" would be, to some degree, "detrimental.'
Other aspects of government intervention received
similar endorsements. For example
--70 percent felt eliminating Federal chartering
would be detrimental,
-- 72 percent felt eliminating State chartering
would be detrimental, and
--88 percent felt eliminating bank examinations
would be detrimental.
Approximately 80 percent of the bankers responding
to our questionnaire opposed any bank regulatory arrangement which does not include the States. Bankers clearly
favored the dual balking system over - solely Federal system.
The present regulatory system was supported by 58 percent
of the bankers surveyed. For those who responded to the
three alternatives to the present system (three Federal
agencies but no State involvement, one Federal agency with
State involvement, and one Federal agency with no State involvement) the most favored (42 percent) was one Federal
agency with State involvement.

1/ We mailed a questionnaire to 1,678 commercial banks, of
which 1,501 or 89.5 percent responded.
The sample included banks of varying sizes from each of the three
agencies, including both problem and nonproblem classifications.
54

Officials at large banks considered the present strucRespondents from small banks aptural arrangement best.
peared to be somewhat indifferent to the number of Federal
supervisory agencies.
Almost half of the problem banks, as compared with
60 percent of the nonproblem banks, supported the present
arrangement.
Only 11 percent of problem banks supported
the two alternatives which exclude State government participation, and only 48 percent supported the alternative consisting of one Federal agency together with State involvement.
Problem banks were clear about what they opposed,
but they were not clear about what they supported.
Our data revealed a seemingly contradictory pattern.
While bankers from small banks tended to be more supportive
of Federal bank examiners and the examination process than
bankers from large banks, these same small bankers were less
inclined to support the present structure of bank supervision. Bankers from small banks appear to strongly support
what is being done, but they are somewhat ambivalent about
who does it so long as the dual Federal-State involvement is
preserved.
Bankers as a group have a generally favorable opinion
of Federal bank examiners and of the bank examination proOver 90 percent of our respondents rated the senior
cess.
Federal examiners' competence as "adequate" or better.
Also, 84 percent rated the senior Federal examiners' understanding of specialized areas as "adequate" or better.
Finally, 67 percent rate.i the examination process as
"effective" or better in achieving various objectives.
Respondents grouped by deposit size and problem classification also generally support the examination process.
However, big banks and problem banks were less supportive
than others.
Besides asking our respondents to indicate how effectively bank examinations achieve certain objectives, we
also asked them to indicate, from a list of 15 possible
bank examination objectives, th? 5 they believed to be
most important.
For the respondents as a whole, the five
objectives in order of importance were:
-- Protection of the safety of depositors'
-- Evaluation of asset quality.
55

funds.

-- Compliance with laws and regulations.
-- Evaluation of management.
-- Evaluation of internal control,
internal audit.

including

We also asked them to indicate from the same list
the five objectives that they believed the Federal agencies
considered as most important. The five Federal agency
objectives in order of importance, as perceived by our
respondents, were:
-- Compliance with laws and regulations.
-- Evaluation of asset quality.
-- Evaluation of capital adequacy.
-- Protection of the safety of depositors'

fu dt

-- Evaluation of internal control, including ir_-iilal
audit.
The two lists are nearly identical, although the ranking
order differs. Although bankers rank "evaluation of management" in the top five, they don't think the dgencies do.
This is somewhat surprising considering the importance
that management evaluation receives in agency rating systems.
Also, while the bankers did not place "evaluation ;f capital
adequacy" among the top five objectives, they believe the
agencies do.
Regardless of how the data was grouped, except for
large banks, our respondents did not believe that the
agencies rate "evaluation of management" among the top
five.
Large banks also rank "compliance with laws and
regulations" somewhat lower than small banks.

56

APPENDIX I

APPENDIX I

C)
Comptroller of the Currency
Administrator of National Banks
Washington, D.C 20219

January 14, 1977
Honorable Elmer B. Staats
Comptroller General of the
United States
General Accounting Office
441 G Street, N. W.
20548
Washington, D. C.
Dear Mr. Staats:
Enclosed find original and one copy of our comments on the
recommendations advanced in the draft General Accounting
Office Report entitled "Study of Federal Supervision of our
Nation's Banks".
I understand that in accordance with the usual procedures
our comments will be included in toto in the final report.
Sincerely,

Robert Bloom
Acting Comptroller of the Currency
Enclosures

Note:

Page references have been changed to conform to the
final report.

1-1

APPENDIX I

APPENDIX I

PREAMBLE
Bank Examination and the Office of the
Comptroller of the Currency

The Office of the Comptroller of the Currency (OCC) commends the General Accounting
Office (GAO) for the objective and workmanlike quality of GAO's report and for the
positive attitude s )wn by the GAO staff which prepared the report.
The GAO report co -retly states that one important goal of bank regulation is
maintaining the *vundness of the banking system; achievement of that goal requires
minimizing the number of bank failures. We agree with that goal, and suggest that
the banking agencies record over the last forty years has been a good one. For
example, 1974 witnessed a severe economic recession and the two largest bank
failures in the history of the United States -- yet no depositors in these banks
lost money and confidence in the banking system was maintained. The average annual
bank failure rate since 1937 has been 0.08 percent -- a remarkably low frilure rate
for any human endeavor.
But it is the other goal of supervision which is not stressed in the GAO re:,ort.
The ultimate measure of how well a bank supervisory agency operates is how vell
the barking system operates. The OCC believes that one of its major funct is to preserve a competitive, responsive and innovative system. Bank sup,
-ion's
role is to ensure that the banking system is able to provide the widest possiblarray of banking services to both the depositor and the borrower.
Thus, the bank supervisory agency has two contradictory goals: monitoring soundness
and sponsoring the competitive, innovative response. It is this dual role which
presents the basic paradox for the bank supervisory agenc,
kn intensely
competitive industry can never be completely safe.
Striking the balance between these two goals is the basic problem of the bank
supervisory agency. According to a former Comptroller of the Currency:
One regulatory approach is to identify a proble- in one area
and remedy across the board, taking no notice of the different
characteristics, or idiosyncracies of the compcnents of the
whole. That approach is acceptable if the object is to produce
a "fail-safe" banking system. Believe me, I can screw down the
National Banking System with enough regulations to prevent bank
failure. But, under that regime, the banking industry would be
financing the capital needs of the country and its citizens at
about 60% of capacity, and that is not in the public interest.
Equally important, it is contrary to the economic principles of
our nation. Instead, I would advocate that we free up the system
to manage itself, loosen the bonds and tak. the quite limited
risks that some unit will slip through the supervisory nec and
founder.

I-2

APPENDIX I

APPENDIX I

A well known critic of bank supervision, economist George J. Benston, has addressed
the question of the costs of bank regulation -- both the direct cost of running
the agencies and the indirect costs of limiting competition by the banking Lndustry -- and has suggested that the best solution is improved supervisory techniq:es. Specifically he recommended:
1.

A primary responsibility of the supervisory agencies is to determine
the most effective method cf examining banks.

2.

Supervisory agencies should be able to use bank reporting as a guide to
self-eyamination by the banks and as a preliminary examination tool.

3.

Models should be developed that predict possible problems.

4.

Banks that are likely to get into trouble should be examined more
frequently and in greater depth.

That list, although not complete, is similar to the revisions of examination
procedures proposed by the consulting firm of H skins & Sells and implemented by
the Office of the Comptroller of the Currency. Examination of the larger banks
has moved from a detailed examination of the bank's assets to an in-depth evaluation of the bank's management, auditing, and control systems. Instead of concentrating on the bank's loan customers, the OCC has moved to an evaluation of the
bank itself. During 1976, the OCC began to use bank financial reports as a
preliminary examination tool, identifying potential difficulties at individual
banks.
GAO reviewed these and other new procedures being adopted by the OCC, and concluded:
As discussed in Chapter 4, we believe that the traditional examinations of the three agencies have concentrated too much on the review
of loans and not enough on bank policies, procedures, practices, controls,
and audit. The changes made by FDIC and FRS will not substantially remedy
this defect. In our view the new procedures being implemented by the
OCC offer the best opportunity for improvement. The OCC's revised
commercial examination procedures should provide the agency with more
meaningful information regarding the banks it supervises and result in
more complete and consistent examinations. More importantly, the new
approach should result in early detection of situations which could
lead to deterioration in some aspect of banking operations. This approach could help avoid bank problems after they have occurred.
Thus the OCC is not attempting to improve bank supervision through arbitrary regulations which might limit bank services to the public. Instead the OCC is attempting
to foster procedures in each bank through which that bank can better manage itself.
The GAO report -- while endorsing the new OCC procedures -criticism of the OCC for not developing its new programs in
two other agencies. As pointed out in the OCC responses to
the OCC has attempted to share its new ideas with the other

I1-3

makes the implied
conjunction with the
the GAO recommendations,
two agencies. The

APPENDIX I

APPENDIX I

OCC also endorses the GAO recommendation of more formalized communication among
the agencies concerning new examination techniques. The OCC takes issue, however,
with the apparent GAO assumption that the best way to generate new ideas is through
an interagency committee (or, as some have proposed, through a giant monolith
combining the three agencies). A primary virtue of three agencies, each with somewhat differing statutory responsibilities, is the ability of a single agency to
experiment with a new idea or procedure. It is doubtful that the new OCC examining
techniques endorsed by GAO could have been developed otherwise. A unified approach
is important and appropriate after a new idea has been proved successful, not
when it is being first developed.
In summary, the purpose of the OCC is to operate so that economic progress and change
is not inhibited while simultaneously, preventing unsound banking practices. It
is that fine line of promoting innovative response while supervising the banking
system that makes bank supervision so difficult. The banking system has just come
through its first major economic crisis since the world wide depression of the
1930s. There were some casualties. But, in fact, the threatened financial crisis
did not develop, and the banking system seems to be stronger today than it was
before. New procedures have been developed by the banking system and the continuing
dynamic future of American banking is assured. For che first time we are assured
that, just as the industry has changed, the tactics and techniques of a major bank
supervisor, the Office of the Comptroller of the Currency, has changed in a similar,
positive, fashion.

I-4

APPENDIX I
Recommendation

APPENDIX

I

(2-21)

Accordingly, we recommend that the Comptroller of the Currency (1) develop
more definitive criteria for evaluating charter applications and (2) thoroughly
document the decision-making process, including an identification by reviewers
of each factor as favorable or unfavorable.

OCC Response:
The OCC is the only federal agency with the responsibility for chartering
banks. It charters banks in all of the 50 states and in Puerto Rico and the
Virgin Islands. The widely differing banking environments found in the U.S.
make it almost impossible to develop definitive criteria which can be universally
applied such as in states like Arizona, which has 6 National Banks, and in
Illinois which has over 400 National Banks. The diversity of criteria therefore,
is a function primarily of the differing political, social and economic environments in which the OCC must operate. The OCC's chartering criteria, of necessity,
must be somewhat flexible. That is only to be expected since the OCC does not
charter in one environment. Also, under the terms of the McFadden Act, the
OCC's actions are often affected by applicable state law.
The new corporate guidelines, development of which began in September, 1975, and
which became effective on November 1, 1976 answer many of the criticisms of the
GAO. Written opinions containing reasons are now sent to applicants receiving
denials. As examples, we quote from three recent letters sent to applicants
denying their charters. One letter in part, states:
Based upon the population and the median income per household, it
would appear difficult for many individuals in the primary service
area to qualify for a loan. Furthermore, income levels are inadequate
to provide a sufficient deposit base for the proposed bank to become
a viable institution.
In another case, we quote in part:
In view of the Supreme Court decision
in Whitney and the Federal Reserve Board's decision in InterMountain Bank
Shares, it would be an exercise in administrative futility for this Office
to approve the present charter appli:ation...Should West Virginia change
its statutes or should the statute be successfully challenged, then this
Office could consider a new application inrlight of these changed circumstances.
In still another case, the denial letter to the applicants stated: The
new guidelines state that a new banking office will not be approved, if
its establishment would t-reaten the viability of a newly chartered indepandent bank. Such protecition will typically not exceed one year. As
you are aware, the new bank opened on September 27, 1976. It is the
opinion of this Office that this newly chartered independent state bank
is entitled to the protection set forth in the Comptroller's policy
statement.

1-5

APPENDIX I
Recommendation

APPENDIX I
(2-21) Continued

Every attempt is now ride to document thoroughly the decision-making process.
Further efforts will be made by our Office to idenfity each factor as favorable or unfavorable.
Our decisions have been subject to judicial review for many years. In the long
series of court cases covering our chartering process, the Comptroller's
decision on a charter application has never been finally overturned by d
reviewing court. See annotations to 12 U.S.C. 21 et seq.
Our Department of Research & Economic Analysis has undertaken a market study
of 35 mational banks chartered between 1969 and 1971. The economic study
attempts to identify, statistically, those factors which can be identified with
the growth or lack of growth of these new banks. The results of that study,
if positive, will be incorporated into our decision-making process. We are
hopeful that quantification of a sufficient number of pertinent factors applicable to a majority of cases will result.

Recommendation

(4-7)

Therefore, we recommend that the Board of Directors, FDIC, the Board of
Governors, FRS, and the Comptroller of the Currency establish scheduling
policies and procedures which would avoid the setting of examination patterns.

OCC Response:
Historically, the OCC has viewed surprise as an important element of an
examination. However, a primary feature of our new examination approach
entails the pre-examination analysis wherein the examiner will determine
the adequacy )f internal control and audit activity. The OCC feels the
best deterrent for fraud is not periodic unannounced visits by examiners
but rather .he existence of sound bank policies, procedures, internal
control and audit activity on a continuing basis. The element of surprise
is necessary only in those cases where such factors are suspect.

Recommendation

q
(4-0)

We recommend that the Board of Directors, FDIC, and the Board of Governors,
FRS, adopt flexible policies for examination frequency which would allow
them to concentrate their efforts on banks with significant problems.
We recommend that the Congress amend the National Bank Act to allow the
Comptroller of the Currency to examine National Banks at his/her discretion.

OCC Response:
We support the recommendation of legislation to permit OCC discretion in
scheduling the frequency of examinations. The current method of adapting
the depth of examinations to the needs of each bank, based on NBSS data
and pre-examination analysis, fully complies with law. However, greater
statutory discretion would enhance our effectiveness in this regard.

I-6

APPENDIX I

APPENDIX I
Recommendation

(4-29)

We recommend that the Board of Governors, FRS, and the Comptroller of the
Currency develop and use a single approach to the classification of loans
subject to country risk.

OCC Response:
The OCC has a well established procedure using a single approach to the
classification of country credits. This procedure makes use of information
from many governmental and non-governmental sources and examiners in all
fourteen national bank regions.
Copies of tute minutes of our committee meetings and a,,, .c,,ilting classifications
have always been provided to members of the staff of the Board of Governors.
The process of country risk evaluation is more precisely an art than a science.
Most of che evaluation process is judgemental. However, the interagency
meetings held to date have been beneficial in determining basic differencce in
philosophies.

Recommendations

.3
(4-3.)

We recommend that the Board of Governors, FRS, and the Comptroller of the
Currency implement procedures whereby major foreign branches and subsidiaries,
including subsidiaries of Edge Act corporations, are examined periodically and
whenever adequate information about their activities is not available at the
home office.
Also, we recommend that the Board of Governors, FRS, and the Comptroller of the
Currency exchange each other's examiners' to cut expenses when conducting
examinations in foreign countries.

OCC Response:
a)

Overseas Examination

National Banks are required by Regulations K & M to provide examiners with
whatever credit and financial information the examiner deems necessary to
evaluate the condition of the bank's foreign branches and subsidiaries. Those
regulations require such information be transmitted to and maintained at the
bank's head office. The OCC has for practical purposes defined "head office"
to include any foreign or domestic office of the bank which is readily accessible
to its examiners. For example, all international credits of one large national
bank are examined from two domestic offices and four foreign offices located in
London, Caracas, Tokyo and Manila. All of that bank's many branches and subsidiaries located in Europe the Middle East and Africa are examined from duplicate
records in London.

I-7

APPENDIX I
Recommendations

s¢

APPENDIX I

(4-30) Continued

Supplemental examinations to determine the quality of the bank's operations
are made on-site overseas when necessary. For purposes of performing asset
and operational examinations, the OCC established in 1972 a London office
permanently staffed by six examiners. In fulfilling its overseas examination
obligations, the OCC in 1976 examined 141 overseas branches and subsidiaries
of 25 banks located in 37 countries; 154 on-site examinations were performed
by 215 National Bank Examiners.
b)

Joint Examinations

The GAO recommendation has merit. As a bare minimunm the r'ysical support
of the three agencies could be jointly provided.
Further arrangements could
be made so that any of the agencies could jointly commission overseas examiners.
In this regard, the OCC is willing to seek a cooperative solution with our
sister agencies.
Under present statutes, however, such a sharing of examiner forces may be
difficult. Section 481 of Title 12 (12 U.S.C. 481) directs the Comptroller
of the Currency to appoint examiners who shall examine every national bank.
That same section empowers the Comptroller to make a thorough examination of
all the affairs of the banks under his jurisdiction including the affairs
of all affiliates of National banks "other than member banks", in order to
disclose fully the relations between the bank and its affiliates and the
"effect of such relations upon the affairs of such bank". (Emphasis added.)
Recommendation

(7-25)

We recommend that the Comptroller of the Currency invite FDIC and FRS to
joint'v review and evaluate its new examination approach. Further, we
recom. and that, in the event cf a favorable assessment of the new process,
the Board of Directors, FDIC, and the Board of Governors, FRS, revise their
examination processes to incorporate the features of OCC's new examination
approach.

OCC Response:
Examination Approach
On November 23, 1976 OCC staff members made a presentation to approximately
20 FRS and FDIC staff members on the revised examination procedures. Copies of
our draft Handbook of Examination Procedures were furnished. Their review and
evaluation on an ongoing basis is welcomed. The Acting Comptroller has proposed
to the Interagency Coordinating Committee that a permanent staff group be set
up for this purpose.

I-8

APPENDIX I

Recommendation

APPENDIX I

l'a
(7-46)

the Board of
Additionally, we recommend that the Board of Directors, FDIC,
a group
staff
jointly
Currency
the
of
Governors, FRS, and the Comptroller
under
banks
lead
National
and
State
at
credits
national
shared
to analyze
classificauniform
the
use
Federal supervision and that the three agencies
tion of these loans when they examine the participating banks.

OCC Response:
Shared National Credits
FDIC
In 1974, meetings were held with representatives of the OCC, FRS and
review
the
for
program
uniform
a
using
of
present to discuss the possibilities
in the
of selected large shared loans. loth the FRS and the FDIC found merit
particitheir
delay
to
existed
program but they believed sufficie'nt pitfalls
representapation in the program. Also, in March of 1974 this Office met with
proposed
the
discuss
to
Supervisors
Bank
State
of
Conference
tives of the
with
program. They indicated interest and agreed to work out arrangements
various bank supervisors.
reviews
In 1975, the Office of the Comptroller of the Currency conducted uniform
write-ups
loan
The
Banks.
National
applicable
in
credits
of shared national
the FDIC.
generated by these reviews were made available to both the FRS and
and hoped
program
the
in
interest
continued
In March, 1975 FRS expressed their
1975
November,
In
overcome.
be
could
"pitfalls"
the
if
they could participate
member
state
involving
program
review
y were instituting a test
FRS revealed
,g our methods and procedures. In July, 1975 FDIC again expressed
banks parallE
of the
interest and a meeting was held in September, 1975 with representatives
way
whatever
in
welcomed
be
would
involvement
FDIC. This Office indicated FDIC
they deemed appropriate.
the data
During May, 1976 the second uniform review was conducted and again
FDIC.
and
FRS
the
to
available
generated was made
of the Federal
In July, 1976 the Comptroller of the Currency and the Vice Chairman
national
shared
to
agencies
two
the
of
approaches
the
discuss
to
met
Reoerve Board
the
with
FRS
provide
to
credits. It was agreed that the OCC should continue
whether
information developed under its program and to explore at a staff level
would be
which
agencies
two
the
between
developed
be
could
procedures
uniform
that
acceptable to all of the Federal Reserve Banks. It is our understanding
shared
involving
project
pilot
a
conducting
is
Bank
the New York Federal Reserve
associated
credits which may assist in resolving some of the anticipated problems
with a combining of the approaches of the two agencies.

2'
Recomme.dction

(7-24)

of Governors,
We also recommend that the Board of Directors, FDIC, the Board
their
refining
in
together
work
Currency
FRS, and the Comptroller of the
examinations.
compliance
credit
consumer
to
approach
their
and
systems
monitoring

I-9

APPENDIX I

APPENDIX I

OCC Response:
Monitoring
The OCC has met on several occasions with officials of the other two
federal supervisory agencies to present its NBSS system. Those orientations
were given both orally and with complete submission of all relevant documents.
Further, we have offered the other supervisory agencies computer programs and
technical knowledge to implement the programs.
Consumer Credit Compliance
With reference to consumer credit compliance examinations the draft report
does not fully recognize that our new program is already operational. Over
6Z of our field staff is currently allocated to the consumer area. We have
conducted three two week schools which trained over 140 examiners in the new
procedures; a second series of three schools is scheduled for March and April,
and a third series will take place in the Fall. The schools stress examination
techniques and feature heavy reliance on case studies to give experience in
examining for compliance. The procedures are tailored to spot problems most
likely to result in harm to consumers. We make use of sophisticated advanced
financial calculators, specially programmed for banking applications, and
sampling techniques designed to increased our effectiveness.
Eleven percent of the country's 4,700 national banks have been examined under
the new procedures. Preliminary analysis of these reports indicates that our
expanded efforts in this area are both justified and effective.
The draft report also does not reflect the extent to which other agencies have
cooperated in developing our new program. The Federal Reserve Board and H.U.D.
aided in reviewing our procedures. Speakers from the Federal Reserve Board,
H.U.D. and the Justice Department participated in our schools. Observers
from the Federal Reserve Board, FDIC, N.C.U.A. and H.U.D. attended the schools
to assess the new procedures. As a result many of our examination procedures
and teaching materials have been adopted by these four agencies. This experience has reinforced our awareness of the benefits of such cooperative
efforts.
Recommendation

is
(8-2e)

a). We recommend that the Board of Directors, FDIC, the Board of Governors,
FRS and the Comptroller of the Currency establish more aggressive policies
for using formal actions.
OCC Response:
We believe that in supervising the vast majority of national banks, our most
effective remedy continues to be the examination process and the meetings held
as part of that process between the board of directors of the bank and OCC
personnel. Since December 23, 1975, the OCC has required meetings with boards
of directors of each national bank at least once every calendar year and, in
certain Lases, following every examination. We believe that the increased use
of such meetings together with our new examination procedures and early warning
system will make our first-line, informal supervisory techniques even more effective.

I-10

APPENDIX I
Recommendation

APPENDIX I

Ib
(8-28) Continued

As the GAO report elsewhere notes, our informal supervisory techniques even
without the improvements noted above, have proven effective in rehabilitation
of most of the so-called problem bank situations. For example, over the period
reviewed by GAO informal procedures utilized by OCC were successful 84% of the
time. Nonetheless, we agree that increased use of formal agreements and cease
and d-sist orders under the Financial Institutions Supervisory Act may accelerate
correction of problems in the more recalcitrant institutions.
OCC use of such formal agreements and orders has increased tenfold from 1q70 to
1975. The OCC has originated slightly more than half of the combined total (179)
formal agreements and cease and desist orders issued by all three agencies during
the last five years. The OCC, however, supervises fewer than half the number of
banks supervised by the other two agencies combined. When compared to the
number of banks supervised, the OCC over tie past five years has used the formal
enforcement tools of Financial Institutions Supervisory Act about two and one
half times as often as the other two agencies.
It should also be noted that the three banking agencies jointly requested
Congress in 1975 to refine and increase the agencies' formal enforcement powers.
Congress failed to pass the necessary legislation.
b). Written criteria should be developed to identify the types and magnitude
of problems th'at formal actions appropriately could correct.
OCC Response:
The OCC has developed as part of its National Bank Surveillance System a severity
anomaly ranking system which identifies every three months the national banks
most likely to requil special supervisory attention. A computerized action
control system is designed to assure that the OCC responds promptly and appropriately to these situations. The criteria built into these systems identifies
more systematically and promptly those cases in which formal enforcement action
is appropriate.
4q
Recommendatiov (8-4,f)
We recommend that the Board of Directors, FDIC, the Board of Gover.ors, FRS,
and the Comptroller of the Currency develop uniform criteria for identifying
problem banks.
OCC Response:
The term "problem bank" is banking agencyv jargon for many different fact patterns.
To an outsider, it appears reasonable and logical to expect a uniform definition
of the term. The agency staff person recognizze the difficulty of reducing all
the variables to a single definition. At the same time, he has little difficulty
in communicating with colleagues in other banking agencies on particular bank
situations.
OCC's approach is to computerize to the greatest extent possible the many variables
which characterize a bank's condition and management from time to time. This results in a capability to rank all banks in relation to their peers. The final
selection of banks needing special supervision can only be done subjectively by
trained personnel using all the tools available and the results of our revised
examinations. The dividing line on the spectrum between "problem" and "non-problem"
status is hard to define but OCC is more than willing to consult and cooperate witn
the other agencies in seeking such dividing lines.

I-11

APPENDIX

APPENDIX I
Recommendation

I

(10-6)

We recommend that where feasible the Comptroller of the Currency, Board of
Directors, FDIC, and Board of Governors, FRS, combine their examiner schools
and standardize their curricula.

OCC Response:
The OCC recognizes that a common training effort and a combined examiners'
school would be highly desirable both in terms of expense and coordination
of examination policy. Our Office stands ready to cooperate fully with all
such efforts. Indeed, our Office is in receipt of a letter from Chairman
Barnett of the FDIC asking our cooperation and financial support for a
combined training facility to be constructed at a Rosslyn, Virginia site.
This matter is receiving serious attention.
The practical difficulty is that our Office has implemented the Haskins and
Sells Report which has created fundamental changes in oar examination process.
These changes are so basic to our examination process that it would be
difficult to coordinate a curriculum. A combined examiners' school is viable
only if the other agencies modernize their techniques in line with those being
implemented at the OCC. It would be possible, however, to offer jointly
courses in more generalized subjects such as Economics and Accounting.

Recommendation

II
(10-1)

We recommend that the Board of Governors, FRS (1) establish a full-time
training office to operate its examiner training program and (2) carry out
the revision of examiner school curricula which it has recognized as needed
for some time.
We also recommend that the Comptroller of the Currency, Board of Directors,
FDIC, and Board of Governors, FRS, increase their training in EDP, Law and
Accounting as desired by their examiners.

OCC Response:
As part of our acknowledged need for specialized training, and consistent with
the advice of our consultants, the Training Division of the Personnel Management
Department has identified a multitude of different specialized courses which
selected examiners will take: they include 7 different commercial examination
schools, 3 trust examination schools, an EDP school, and International school
and a consumer examination school. That program has now been implemented and
is in full operation. The schools are programmed for examiners at different
stages of their professional development. Among the many courses that will be
offered by lkilled personnel, both from within the OCC and, where necessary,
from outsiae, are ones in EDP, Law and Accounting. Among the other areas that
will be covered in that curriculum development will be specialized work in
Economics, Bank Marketing, Finance, Auditing and similar topics.

1-12

APPENDIX I

APPENDIX I
Recommendation

(11-8)

We recommend that the Board of Directors, FDIC, the Board of Governors, FRS,
and the Comptroller of the Currency either (1) jointly establish a more
effective mechanism for the three agencies to combine their forces in undertaking significant new initiatives to improve the bank supervisory process
or in attacking and resolving problems common to the three agencies, or
(2) the Congress enact legislation to establish a mechanism for more
effective coordination.

OCC Response:
The OCC has always stood for the strongest possible working relationships
between federal supervisory authorities. At the December, 1976 meeting of
the Interagency Coordinating Committee, Mr. Robert Bloom, Acting Comptroller
of the Currency, asked that the commiLtee take up at its next meeting the
subject of strengthening coordination of examination procedures. It will be
proposed that a permanent staff group be set up for this purpose. We
anticipate modification and refinement of our newly implemented examination
approach on an ongoing basis. Review and evaluation of such changes as they
affect problems common to the three agencies would be most useful.

I-13

APPENDIX II

APPENDIX II

CHAIRMAN OF THE BOARD OF GOVERNORS
:::::.

1%
=.iwt

FEDERAL RESERVE SYSTEM
WASHINGTON A. C. 20551

January 16, 1977

The Honorable Elmer B. Staats
ComptrolleL General of the United States
Washington, D. C. 20548
Dear Mr. Staats:
We appreciate the opportunity to review the General Accounting
Office's draft report on the "Study of Federal Supervision of our Nation's
Banks."
The data contained in the report reflect favorably on the
Federal Reserve's superviscry performance with respect to both banks
and bank holding companies. However, the report does suggest a number
of refinements in examination procedures and urges more uniformity of
standards among the Federal bank regulatory agencies. In most instances,
the Federal Reserve had already taken steps to accomplish the objectives
of the recommendations.
The Board's specific comments concerning individual recommendations and its general comments concerning the GAO report are enclosed.
It is our understanding that our responses to specific recommendations
will appear verbatim in the report immediately following the applicable
recommendation and that our general comments entitled "Statement by
the Board of Governors of the Federal Reserve System" will appear in the
Highlight Section of the final report.
Sincerely yours,

Arthur F. Burns
Enclosures

APPENDIX II

APPENDIX II

STATEMENT' BY
THE BOARD OF GOVERNORS OF THE FEDERAL RESERVE SYSTEM
ON THE GAO REPORT

The agreement between the General Accounting Office and the
three bank regulatory agencies, pursuant to which the special GAO review
of the bank supervisory process was commenced in May, 1976, provided
that each of the agencies involved would have an opportunity to comment
on the conclusions and recommendations set forth in that report.

In

addition to its specific comments on individual recommendations which
are set forth throughout the main body of the report, the Board also
believes that some introductory comments are appropriate.
--The report confirms the basic health and soundness
of the banking system. The number of banks requiring
special supervisory attention is a small percentage
of the total number and the percentage which have
in fact failed is much smaller still.
--The report confirms the basic soundness of the
current system of supervision. Refinements, rather
than basic revisions in the current system, are
recommended.
In most instances, the Federal Reserve
Board had already taken steps to implement such
refinements and we believe the same to be true of
the other agencies.
--The report confirms the necessity for the legislative
improvements in the bank supervisory and regulatory
area which the Board recommended to Congress as
early as September, 1975, as well as the Board's
proposals for a Federal Bank Examination Council.
The bulk of the GAO review focuses on banking institutions
which have required special supervisory attention and the responses
of the various agencies to this requirement.

Despite this limitation,

the report establishes that, at any one time, the percentage of banking

II-2

APPENDIX II

APPENDIX II

institutions in the country which for various reasons can be considered
to require special supervisory attention is extremely small, in the
neighborhood of 5 per cent.

The data in the report show that between

1970 and 1975, encompassing an exceptionally difficult economic period
for the country, only 42 of the approximately 14,000 commercial banks
in the cour.try failed.

Most of those institutions were relatively small

and aggregate losses to depositors were minimal.
Also relevant to an evaluation of the supervisory process
is the conclusion of the report that the group of banks identified as
requiring special supervisory attention is fluid.

The composition of

the problem lists changes with some frequency as the regulatory agencies
identify problems and the banks respond to the need for corrective action.
The recommendations made throughout the !eport indicate that
no need was found for any basic revisions in the country's present system
of bank regulation.

Rather, the report identifies a number of areas

which GAO believes need further attention by the agencies.

As noted

in our specific comments on the recommendations, in most instances the
Federal Reserve and the other regulatory agencies hrq already taken
actions in harmony with the basic thrust of the recommendations.

For

example, the Board was already focusing more of its supervisory resources
on institutions with known problems and less on those thought to be
in good condition.
The majority of the recommendations in the area of bank examination and supervision relate to a desire for greate

II-3

unaiformity in supervisory

APPENDIX II

APPENDIX II

treatment among the agencies.

These recommendations support the Board's

conclusions and initiatives in this area.

In December, 1975, Governor

Holland testified before the Senate Committee on Banking, Housing and
Urban Affairs and in that testimony maJe reference to the concept of
a joint bank examination council which at that time had received substantial
support within the Board.

In that regard, he stated:

Such a Council would be focused on the areas that
we believe are most in need of improvement; that
is, efficient and uniform modernization of bank
examination and vigorous and consistent follow-up
procedures when bank weaknesses are revealed. Such
a Council could be established administratively
or by statute. Its statutory authorization would
undoubtedly give more impetus to the establishment
of such a Council, and would also provide it with
more clear-cut authority to take
Jfinitive action
within its statutorily defined areas of administration.
The Federal Bank Examination Council should have
authority to establish standards and procedures
for bank surveillance, examination and follow-up,
applicable to all the Federal banking agencies,
and it should review significant problem cases wher
and as they develop. All three Federal banking
agencies should be represented on the Council.
Subsequently, at the Board's request, Senator Stevenson introduced
the Federal Bank Examination Council Act (S. 3494).
would es

Wlish mandatory uniform standards an't

Such a council

procedures for Federal

examination of banks as well as uniform reporting systems and conduct
joint schools for examiners.

The Board strongly supports such legislation

and believes a proposal along those lines could accomplish most of the
objectives set out in the report's r Commendations in the examination
area.

II-4

APPENDIX II

APPENDIX II

The report focuses extensively on the method by which the
agencies deal with problem bank situations and makes a number of recommendations for improvements in this area which will be discussed later.
Further, in a number of instances the report specifically supports enactment
of various legislative proposals made by the Board on behalf of the
As Chairman Burns stated in a letter of September 5,

regulatory agencies.

1975 proposing such legislation, "All of these recommendations arise
from the agencies concern over 'problem bank' situations anid are designed
to help prevent or correct such situations."

H.R. 9743 and S. 2304,

which embodied these recommendations, would have provided civil penalties
for violations of various provisions of Federal law by banks and bankers;
imposed new restrictions on a bank's transactions with insiders; and
placed the agencies in a position to make more effective use of the
Financial Institutions Supervisory Act of 1966.

The Board believes

that the report provides strong support for this legislation.

Tn this

regard, we note that Senator Proxmire his just introduced a bill in
the 95th Congress which encompasses these recommendations.
The GAO report stated that "Examiners found problems in nearly
all of the banks in our samples, including those not on the agencies'
problem lists . ..

."

The tables contained in the review of this element

of bank supervision showed that examiners applied
e.g.,

:trict standards;

in 70 per cent of the banks the examiners criticized the

.ume

of classified loans; violations of law and regulations, whether Snorely
technical or substantive, were identified in 55 per cent of the bank,;

II-5

APPENDIX II

APPENDIX II

inadequate routines and controls were noted in 44 per cent of the banks.
We agree with the GAO and believe the report readily confirms that bank
examiners are effective in identifying problem areas in commercial banks.
We believe that the study also demonstrates that the supervisory
agencies are effective in resolving significant problems once they have
been identified by the examiners.

We are ?

Convinced that analysis

of the dynamics of so-called "problem lists' -- one of the techniques
employeu by the GAO -- is a proper basis fo. measuring supervisory effectiveness.

Thus, we are somewhat concerned with the report's focus on the

length of time an institution remains subject to special supervisory
attention as being an indication of whether or iot the supervisory process
is, in fact, working.

We believe that substantial weight should also

be given to the percentage of banks which fail as an indication of whether
or not the process works.

Daring the period examined by the GAO, only

two State member banks failed, and they were relatively small.
However, we believe that even the focus of the report on the
dynam'

Jf the list of institutions subject to special supervisory

attention demonstrates Lhe effectiveness of the present system.

The

report shows that the composition of 'he lists changes with some frequency
as problems are identified by the regulators and resolved by the institutions in conjunction with the regulators.

During the period examined

by the GAO, 1,180 banks were added to these lists and 897 were removed.
Furthermore, as set forth An more detail in our specific responses to
individual recommendations, we believe the data gathered in chapter 8

11·-6

APPENDIX II

APPENDIX II

demonstrate that the performance of the three agencies is roughly the
same and, in fact, good for all three agencies.
In addition to the recommendations for greater uniformity
in examination and follow-up among the agencies, the report makes a number
of recommendations relating to examination techniques and training.
As more specifically set forth in our responses to the particular recc-mmenJations, the Board has taken, or is in the process of taking, effective
action compatible with the major thrust of most o' the recommendations.
For instance, a major portion of the recommendatlors deal witb the desire
that the agencies focus more of their resources on institutions with
known problems.

In this regard, the Board requires all problem banks

to he examined at a minimum of six-month intervals.

Further, the Board

has recently adopted limited scope Asset Quality and Management Performance
Examinations to be used on banks thought to be in relatively good condition.
The Board believes that these procedures give us the needed flexibility
while at the same time minimizing the likelihood that problems will
be overlooked.
In the area of training, the Board has, among other things,
recently revised curricula for its various examiner schools and has
instituted new schools in the areas of consumer regulations and bank
holding company analysis.
The Board would also like to comment on the broad purposes
of the bank examination process lest the sum of the report's recommendations
be mrisconstrued.

We believe that bank examination and supervision should

11-7

APPENDIX II

APPENDIX II

be directed at securing compliance with banking laws and regulations
and determining that a bank is operated soundly so as to assure, to
the greatest extent possible, safety and soundness of depositors' funds
and continued banking services to the community.

A system of bank regulation

which goes beyond these goals imposes certain social costs and dangers.
It is not the job of the supervisors to determine whether specific loans
should or should not be extended or whether a bank's resources should
be used in a particular manner unless such decisions contravene law
or affect the safety and soundness of the bank.

Rather, private initiative

should be encouraged to the greatest extent possible.
Finally, the report also comments on the Board's supervision
of bank holding companies.

The data in the report confirm t)at, as in

the case of banks, the percentage of problem institutions is relatively
small.

Even utilizing a sample biased toward problem institutions,

th,;re were limited instances in which bank holding companies were found
to have caused problems in the subsidiary banks.

Out of the sample

of 344 which were affiliated with bank holding companies, there were
22 banks in which the report stated that the problem was caused by the
parent holdin% company.
banks

This constitutes 6.5 per cent of the sample

ffiliated with hank holding companies.

However, the Board's

examination of the parent bank holding company in each of these instances
demonstrates that, in fact, the actions of only five holding companies
could be said to have caused any serious problem in the subs

l.ary banks.

In addition, the Board believes that it is taking effective supervisory

II-8

APPEiNDIX II

APPENDIX II

action in those cases where holding companies are causing problems for
the subsidiary banks.

In October, 1974, the Board's request for ceaseSince

and-desist authority over bank holding companies was granted.

that time, the Board has significantly expanded its supervisory efforts
with respect to bank holding companies, concentrating primarily ,.a those
exhibiting problems.

With respect to formal actions, in the 26 months

the F:ard has had the authority, it has issued 12 cease-and-desist orders
and 12 written agreements against holding companies.
In concluding our general statement, we wish to note once
again that our banking system has weathered an extremely difficult period
successfully.

The bank supervisory process of this country, which by

no means is perfect, has materially contributed to this achievement.
********************

The Board's further responses to individual recommendations
may be found throughout the body of the main report as follows:
Page

Recommendations relating to

7

-- Scheduling of examinations --------------

4

-- Flexible examination policies --

4- 9

-- Use of State examinations --------------

4-13

-- Scope of examinations ---------------------

4-17

-- Examination workpapers -------------------

4-19

-- Coiintry risk evaluation -------------------

4-33

-- Examination of foreign operations-

4-35

-- L ? examinations -------- -----------------

4-39

II-9

APPENDIX II

APPENDIX II

Recommendations relating to

Page

-- Supervision of holding companies ----------

4-51

-- Examiners' meetings with directors --------

6- 5

-- Examination report format -----------------

6-13

-- Evaluation of the Comptroller of the
Currency's examination procedures ---------

7-25

-- Analysis of shared national credits -------

7-25

-- Policies for formal enforcement actions ---

8-18

-- Criteria

for

identifying problem banks ----

8-49

-- Combined examiner schools ----------------- 10- 6
-- Separate training office ------------------ 10-11
-- Evaluation of examiners ------------------

10-15

-- Uniformity in the supervisory process ----- 11- 8

If-in

APPENDIX II

APPENDIX II

Recommendation
Therefore, we recommend that the Board of Dire tors, FDIC,
the Board of Governors, FRS, and the Comptroller of the Currency establish
scheduling policies and procedures which would avoid setting examination
patterns.
Comments
This recommendation is based upon the premise that the agencies
view surprise as an important element of an examination.

The Board

believes that,in many cases, there is serious doubt as to the benefits
to be gained and hence the desirability of surprise examinations.

In

those instances where surprise is considered important, it has been,
and will continue to be,our practice to schedule examinations so that
they cannot be predicted in advance.
Recommendation
We recommend that the Board of Directors, ?DIC, and the Board
a: Governors, FRS adopt flexible policies for examination frequency
which would allow them to concentrate their efforts on banks with known
serious problems.
We recommend that the Congress amend the National Bank Act
to allow the Comptroller of the Currency to examine national banks at
his/her discretion. We would be glad to assist the committees in drafting
appropriate legislation.
Comments
The Board already has established policies that are flexible
enough to allow us to concentrate our efforts on banks with known serious
problems.

Some years ago, the Board adopted the policy, which was reaffirmed

in 1975, that all banks considered to be in a problem status be examined
at a minimum of six-month intervals.

However, we will continue to schedule

periodic examinations of all banks under our supervision
may deteriorate with the passage of time.

Il-li

since a

Ink

As pointed out in the GAO

APPENDIX II

APPENDIX II

report, the Board recently approved the usage of Asset Quality and Management
Peformance Examinations in the case of banks thought to be relatively
free of major problems.

If this limited scope examination detects

major changes or deterioration, a full scale examination is then commenced.
These procedures give us flexibility while at the same time insuring
that problems are not overlooked.
Recommendation
We recommend that the Board of Directors, 'DIC,and the Board
of Governors, FRS, extend their current efforts to use State examinations
and, if they do, we also recommend that they
--develop minimum standards for acceptable State examiner
training and examination procedures and
--use only reports of State examinations meeting those standards.
Comments
The report recognizes our current extensive efforts to eliminate
unnecessary duplication by utilizing State examiners and State examination
reports.

If experience with our existing program in. InLiana should

indicate thlt expansion of this program is desirable, GAO's recommendations
regarding standards would be appropriate.

Indeed, the purpose of the

existing experimental program is to develop such standards.

In this

connection, however, it should be recognized that written standards
: 'one

will not insure the success of any program.

Recommendation
We recommend that the Board of Directors, FDIC and the Board
of Governors, FRS, establish procedures to base the scope of each examination
on the examiners' evaluation of the quality of the bank's controls,
policies, procedures, and audit.

II-12

APPENDIX

II

Comments
This recommendation encompasses what we are already doing.
We review the policies, procedures, and controls in connection with
all bank examinations.

In most large banks, our examiners currently

perform a preexamination review specifically focusing on controls, policies,
and procedures.

The results of such review are used to determine the

amount of scrutiny given to each area.

In -maller banking institutions,

a review of the controls, policies and procedures in effect at the last
examination is

used to develop the scope of the examination.

Recommendation
We recommend that the Board of Directors, FDIC, and the Board
of Governors, FRS, develop standards for the preparation, maintenance,
anj use of examination workpapers.
Comments
We believe that, in the vast majority of examinations, the
examination workpapers and line sheets prepared are adequate to meet
the System's needs.

The manner in which examination workpapers should

be prepared and maintained is extensively covered in connection with
the training of our examiners.
Recommendation
We recommend that the Board of Governors, FRS, and the Comptroller
of the Curryrzy, using all available information, develop and use a
single approach to classify loans subject to country risk.
Comments
The evaluation of the country risk element in international
loans calls for difficult analysis and judgment at the time lines of
credit are established or loans extended since "country risk" involves
1-13

APPI ,DIX II

APPE,,DI,,

II

an estimate of a country's political, economic, and social fortunes
over the life of the loan as they may affect the collectability of such
loans.

There is serious question as to the validity of generalized

characterizations of credits based on the country of residency of the
borrower, particularly where the characteristics of thu credit may well
vary with the borrower - private or governmental - as well as the nature
and extent of external resources available to support the loan.

For

a number of months now,the Federal Reserve has had underway a review
of country risk problems in international lending as well as appropriate
supervisory treatment of the problem.

This review has included an on-going

appraisal of the system employed by the Comptroller of the Currency.
In this rega.d, we believe that, while there may be general agreement
on the desirability of uniform evaluation of the country risk element
in individual international credits, there is a real question as to the
desirability of rating individual countries.

It mig'

be noteu, for

instance, that the Comptroller's system focuses almost exclusively on
credits to individual governments.

In any event, we believe that we

should strive toward uniform treatment.

Of course, as with respect

to many of the recommendations, the Federal Bank Examination Council
proposal would accomplish this.
Recommendation
We recommend that the Board of Governors, rRS, and the Comptroller
of the Currency implement procedures to examine (where permitted by
the country involved) major foreign branches and subsidiaries, including
subsidiaries of Edge Act corporations, periodically and whenever adequate
information about their activities is not available at the home office.

APP ,iDI< II

APPEIJDIX II

Also, we recommend that the Board of Governors, FRS, and the
Comptroller of the Currency utilize each others examiners to cut expenses
when conducting examinations in foreign countries.
Comments
The development of widespread networks of foreign branches
and subsidiaries by the major banks has brought the question of the
supervision of the banks' international operations to the forefront
in recent years.

We concur with the principle that examinations, wherever

conducted, should be adequate to provide the necessary supervisory information.
However, one constraint with which the Board has had to deal is, as
noted in the report, that, in many cases examinations of foreign subsidiaries
are not possible because of host country laws which preclude direct examinations by other governmental authorities of banks chartered in those
countries regardless of the ownership.

The Systen has not only required

that banks maintain records at the head office adequate to appraise the
risk and exposure of the banks through their foreign operations, but the System
has also provided for direct visitations of examiners to major foreign
branches in those cases where such visitations have been legally possible.
The Board believes that, on the whole, this system has worked
well.

The information available at head offices has, in general, been

adequate to assure that the banks were not unduly exposed to loss or
serious financial difficulties.

At the same time, there has teen a

continual search for better and more efficient ways of satisfying the
Federal Reserve's supervisory responsibilities in the international
field.

APPENDIX II

APPENDIX II

Beginning in the fall of 1976, on-site examinations were made
of foreign branches of State member banks where we had previously utilized
on-site inspections by State examiners or information at tne head office.
Moreover, a number of foreign subsidiaries were directly examined for
the first time with the agreement of the host government.
of those examinations has not yet been completed.

A full evaluation

One preliminary result

of that exercise has been to provide assurance that a large portion
of the material needed for proper supervision of foreign branches and
subsidiaries is in the management information systems at head offices.
In this connection, it should be noted that consultations are continuing
with foreign bank s!iperisory authorities about the ways in which access
to foreign sub

-diaries may be broadened to accommodate on-site reviews.

These consultations are part of a wider effort of international cooperation
in bank supervision.
Recommendation
We recommend that the Board of Directors, FDIC, and the Board
of Governors, FRS develop reports of examination for EDP operations
which present the problems found, corrective action needed and any necessary
explanatory data in a clear and concise manner.
Comments
The Board wishes to note that it believes its present EDP
examination report adequately presents the major problems found and
corrective action needed.

Furthermore, the System has already undertaken

a review of EDP examination procedures to determine whether there are
possible improvements, particularly in the review of $nternal controls,
and, in connection with that review, is preparing a revised examination
report.
II-16

APPENDIX II

APPENDIX II
Recommendation

We recommend that the Poard of Governors, FRS implement a
system of supervision which is based on onsite inspections of holding
companies and their major nonbanking subsidiaries. We also recommend
that the Board strengthen its oversight of holding company supervision
by establishing
--a systemwide manual of inspection procedures,
--a standard inspectic.o

report, and

--periodic onsite evaluations of Reserve bank supervisory
activities.
Comments
The System has for some time conducted on-site inspections of
selected holding companies.

Partly as a result of these inspections and problems

which came to its attention, the Board in late 1974 requested and was
granted legislative authority to impose the same supervisory remedies
on holding companies that were applicable to banks under the Financial
Institutions Supervisory Act of 1966.

In early 1976,the Board directed

that this inspection program be significantly expanded with initial
efforts directed toward holding companies requiring special supervisory
attention.
In addition, in 1975 the Board commenced work on a computer
based monitoring system in order to identify those holding companies
which might require special attention.

This program is partially operational

at the present time and is expected to be fully operable within the
next few months.
A manual of inspection procedures is currently under development.
However, completion of such a manual has of necessity awaited experience

II-17

APPE,,DIX iJ

APP-i.DIX II

gained frcm the direct an-site inspections which have been carried our.
We believe that the recommendations relating to a standardized inspection
report as well as periodic on-site evaluations of Reserve Banks supervisory
activities warrant further consideration.

We might note that the initial

steps to set up such periodic evaluations already have been commenced
by the Board.
While we see no difficulty with the thrust of the recommendations,
the Board is concerned that the method used in the GAO report may lead
to unwarranted fears as to the general health of bank holding companies.
The sample chosen was one in which problem banks were at least six times
more likely to occur than in the industry as a whole.

A sample biased

toward problem banks is naturally biased toward problem holding companies.
Under the heading "Unsound Holding Companies' Expansion Applications
Approved" the report states that th' Board approved applications by
15 of 20 "detrimental holding companies" to acquire additional banking
and nonbanking subsidiaries.

Our review cf these companies indicates

that the problems of over two-thirds of these companies were problems
centered in the banking subsidiaries as opposed to problems in either
the parent holding company or a nonbanking subsidiary.

As such, these

problems would be most effectively handled by the primary examining
authority of the bank involved.

Furthermore, L,..

majority of the applications

involving these institutions which are referred to were acted on in
the early 1970's, lcng before any of the institutions had experienced
difficulty or had been identified as requiring special supervisory attention.

II-18

APPENDIX II

APPENDIX II

In fact, only three applications were approved at a time when any of
the institutions involved was considered to be in a problem condition.
Two of these applications consisted of corporate reorganizations having
no financial impact on the institution whatsoever.

The third application

involved permission to engage in a nonleveraged, potentially profitable,
operation which was considered to be a positive factor to improve the
ccndition of the company involved.
The Board regards its policy, adopted in June 1974, of curbing
bank holding company expansion into nonbanking activities, particularly
with respect to bank holding companies with financial problems, as being
an effective supervisory tool.

In fact, the Board has acted to deny

a significant percentage of applications on financial and managerial
grounds since t'is policy was introduced, and many more have been withdrawn
by applicants after discussions with staff.

The Board believes it has

applied this policy responsibly and it remains in effect.
Recommendation
Therefore, ws- recommend that the Board of Directors, FDIC,
and the Board of Governors, FRS require their examiners to meet with
the bank's board of directors or audit or examining committee after
each examination.
Comments
The System has for many years been concerned that the board
of directors be particularly aware of the results of an examination.
Thus, the System has histor:ically required that the examination report
be considered and disc.ussed at a meeting cf the board of directors.
To insure that this is done, directors are required to sign a statement

II-.l)

APPENDIX II

APPENDIX II

attached to the report that it has been so read and considered.

Further,

examiners are instructed to review the minutes of board of director
meetings to insure that the spirit of these requirements has been fully
carried out.
With respect to meetings, the Board in 1975 directed that
an earlier existing policy for most of the System be expanded to all
Reserve Banks.

This policy requires that Reserve Bank staff meet with

the board of directors of all so-called problem banks.

The Board believes

that such meetings are important where significant problems are revealed.
Recommendation
We recommend that the Board of Directors, FDIC, and the Board
of Governors, FRS, develop and use reports of examination which provide
the banks with the results of the examination and any necessary supporting
information.
Comments
We believe the bank examir.dtion report presently provides
the banks with the results of an examination and necessary supporting
information.

We also believe it should provide the System with the

information it needs to carry out its supervisory functions.
examination report adequately carries out these needs.

The present

It should not be

forgotten that the System also uses other methods of communicating its
views to its member banks, such as correspondence, informal meetings, and
consultations on applications.

Of course, the System is continually

exploring methods of improving communlicationls.

II-20

APPENDIX II

APPENDIX II

Recommendation
We recommend that the Comptroller of the Currency invite FDIC
and FRS to jointly evaluate its new exar.inarion approach. We further
recommend that, in the event of a favorable assessment of the new process,
the Board of Directors, FDIC, and the Board of Governors, FRS revise
their examination processes to incorporate the concepts of OCC's approaches.
Comments
The Comptroller's new procedures are based in large part on
the Haskins and Sells report.

At the time that report was prepared,

the Comptroller furnished it to the other banking institutions in the
belief that some of the recommendations might be jointly applicable.
K

task force at the "ederal Reserve reviwed the report shortly after

its issuance and concluded that, in most instances, the System had already
implemented those recommendations involved which would have been applicable
to the System.

Subsequent to that time, the development of new examination

procedures at the Comptroller's office has been substantially completed.
Recently, senior members of the Board's staff attended a briefing by
the Comptroller's office on these new examination procedure- and the
report form to be used by that agency.

The Board believes that the

Comptroller has been most cooperative in sharing his new systems with
us and fully intends to use whatever benefits may be derived from the
Comptroller's efforts in this area in our on-going review of our examination
procedures.

11-21

APP[,4DI,

PPLNrPIX

11

1T

Recommendation
We recommend that the Board of Directors, FDIC, the Board
of Governors, FRS, and the Comptroller of the Currency jointly 3taff
a group to analyze shared national credits at State and nationel lead
banks under Federal supervision and that the three agencies use the
uniform clac.sification of te.ase loans when they examine the pa.ticipating
banks.
We recommend that the Board of rirectors, FDIC, the Board
FRS, and the Comptroller of the Currency work together
Governors,
of
monitoring systems and their approaches to examining
their
refine
to
for compliance with consumer credit laws.
Comments
A joint approach to shared national credits it clearly desirable.
In fact, in June 1976 the Board and the Office of the Comptroller of
the Currency entered into a preliminary agreement which provides for
a sharing by each agency of examiners' classifications of a national
credit.
The second portion of this recommendation deals with the desirability
of uniform refinement of consumer credit enforcement and compliance
po. Lcies.

in the report, the GAO states that some agencies believe

there is a possible "conflict between a bank's objective of financial
soundness and strict compliance with consumer credit laws."
does not agree with this statement.

The Board

On the contrary, we believe that

stringent enforcement of consumer laws and regulations will achieve
compliance and thereby reduce the likelihood that banks will incur
substantial liability as a result of consumer suits.
The Federal Reserve has had the major responsibility for drafting
regulations to implement the explosion of legislation that has taken
place in this area over the past two years.
II-.:

In this connection, the

APPE;JDIX 11

APPEi;DIX II

Board's Division of Consumer Affairs has worked very closely with the
other agencies.

It has formed a Federal Reserve task force to develop

approaches to the enforcement of newly enacted consumer credit laws.
A cadre of examination specialists who will concentrate on inspection
and compliance is being trained.

Two schools on consumer regulations

were conducted in 1976 and four have been planned for 1977.
Additionally, examination manuals that deal with the full
array of consumer regulations have recently been prepared.

A new examination

report form dealing exclusively with this area has been prepared and
is expected to be in use in the near future.
Recommendation
We recommend that the Board of Directors, b1DIC, the Board
of Governors, FRS, and the Comptroller of the Currency establish more
aggressive policies for using formal actionas.
Written guidelines should
be developed to identify the types and magnitude of problems that formal
actions could appropriately correct.
Comments
In this section, the report notes that each problem situation
has to be evaluated on a case-by-case basis and formal action would
not always be appropriate.

The report goes on to recommend that more

aggressive policies be used for formal actions and that written guidelines.
be developed to identify the types and magnitude of problems that formal
actions could appropriately correct.

In this regard, we note that the

report confirms that all of the agencies have already markedly expanded
their formal enforcement activities.

On November 3, 1975 the Board issued

a policy statement emphasizing its intention to take formal action where
appropriate in connection with violations of the Bank Holding Company
Act.
11-23

APPENDIX II

APPENDIX II

Further, we do riot believe that adequate weight has been given
in the report to existing hindrances to formal action under the Financial
Institutions Supervisory Act of 1966.

This chapter does, however, support

the Board's existing recommendations for changes to the Financial Institutions
Supervisory Act which would enable the supervisory authorities to remove
bank officers for gross negligence and to assess civil penalties for violations
of laws and regulations.

These legislative recommendations were made

in response to procedural and substantive problems inherent in making
effective use of the present formal procedures set fcrth in the Financial
Institutions Supervisory Act.

In this regard, the Board's letter of

September 5, 1975, to the banking committees of both Houses of Congress
setting forth the Legislative proposals made it clear that there were
a number of situations in which the existing formal regulatory remedies
would have little or no value in preventing or ameliorating problem bank
situations.

We believe that those recommendations, embodied in H.R.

9743 and S. 2304, would help to substantially reduce the incidence of
problem banking situations.

Further, the Board has cor.tinued to review

areas in which it appears that changes may be of substantial aid.

The

Board intends to submit further legislative proposals to this end in
the very near future. 1v tnir regard, Chairman Proxmire has introduced
legislation in the 95th Congress which encompasses the earlier
recommendations.
The Board is further concerned that the discussion in this
chapter of the manner in which the agencies are handling problem bank
situations may not present an accurate view in all respects.

II-,'>

The major

APPENDIX II

APPENDIX II

shortcoming ir this regard stems from the fact that the different agencies
utilize problem bank lists for varying purposes.

Furthermore, even

between agencies with similar goals, different judgments may occur as
to the severity of an institution's problems and the length of time
monitoring is required.

Meaningful comparison between agencies' enforce-

ment activities in this area is therefore impossible.

We would, however,

note that the report's conclusions relating to the agencies' effectiveness
in returning institutions to nonproblem status are not supported by
the tables since the percentages used excluded institutions withdrawing
from membership and ,irging. Presumably, the approving agency found
in the case of the mergers, as required by the Bank Merger Act, that
the financial and managerial condition of the resulting bank was satisfactory
and, in the case of withdrawals from membership, supervisory pressure
may well have contributed to such withdrawals.

Further, as noted in

the table, withdrawals and mergers are disproportionately high in the
sample for the Federal Reserve.
Recommendation
We recommend that the Board of Directors, FDIC, the Board
of Governors, FRS, and the Comptroller of the Currency develop uniform
criteria for identifying problem banks.

APPELVDIX II
APPENDIX II

APPENDIX II

Comments
As previously noted in earlier responses, the rating systems
are utilized for different purposes within different agencies.

However,

we believe there is certainly room for much common ground in this area.
The legislative proposals for a Federal Bank Examination Council referred
to earlier would aid in this development, though judgmental evaluation
of any common criteria will likely lead to some diversity.
Recommendation
We recommend that where feasible the Comptroller of the Currency,
the Board of Directors, FDIC, and the Board of Governors, FRS, combine
their examiner schools and standardize their curriculums.
Comifents
The examiner schools were a combined effort of the three agencies
when they were established in 1952 by the Federal Reserve.

However,

in 1962 the Office of the Comptroller of the Currency withdrew from
the program, believing it preferable to operate its own school.

In

the early 1970's the number of FDIC students necessitated some sessions
held for FDIC examiners only and, when the FDIC enrollment needs continued
at this high level, it was decided that the only practical course of
action for the FDIC and the Federal Reserve System was to establish
separate schools.
The Board believes that a joint effort in this area would
be appropriate and desirable.

This is among the reasons the Board supports

the concept of a Federal Bank Examination Council.

Short of this proposal,

the Board will explore with other agencies the feasibility of conducting
joint schools.
IIi-L.

APPEi,.DIX II

APPENDIX II

Recommendation
We recommend that the Board of Governors, FRS, (1) establish
a full-time training office to operate its examiner training program
and (2) carry out the revision of examiner school curriculums which
it has recognized as needed for sometime.
We also recommend that the Comptroller of the Currency, the
Board of Directors, FDIC, and the Board of Governors, FRS, increase their
training in EDP, law, and accounting, as desired by their examiners.
Comtents
One individual currently administers the various Federal Reserve
examination school_. eld in Washington.

In addition, one full time

staff member is assigned to handle preparatory and procedural aspects
suc.h as registration, printing and distribution of instructional materials
and day-to-day dealings with instructors and students.

Other responsibilities

for the different schools have been assigned to various members of the
Board's staff who are experts in each field of training.

For instance,

the curriculum for the newly established Holding Company School was
devised by members of the Federal Reserve staff expert in matters relating
to holding companies and the new Consumer Regulations School is handled
by individuals who have been actively involved in implementing the recent
consumer legislation.

The Board believes that this system ha3 met its

needs.
If the report's recommendation for a joint school is adopted,
this would reduce the need to consider a separate office at the Board.
However, if such arrangements cannot be worked out, the Board will consider
establishing such an office.

II-27

AP"ENDIX II

APPENDIX II

We might note that the portion of this recommendation relating
to a revision of exdmination curricula had been started prior to the
report.

At the direction of the System Education Committee, the curricula

for the schools for assistant examiners and examiners were updated and
revised in the spring and summer of 1976 and the curriculum for the
EDP school was revised in the fall.

The Holding Company School and

the Consumer Regulation School have been recently established and therefore
have new curricula.
With respect to that portion of the recommendation relating
to additional training in specific areas, the Board has a previously
scheduled session of the EDP school set for 1977 which will use a recently
updated curriculum.

The laws relating to consumer affairs are extensively

covered in schools developed by the Office of Consumer Affairs now conducted
in Washington as part of the overall examination program.

The Board

will study the question whether additional training in the areas of
law and accounting should be provided to examiners.
Recommendation
We recommend that the Board of Governors, FRS also establish
formal evaluation process to measure the competence of persons seeking
advancement to examiner status.
Comments
We note that this recommendation is not based upon a conclusion
that the examiners of any one agency are more or less competent than
those of another agency.

Standardized tests are merely one way of arriving

at e formal evaluation, and we would not want to rely on them exclusively.

II-28

APPENDIX II

APPENDIX II

However, there is something to be said in favor of formal tests as a
supplementary evaluation device, and the Board intends to investigate
their feasibility.
Recommendation
We recommend that either (1) the Board of Directors, FDIC,
the Board of Governors, FRS, and the Comptroller of the Currency jointly
establish a more effective mechanism to combine their forces in undertaking
significant initiatives to improve the bank supervisory process or in
attacking and resolving common problems, or (2) the Congress enact legislation
to establish a mechanism for more effective coordination. We would
be glad to assist the committees in drafting appropriate legislatiot..
Comments
The Board is pleased that this portion of the report supports
its previous conclusions and initiatives in this area and favors the
legislative approach.
In December, 1975, Governor Holland testified before the Senate
Committee on Banking, Housing and Urban Affairs and in that testimony
made reference to Lhe concept of a joint Bank hndmination Council whichi
at that time had received substantial support within the Board.
that regard, he stated:
Such a Council would be focused on the areas that
we believe are most in need of improvement; that
is, efficient and uniform modernization of bank
examination and vigorous and consistent follow-up
procedures when bank weaknesses are revealed. Such
a Council could be established administratively
or by statute. Its statutory authorization would
undoubtedly give more impetus to the establishment
of such a Council, and would also provide it with
more clear-cut authority to take definit:Je action
within its statutorily defined areas of administration.

II-29

In

APPENDIX II

APPENDIX II

The Federal Bank Examination Council should have
authority to establish standards and procedures
for bank surveillance, examination and follow-up,
applicable to all the Federal banking agencies,
and it should review significant problem cases when
and as they develop. All three Federal banking
agencies should be represented on the Council.
Subsequently, at our suggestion, Senator Stevenson introduced
the Federal Bank Examination Council Act (S. 3494).

Such a Council would

establish mandatory uniform standards and procedur 3 for Federal
examination of banks and uniform reporting systems and conduct
joint schools for examiners.

The Board believes that a proposal along

these lines could accomplish most of the objectives set out in the report's
recommendations in the examination area.

II-30

APPENDIX III

APPENDIX III
':;

FEDERAl. DEPUSIT INh;'lRANC[ CORPORATION. .- asljli,,u

l .o i. 2,'

January 17, 1977

Honorable Elmer B. Staats
Comptroller General of the United States
Washington, D.C. 20548
Dear Mr. Staats:
I appreciate the opportunity to comment on the draft of your report
to Congress on federal supervision of the commercial banks in this country.
In general, I believe that the General Accounting Office has done a
workmanlike job with an extremely difficult task, male more difficult by a
relatively tight time frame. We feel that your comments as an impartial
professional observer should be studied carefully by us in an atmosphere
of cooperativeness and receptiveness. In that vein, I would like to cormmnent
on a few points in the draft.
1. The day-to-day relationship which the FDIC has with state banking
supervisors is extremely important in our supervisory effort. Unlike the
Comptroller of the Currency, we supervise banks who are operating under
50 state laws as well as the Federal Deposit Insurance Act. Those banks
are chartered by 50 different state supervisory authorities and the manner
of supervising those banks at the federal level differs as a result from state
to state.
2. It is important to realize that the FDIC is the sole federal
regulator for the entire mutual savings bank industry, a $100 billion industry.
While I appreciate that your report is directed only to commert ial banks,
I believe it is essential to take into account its activities with respect to
the mutual savings bank industry in order to understand the supervisory
effort of the FDIC.
3. Your report emphasizes the need for flexibility in examination
techniques. We wholeheartedly concur and as a result of a continuing study
going back a number of year.,
we amended in early November of 1976 our
basic memorandum which governs our examination policy. This amended
General Memorandum No. I is quite consistent with the thrust of your report
and I am sorry that you did not include it and a full discussion of it in your
report. We like to think that the philosophy outlined in this memorandum,
which we have tested during the past few years by experimenting in different
III-1

APPENDIX III

APPENDIX III
Honoralle Elmer B. Staats
January 17, 1977
Page Two

regions, is the best philosophy for the FDIC to pursue in the examirnation
of nonmember banks. Since it is so central to our operations, and since
it is a rel Lvely new statement of a flexible examination policy, I would
personally have liked to have seen your in-depth comments about it.
4. We believe, as your report recommends, that more formal
actions should be taken in the supervisory process by federal regulators.
We have attempted to pursue that policy, particularly since late spring and
early summer of 1976, and have requested from the Congress additional
supervisory powers.
5. The report notes the large number of violations of the law during
a typical examination. I was pleased to note that ojul point out that some of
the laws and regulations are complex and that some of the violations were
of a technical nature that would in no way affect the soundness of a bank.
Rightfully, you also point out that other types of violations, such as a loan
in excess of a bank's legal lending limit, could result in losses to a bank.
In our experience, the major portion of violations of laws set forth in reports
of examination do not affect the safety and soundness of a bank. All violations
of laws or regulations are a matter of concern, of course, but it is the particular responsibility of the bank regulator to consider each violation in terms
of whether it was intentional or willful, the consequences flowing therefrom,
the likelihood of continued violations, and other similar matters, and to then
take the appropriate corrective action.
6. Finally, the report implicitly argues that Corporation examiners
should be criticizing loan policies before bad loans are made. I certainly
agree that a closer review of loan policies is important, and criticism of
such policies in advance of their implementation be made where the policies
will obviously lead to an unsafe and unsound condition for the bank or to
violations of law. Most written loan policies will be stated in such a way,
however, that a reasonable examiner will find it extremely difficult to find
something significant in them to criticize. I suspect that the written
policies themselves are not the problem but rather the implementation
of those policies. I certainly see no expertise in our Corporation for
drafting standard written policies that banks we supervise should pursue.
The FDIC was not created to manage banks, nor do I believe that it is your
intention to have your report suggest that. Nevertheless, it does suggest
it, and I do feel obliged to make these comments about that implication.

I II-2

APPENDIX III

APPENDIX III

Honorable Eimer B. Staats
January 17, 1977
Page Three

Finally, the FDIC Division of Bank Supervision has prepared extensive detailed comments concerning recommendations and comments made
in your report which I enclose for your consideration. Please excuse the
length and the aetail of those comments; I believe they reflect, however,
the thoughtfulnes i with which we have reviewed your report.
Thank yeo. for permitting us the oppo:tunity to comment on the
draft of your report.
Very truly yours,

Robert E. Barnett
Chairman
Enclosure

III-3

APPENDIX III

APPENDIX III

FEDERAL DEPOSIT INSURANCE CORPORATION
DIVISION OF BANK SUPERVISION

Staff General Camments and Agency Recommiendations

Note:

Page references have been changed to conform
to the final report.

III-4

APPENDIX III

APPENDIX III
CHAPTER 4

The GAO report indicates that the agencies have not established criteria or levels of acceptability with respect to financial ratios and comparisons used in the examination process.
FDIC uses financial ratios as general guidelines for initial
screening purposes.

In banking and in finance, ratios are only

indicators, and as such need to be individually assessed.

The key

element in banking, as in a number of other industries, aside from
management, is the quality and turnover rate of the inventory.

In

banking, of course, inventory is principally made up of loans

i

securities.

Since no two banks have identical inventories, it

logically follows that where the relevant ratios for two banks
are identical or in the same range, further analysis is required
before a meaningful evaluation of a bank's condition can properly
be made.
The examination is designed to and does enable the FDIC to
ascertain the overall condition of the bank, the quality of its
management, and the extent of compliance with applicable laws and
regulations.

Moreover, the examination report, including the ex-

aminer's recommendations, is thoroughly reviewed and analyzed at
the appropriate Regional Office.

During these reviews, the re-

viewer also considers the Statements of Condition and Reports of
Income and Dividends filed by the bank; the bank's complete correspondence file, showing its history and the attitudes and abilities of the bank's management; reports of loans to the bank's
III-5

APPENDIX III

APPENDIX III

officers at other banks, reports of loans against the bank's stock
at other banks, and any supervisory programs which are in effect;
and, computerized monitoring systems which subject the bank to a
number of financial checks.

The major purpose of this review is

to determine the extent and type of supervision which may be
needed, not just "...for arithmetic accuracy, grammar, logic,
support for statements, and internal consistency," as the GAO report states.
After review of the examination report as well as other relevant data at the Regional level, another review process is conducted at the Washington level for each bank.

Corrective and

follow-up programs are initiated at the conclusion of the examination, and in addition to possible on-site vi:sitations or follcwup examinations, the bank's "vital signs" are monitored via an
automated monitoring system fed by data from call reports, Reports
of Income and Dividends, and examination data.
The GAO report also states that examinations have not given
enough emphasis "to the bank's basic management practices, operations, and controls."

Both from a policy standpoint and the prac-

tical application of that policy, the FDIC has been and is in the
forefront of stressing the need to review, analyze and evaluate
the policies and controls of a bank under examination.

Thus, the

following quotations from the Manual of Examination Policies typify our basic approach to this phase of the examination process:

III-6

APPENDIX III

APPENDIX III

"The Examiner's evaluation of the loan portfolio involves much more than merely appraising the individual
loans therein. Present management and administration
of the overall loan account, including the establishment
of sound lending and collection policies are of vital
importance if the bank is to be continuously operated in
an acceptable manner."
(Section H, page 3, paragraph
III.)
"Management of a bank's securities portfolio is facilitated by the adoption of a definite investment policy.
*** Details of the investment policy, expressed in
writing, should establish standards for selection that
thoroughly consider: (a) Quality, (b) Maturity, (c)
Diversification, (d) Marketabilit5, and (e) Income."
(Section G, page 1, paragraph 1.)
"Sound portfolio management dictates that procedures be
established and adhered to relative to the execution of
purchases and sales, review of portfolio and maintenance
of credit information."
(Section G, page 3, paragraph
III.)
"An important part of the Examiner's duties is the appraisal of the bank's internal controls to determine
their adequacy for assuring both the necessary degrees
of accuracy in recorded information and reasonable protectisal of the bank's assets."
(Section P, page 7,
paragraph III. A.)
In addition, the essential thrust of the examination is premised on the concept that the entire posture of the bank rests on
its management practices, operations and controls and these areas
of concern are carefully reviewed and evaluated in the course of
an examination and at other key points in the supervisory process.
For example, the examination report, which, by necessity, must be
limited to essentials, includes 13 schedules dedicated to the practices, operations and controls of the bank's management oat of a
total of approximately 30 schedules in the report.

III-7

APPENDIX III

APPENDIX III

The GAO report states, in part:
"While the agencies reportea some violations of consumer
protection laws and regulations, they acknowledged that
they have not aggressively monitored consumer protection
law compliance, and they have begun revising their ap(See Ch. 7.)"
proaches.
While we dc not argue with the implication of the above statement, the FDIC has expended considerable resources in the area of
consumer protection.

It is estimated that about 10% of our super-

visory effort is taken up with examining for compliance with consumer laws and other matters not related to safety and soundness.
We recognize, however, that additional efforts will be necessary
to enforce the many recently enacted consumer laws and regulations.
Some of the major activities of the Corporation in the area of consumer protection are:

(1) adoption of a separate compliance report

for reporting examinations for compliance with consumer laws, which
has significantly increased the volume of violations cited over
the former method used; (2) establishment of the Office of Iank
Customer Affairs which serves as a focal point within FDIC for
protecting the legitimate interests of bank customers;

(3) expand-

ed training for examiners and assistant examiners in consumer laws
and regulations, including an orientation in consumer laws for assistant examiners, a week of training for senior assistant examiners, and case problems and additional training for commissioned
examiners; and (4) providing information and education to bankers
and to a lesser extent to consumers (e.g., FDIC has under active
consideration issuance of a series of pamphlets to consumers covering consumer laws and banking and FDIC's role in that area).
III-8

APPENDIX III

APPENDIX III

In 1972, the FDIC considered issuing regulations and held
hearings on regulatory proposals dealing with the subject of discrimination in granting home loans.

However, for a number of rea-

sons, including that there was a paucity of data needed to write
effective regulations, final regulations were not issued.

However,

a major undertaking conducted jointly by the FDIC and the OCC has
been undertaken to develop a program to insure that the banks under their jurisdiction are complying with federal laws prohibiting discrimination in the granting of home loans.

During the test

phase, approximately 300 banks will use a specially designed form
in connection with their home mortgage lending activity.
expects that the new systems of data retention

The FDIC

and analysis will

provide a reliable indicati~on of where discriminatory lending is
taking place and serve as an adjunct to the examination and complaint mechanisms already used by the Corporation.
The GAO report implies that. t.e FDIC has the authority to examine routinely a'l
banks.

insured banks, including member and national

In point of fact, the legislative history of the Federal

Deposit Insurance Act of 1950 quite clearly indicates that the intent of Congress was to circumscribe the FDIC's examination of
member and national banks in the following manner (H.R. Rep. No.
3049, 81st Cong., 2d Sess. 3 and 4):
"In providing direct authority to the Corporation to
make a special examination of any national bank, Listrict
bank, or State member bank, the conferees were firmly of
the opinion that such authority is not to be utilized by
the Corporation to embark upon a program of regular
III-9

APPENDIX III

APPENDIX III

periodic examinations of such banks, which would only
result in a needless duplication of effort. Such
special examination authority is to be utilized by the
Corporation only in a case where, in the judgment of the
Board of Directors, after a req-iew of the Federal Reserve
or Comptroller of the Currency examination reports, there
are indications that the bank may be a problem case, or
that it is in a condition likely to result in loss to the
depositors or to tho Corporation."
Unless otherwise directed by Congress, the FDIC feels conaccordance with

strained to exercise its examination Pathority
the above statement of Congressional intent,

i..
.

ditioD, the

further implication in the GAO report cf overlapping examination
authority having to be parceled out through voln.,tary agreement
between the three agencies is not, at least with respect to the
FDIC, completely accurate.
The GAO report lists four criteria for scheduling examinations.
We simply note in passing that the list of criteria for scheduling
examinations fails to mention the primary criteria employed by
FDIC, namely the overall condition, compliance posture, and needs
of the bank about to be examined.
The following comments are directed to the statements in the
GAO report relating to Electronic Data Processing (EDP) matters:
The FDIC has recognized the ,heed to devote additional attention to EDP operations and to expand EDP expertise within our examiner and supervisory staffs.

Efforts are continuing to develop

more EDP field examiners and provide an interim career path position for a select cadre of our commercial examiner force.

While

our commercial examination effort addresses all aspects of bank
III-10

APPENDIX III

APPENDIX III

EDP, including the developments in electronic funds transfer, we
also recognize the need for the development of EDP expertise in
trust operations and will be devoting attention to that area during 1977.

We are planning to provide EDP review examiner positions

for each of the 14 Regional Offices, as appropriate, to accommodate
EDP examination needs.
The Corporation offers an introe.uct ry course in EDP for assistant examiners, entitled Course ir Examining a Computerized
Bank-I

(CECB-I), which is designed to prepare them to evaluate EDP

input/output controls and reconcile the automated applications to
the general ledger control accounts.

Approximately 150 FDIC assis-

tant examiners are processed through this school each year.

In ad-

dition, a Course in Examining a Computerized Bank II (CECB-II) is
offered for senior assistant and commissioned examiners to train
them in basic automation concepts and computerized examination
techniques.

Approximately 125 examiners complete this cours3 each

Finally, an eight-week advanced course entitled Field Exam-

year.

iner Advanced Automation Training (FEAAT) was commenced in 1974 to
provide in-depth technical training in EDP matters.

Through 1976,

59 examiners have completed this course and two sessions have been
scheduled for 1977 for approximately 28 more examiners.

According-

ly, all of the FDIC's EDP training needs are provided in-house.
The FDIC has developed and implemented an instalment loan
retrieval package for the use of examiners in conducting examinations.

This package not only eliminates menial data-gathering
III-11

APPENDIX III

APPENDIX III

efforts and saves considerable manhours, but it has improved the
quality of examinations, uncovering some practices which may have
gone undetected heretofore.

Further, during 1976 three other de-

posit EDP capabilities were added to the examiners' software package and a mortgage loan capability will be implemented early in
1977.

Other applications of EDP for use in conducting examina-

tions are in various developmental stages and software will be
considered for trust examinations during 1977.
EDP techniques and capabilities are also being used within
the Washington Office to seek and project solutions to problem
and failing bank situations.
While it is true that many banks do not have enough data
processing activity to justify purchasing an in-house computer and
satisfy their data processing needs through contract servicers, a
number of small banks have acquired so-called mini-computers and
perform their own data processing on-premises.

The evaluation

(examination) of contract servicers presents no unusual problems
for our trained EDP examiners and the evaluation procedures employed parallel those used for bank-operated data centers.

How-

ever, the evaluation of mini-computer operations presents unusual
control considerations and our experience in this area has not
matured.

We are continuing in our efforts to develop a sound ex-

amination approach in this area.
The Division of Bank Supervision Manual of Examination Policies, Appendix C, provides guidan,. for the preparation of EDP
III-12

APPENDIX III

APPENDIX III

checklists, questionnaires, summary comments and report of examination treatment for banks with their own computers.

Memorandums

to Regional Directors, EDP examiner conferences, and EDP seminars
provide communication and input for the redesign of examination
practices and training courses.

Each Region adopts its own EDP

examination program and some variance does occur, depending on the
EDP sophistication found in the banks supervised.

The provisions

for the interim EDP examiner career path and EDP review examiner
positions for each of the Regions during 1977 should result in
improved examination efforts as the circumstances and need dictate.
The average number of man-days per EDP evaluation in 1975 was
3.9 and year-to-date 1976 is 4.1.

Our experience indicates that

Regional Offices with the more sophisticated

banks tend to devote

more manpower to EDP evaluations and to develop more expertise in
EDP matters than Regional Offices with less sophisticated banks.
Further, it seems, within certain limits, the more knowledgeable
the EDP examiner, the more time expended in conducting evaluations.
The FDIC furnishes the results of data center evaluations to
the bank's management or to the independent data servicer of a
state nonmember insured bank.

Where a data center evaluation is

conducted as part of a bank examination, the findings

- the

evaluation are incorporated into the report of examination.

Where

the data center evaluation is conducted independently of a bank
examination, the findings are transmitted under separate cover.
These evaluations findings may consist of the EDP examiner's
II -13

APPENDIX III

APPENDIX III
summary comments with or without the questionnaire.

Where the

questionnaire is included, appropriate explanation is provided to
ensure that the reader understands that a negative response to a
particular question does not necessarily constitute an unsatisfactory finding with respect to that part of the EDP operation
covered by that section of the questionnaire.

Our experience in-

dicates that many data center managements have requested the entire questionnaire for their review and we feel that it serves as
a useful educational tool for management.

However, the question-

naire is viewed by the FDIC as a formal workpaper.

The results

of an evaluation of servicer data centers are available to serviced state nonmember insured banks on request or at the optic n
of the Regional Director without any request.

They are also

available to any other federally insured serviced institution upon
request.

All data center evaluation reports developed by the FDIC

are considered to be confidential and the property of the FDIC and
appropriate statements to that effect accompany each such report
released.
The creation of EDP review examiner positions at the Regional
Office level should provide the capacity to communicate more effectively with all data centers and help to achieve more uniform
correction of operational deficiencies.

III-14

APPENDIX III

APPENDIX III

Recommendation (page 4-7)
Therefore, we recommend that the Board of Directors, FDIC, the Board of
Governors, FRS, and the Comptrcller of the Currency establish scheduling
policies and procedures which wculd avoid setting examination patterns.

FDIC Response
We believe that our recently adopted General Memorandum #1, which has
been under consideration and extensively tested for several years prior
to adoption, largely satisfies this recommendation.

For more extensive

comments on our General Memorandum, please refer to our comments on the
recommendations contained on page 4-$ of the GAO Report.

III-15

APPENDIX III
Recommendations

APPENDIX III

q
(page 4-g)

We recommend that the Board of Governors, VDIC and the Board of
Governors, FRS adopt flexible policies for examination frequency which
would allow them to concentrate their efforts on banks with known
serious problems.

We recommend that the Congress amend the National Bank Act to allow
the Comptroller of the Currency to examine national banks at his/her
discretion.

We would be glad to assist the committees in drafting

appropriate legislation.

FDIC Response
Although it was FDIC's long-standing policy to examine each bank once
a year, it is inaccurate and misleading to suggest that that time-frame
was the only guideline used by the FDIC in scheduling examinations, or,
to state it another way, that examinations were not scheduled and conducted by the FDIC based upon the "bank's soundness; and the quality
of its policies, procedures, practices, controls, audit, and management."

During 1975, FDIC conducted 213 fol.ow-up examinations and a number of
on-site visitations at banks presenting either financial or supervisory
problems.

Further, those banks which were not examined in 1975 largely

consisted of banks which would not fall within the one-year time-frame
guideline under General Memorandum #1.

Although General Memorandum #1

was formally adopted in November 1976 and implemented on January 1, 1977,
the concepts and practices embodied in it are not of recent origin.
Those concepts and practices have been under consideration at FDIC since
II-16

APFENIDIX
early 1974.

III

APPENDIX III

Furthermore, the concepts and practices have been experi-

mented with and tested in five of the FDIC's 14 Regional Offices prior
to formal adoption of General Memorandum #1.

We might add parentheti-

cally that FDIC policy is to experiment on a regional basis with major
policy changes before implementation for the entire Corporation.

Accordingly, while the recently issued General Memorandum #1 expresses
more definitively that scheduling of examinations is not based on timeframe priorities alone, nevertheless, we feel that the criticism of
past scheduling practices expressed in the GAO recommendation is misplaced.

The FDIC has followed and continues to follow a policy so

aptly stated in the said General Memorandum #1, namely:
"The first priority has been and will continue to be, effective surveillance and supervision of the institutions which
present either supervisory or financial problems."

III-17

APPENDIX III

APPENDIX III
19

Recommendations

(page 4-]vI)

We recommend that the Board of Directors, FDTC, and the Board of Gov-,
ernors, FRS, extend their current efforts to use State examinations and,
if they do, we also recommend that they
--develop minimum standards for acceptable State examiner
training and examination procedures and
--use only reports of State examinations meeting those
standards.

FDIC Response
lheFDIC has determined that the Experimental Withdrawal Program conducted in three stat.

during the past three years will not be con-

tinued in its present form.

However, agreement to examine nonproblem

banks on an alternate-year basis has already been consummated with one
state ar-nthe possibility of entering into similar arrangements with
other states is being explored.

Furthermore, termination of the Experi-

mental Withdrawal Program should not be construed as a decline on the
part of the FDIC to cooperate to the fullest extent possible with the
various states or to place less reliance on the efforts of the state
supervisors.

The guidelines set forth in General Memorandum #1 provide

a workable framework for increased cooperation with the states.

Thus,

almost by definition, if the program expressed in General Memorandum #1
proves workable and if a state banking department performs in an acceptable manner, the frequency and scope of FDIC examinations in that
state will be reduced.

111-18

APPENDIX ITI

APPENDIX III
Recommendation (page 4--

and 4-4i-)

We recommend that the Board of Directors, FDIC and the Board of Governors, FPS, establish procedures to base the scope of each examinat _on
on the examiners' evaluation of the quality of the bank's controls,
policies, procedures, and audit.

FDIC Response
With respect to FDIC examinations, the findings and conclusions expressed by GAO are not accurate.

The primary factor influencing the

scope of the examination is not size, but the known history of strengths
and weaknesses of the particular institution.

Furthermore, FDIC examiners

do pre-plan the scope of an examination, by studying applicable files and
previous examination reports, and noting any material changes in the management or style of operations since the last examination.

FDIC examiners have in recent years reviewed a bank's internal controls,
policies and procedures prior to actual commencement of the examination
in order to establish the scope of the examination within the minimumn
standards prescribed.

With respect to smaller banks, however, such a

review tends to be less formal, hence harder for GAO to detect than with
larger banks.

Considerable leeway in this respect is provided for in the

recently adopted General Memorandum #1, and we reiterate that these procedures were considered and extensively tested in five of the FDIC's 14
Regional Offices for several years prior to formal adoption.

III-19

APPENDIX III

APPENDIX III

Recommendation

(page 4-31

We recommend that the Board of Directors, FDIC, and the Board of Governors, FRS, develop standards for the preparation, maintenance, and
use of examination workpapers.

FDIC Response
The standards for the preparation, maintenance, use and importance of
examination workpapers are included in the course of study at the
various schools operated by the Corporation and in our on-the-job training program.

The examination workpapers do, in fact, cover a number of

items other than the details relating to specific loans and securities
in support of comments contained in a Report of Examination.

We be-

lieve our examination workpapers will permit a determination that appropriate examination procedures have been followed, provide support
for the preparation of the Report of Examination, and are uti 'zed at
the next examination.

III-20

APPENDIX III

APPENDIX III

Recommendation (page 4-a3)
We recommend that the Board of Directors, FDIC and the Board of Governors, FRS develop reports of examination for EDP operations which
present the problems found, corrective action needed and any necessary
explanatory data in a clear and concise manner.

FDIC Response
The summary comments page of the FDIC EDP questionnaire provides clear
and concise descriptions of the results of a data center evaluation.
In our judgment, a new evaluation report is not necessary at this time
and our form, if effectively used, is comparable to the new one recently
adopted by the OCC.

However, we view our questionnaire as a constantly

evolving tool which will be revised frequently in order to stay abreast
of industry developments and to meet the burgeoning needs of our field
personnel.

See also our comments regarding

cluded with our general comments.

III-21

.DP evaluation reports in-

APPENDIX III

APPENDIX III
CHAPTER 5

GAO stated, in relevant part, that:
"The relaticn;;nip between the frequency with which
were cited for problems with internal routines and
trols and violations cf laws and regulations--both
which are related to management effectiveness--and
frequency of criticism of management effectiveness
not what would have been expected."

banks
conof
the
was

"While the examiners frequently cited banks for having
problems in two areas indicative of management effectiveness--internal controls and violations of laws and
regulations--they did not often criticize management
effectiveness. As shown below, management effectiveness was most often criticized in problem banks with
less than $500 million in deposits even though 30 to
50 percent of larger banks in the general and problem
samples were also criticized for violations of laws and
regulations and poor internal routines and controls."

"Violations of laws and regulations reflect on management's capability."
Generally, the size and character of the operation engaged
in by a bank defines the scope and requirements of sound internal
controls for that particular bank.

Clearly, the internal con-

trols deemed appropriate for a large, sophisticated operation are,
in most cases, not appropriate for a smaller, less complicated one.
Management is charged with the responsibility of deciding the internal controls best suited for ivs bank in order to provide adequate protection for its assets and a meaningful flow of information to senior management.

Recognizing the practicalities of the

situation, FDIC closely monitors the various internal controls
III-22

APPENDIX III
employed by banks un

APPENDIX III
-_

our direct supervision and our examiner

personnel may comment on apparent weaknesses observed.

However,

if the particular system has worked with reasonable effectiveness
for a given bank, is within the general bounds of prudence, and
does not constitute an unsafe or unsound practice, corrective
measures are not aggressively pursued, notwithstanding the critical comment in the examination report.
Banking is a highly controlled industry and, thus, is subject
to a plethora of laws and regulations on both the federal and state
levels.

It is, therefore, not unexpected that banks will on occa-

sion be found to have violated, intentionally or unintentionally,
a particular statute or regulation.

It is the job of the bank

regulator to consider each violation in terms of whether it was
intentional or willful, the consequences flowing therefrom, the
likelihood of continued violations, and management's history of
compliance and attitude toward taking appropriate corrective measures.

Accordingly, if the violation is unintentional or merely

technical in nature and not recurring, criticism of management
effectiveness would not seem warranted.

If otherwise, of course,

criticism of management is probably appropriate.

In short, in

this area as well as all areas of its supervisory responsibility,
the FDIC attempts to follow a rule of reason.

Overreaction to

technical, unintentional violations of law or regulations could,
in our judgment, impact adversely on the entire enforcement posture of the Corporation.
III-23

APPENDIX III

APPENDIX III

GAO stated that:
"The agencies rarely criticized a bank's loan policies
until loan problems developed. For example, if a
bank's managers had not adequately diversified the bank's
risks, examiners did not criticize the inadequate diversification policy until those lines of credit actually
became classified."

"For example, inadequate loan policies were not cited
by examiners until the banks had large amounts of classified loans, as shown by data for banks in our general
and problem samples combined."
The FDIC, of course, encourages banks under our direct supervision to adapt sound written loan policies.

Furthermore, in vir-

tually every formal enforcement action, FDIC routinely requires
the offending bank to provide written loan policies acceptable to
the Corporation and the appropriate state authority.

However,

oversight of a bank's written loan policies does not, and is not
inLended to, extend to writing the loan policies for the bank or
specifically prescribing how, when and to whom the credit facilities of the Lank are to be used.

We view such action by the FDIC

as objectionable on two grounds:

(1) as encroaching on manage-

ment's prerogatives, and (2) perhaps constituting a form of credit allocation.

Our task is to review the policies to determine

that they are within the bounds of safe and sound banking practices.
However, it may be somewhat naive to assume that a review of the
written loan policies of a bank will, in most cases, reveal imprudence.

Typically, it is the implementation of such policies

which generates criticism.

III-24

APPENDIX III

APPENDIX III

It is not accurate to suggest that the failure to diversify
risk is only criticized when "those lines of credit became classified."

It is both FDIC policy and practice to comment on a

failure to diversify (concentration of credit) without regard as
to whether or not the assets involved have been adversely classified.

The Division of Bank Supervision Manual of Examination

Policies states in relevant part:
"...the inclusion of a concentration of credit in a report implies criticism of a bank's policies amenable or
susceptible to management control."
(Section H, page 6,
paragraph IV. C.)

GAO note:

Omitted comments pertain to material
in the draft report but omitted from
the final report.

III-25

APPENDIX III

APPENDIX III
CHAPTER 6
The thrust of Chapter 6 may be summarized as follows:

examiners

seldom meet with bank directors, the examination reports do not convey
the bank's problems in a clear and concise manner to the directors, and
the material in the confidential section should be furnished to the
banks.

The recommendations are that the FDIC and the FRS require

examiners to meet with the directors or audit or examining committee
after each examination and that the FDIC and FRS develop and use
reports of examination "which provide the banks with the results of
the examination and any Necessary supporting information."

As we view

it, the implication is that FDIC and the FRS redesign the report of examination along the lines of the OCC's new format.
We believe that the statements and recommendations stem from a
misconception, or perhaps a misunderstanding, of the policies and
practices of the FDIC in the matters covered in Chapter 6.

The

following responses to the GAO recommendations represent a brief
summary of the FDIC's policies and practices, and efforts to improve
those policies and practices, regarding the supervisory areas dealt
with in Chapter 6.

III-26

APPENDIX III

APPENDIX III

Recommendation (page 6-5)
Therefore, we recommend that the Boa--d of Directors, FDIC, and the
Board of Governors, FRS require their examiners to meet with the bank's
board of directors or audit or examining committee after each examination.

FDIC Response
FDIC conducted approximately 7,900 examinations in 1975.

Senior offi-

cials from the various Regional Offices met with bank management on
approximately 1,750 occasions, representing 22% of all examinations.
ThLcughout 1975, there was an average of 224 banks under our supervision
which were formally designated as financial problems.

FDIC policy is

to meet with bank directors at least where problem situations exist.

FDIC staff has in the past year been reconsidering the question of how
often meetings with bank directors should be held.

In consideration of

this subject, the responsibilities of bank directors, the Corporation's
responsibility to bank directors, and our past and present practices in
holding board meetings were weighed.

In a broad sense, the board of directors of a bank is responsible for
the formulation of sound policies and objectives of a bank, the effective supervision of its affairs, and promotion of its welfare.

In dis-

charging these responsibilities, a director's duty is to exercise due
care or be exposed to a charge of negligent performance of his duty.

III-27

APPENDIX III

APPENDIX III

To insure that bank directors are aware of the contents of examination
reports, the Corporation requires that a receipt accompanying each report be signed by the bank's executive officer stating that the report
"...was duly considered by the directors...and a record of the action
taken thereon by the Board has been entered in the minutes."

Moreover,

at each examination, the examiner is charged with the responsibility
of determining that the bank's board minutes reflect a thorough consideration of examination reports and correspondence received from supervisory authorities since the last examination.

To enable bank managements to begin work on problem areas prior to receipt of the coImpleted examination report, a list of adversely classified assets and cther major criticisms is provided to the executive officer at the completion of each examination and most of the FDIC Reqional Offices have Implemented deadlines for receipt of completed examination reports in the Regional Office--usually 10 calendar days after
the close of the examination.

The FDIC Manual of Examination Policies states, with respect to examiners holding meetings with directors (Section Q, page 3, paragraph I.E.):
"Except in instances where authority has been delegated by
the Regional Director, the Examiner should consult with the
Regional Office before calling a board meeting. Ordinarily,
meetings with the board of directors should be held at the
conclusion of all examinations of problem banks. A meeting
of the board may also be required when experience and instinct tells the Examiner a likelihood exists that the bank
will be added to the problem list or will be earmarked for
other special supervision. Additionally, where there is a
substantial volume of classified assets, low capital or other
areas of important criticism, a board meeting may be desirable. This is particularly true when the trend has been unfavorable and previous admonitions have gone unheeded."
III-28

APPENDIX III

APPENDIX III

In keeping with this policy, it is Ln fact the practice in most regions
for the examiner to hold a meeting with bank directors if problems of
consequence are found at the examination, or if significant adverse
trends are noted since the last examination.

In virtually all instances

involving problem banks, a representative from the Regional Office will
to
meet with the directors, and in most cases an invitation is extended
the state authority to participate in -he meeting.
The FDIC is cognizant of the benefits flowing from more frequent meetings with the boards of directors of banks under our direct supervision
and anticipates holding such meetings with increased frequency in the
future.

We are also actively reviewing the posture of the FDIC in

this regard with a view of improving upon the timeliness and conduct of
such meetings.

III-29

APPENDIX III
Recommendation

APPENDIX III
(page 6--.3)

We recommend that the Board of Directors, FDIC,

and the Board of Gov-

ernors, FRS, develop and use reports of examination which provide the
banks with the results of the examination and any necessary supporting
information.

FDIC Response
FDIC conducted an intensive study in 1965 to assess the impact of its
examination report on banks.

As a result in 1969, a new examination

report format was put into use.

We believe this report format, and

the guidelines under which it is used, provides a clear, concise picture of problem areas to bank managements.

Various FDIC staff members

have attended familiarization sessions on the OCC's new examination
report format.

The OCC has tested his new format in only ten banks and

the impression of the FDIC staff members is that the report format is
somewhat cumbersome, especially in problem situations.

There appears to be some misunderstanding with respect to the purpose
and thrust of the confidential (supervisory) section of the report of
examination.

The purpose and thrust of the confidential section are

to allow the examiner to comment on matters uncovered during the course
of the examination which may not lend themselves to complete substantiation, but which may serve to alert his superiors that further investigatory or supervisory efforts may be necessary.

For obvious reasons,

such material is not, and should not, be provided to the management of

111-30

APPENDIX III
the bank.

APPENDIX III

However, a thorough study of the role and use of the confi-

dential section was started some months ago and, when completed, will
probably result in significant changes in its thrust, format and content, or in its elimination.

III-31

APPENDIX III

APPENDIX III
CHAPTER 7

As is indicated in the FDIC comments to the recommendations
made by GAO in this chapter, we view the impact of the changes in
the FDIC examination process set forth in General Memorandum #1
as significant and vital to an understanding of the Corporation's
examination philosophy and practices.

We believe the entire

General Memorandum should be included in the GAO r(port.

However,

in the absence of that we offer the following excerpts from General
Memorandum #1, with emphasis added:
"The first priority has been, and will continue to be,
effective surveillance and supervision of those institutions which present either supervisory or financial
problems."

"Emphasis at these modified examinations should be placed
on management policies and performance; the evaluation
of asset quality, alignment and liquidity; capital adequancy; and, compliance with applicable laws and regulations."
"In those banks with assets of $100 million or more, all
report schedules which are presently in use and are applicable to the given bank will continue to be included in the
examination report. Where the fixed asset investment is
moderate in relation to capital, there are no statutory
violations with respect to fixed assets, and absent other
problems of significance, fixed asset schedules may be
omitted from these examination reports. Further, examiners
are instructed to assess the quality of management systems
and reports as well as audit and control functions, and
where it is permissible to do so without compromising the
integrity of the examinations, utilize the output of those
systems. Cash counts and proof and verification procedures
may be omitted in those banks where it is appropriate to
do so, and branch offices which do not have a significant
volume of important assets need not be examined; however,
in the latter instance, conditions at these offices should
be reviewed with management prior to the conclusion of the
examination."
III-32

APPENDIX III

APPENDIX III

"If believed desirable in the opinion of the Regional
Director, simultaneous examinations may be arranged of
all closely related banks or subsidiaries of bank holding
companies, requiring coordination with other bank regulatory agencies. The type of examination employed in each
bank at simultaneous examinations will be at the discretion
of the Regional Director unless precluded by the guidelines
for modified examinations."

"It is expected that the Corporation's automated bank
examination programs and monitoring systems will be used
wherever possible in an effort to provide increased efficiency and conserve manpower. This use should include the
scheduling of examinations as well as their conduct. Further,
sampling techniques should be used wherever possible."
"It is expected that visitations will be frequently used as
an investigatory and supervisory tool for those banks which
show adverse trends, either at examinations or through a
monitoring system, and to gauge compliance with provisions
of cease and desist orders. Further, visitations subsequent
to management or ownership changes should be used to assess
the attitudes and abilities of the new management/ownership
if the principals are not already known to the Regional Office."
"In addition to the required periodic examinations, it will
be the policy to conduct a visitation at each new bank quarterly
during the first two years of operation (visitations need not
be held during the quarter in which an examination, either by
the Corporation or the state authority, is conducted). The
purpose of these visitations is to gain some measure of the
performance of management and the direction in which the bank
is headed. At the discretion of the Regional Director, findings
of the visitation may be reported in either memorandum form or
examination report format."
The GAO comments on the status of monitoring systems in the Office
of the Comptroller of the Currency and in the Federal Deposit Insurance
Corporation set forthi in chapter 7 of the report have served a useful
purpose in that they focus on an aspect of bank supervision which has
grown in importance in the recent past few years and may be of even greater
importance in the future.

Some clarification is needed of the fundamentals
III-33

APPENDIX III

APPENDIX III

of analysis of bank reports, of the various systems which have been
developed to facilitate such analysis for supervisory purposes, and
of a framework for evaluation of the efficiency of the programs.

For

purposes of illustration, the following comments are based upon a comparison of the National Bank Surveillance System (NBSS) and the systems
in use at FDIC.
a) An essential element of any monitoring system is a data
collection system.

The quarterly Reports of Condition and

the quarterly, in the case of large banks, and semi-annual,
in the case of smaller banks, Reports of Income comprise the
primary data base for both the NBSS and the FDIC systems.
Data from reports of examination are important supplements
to the data base; at the present time FDIC probably relies
more heavily than the NBSS on this source of information.
Obviously, a monitoring system that depends on regular
financial reports submitted by banks is only as good as the
information in the reports.

The information items must be

meaningful; they must be accurate; a..1 they must be available
on a timely basis.

Given that the OCC and the FDIC use the

same format of the Reports of Condition and Income, their
divergence appears to be in the areas of accuracy and timeliness.
The OCC has put into effect an editing system which requires
less stringent tests for mathematical accuracy and internal
consistency in the national bank reports than that used by
the FDIC in processing reports for all insured banks.
III-34

FDIC

APPENDIX III

APPENDIX III
has worked from another angle.

The Corporation has

begun to levy fines on banks that get their reports in
late.

All three federal bank regulatory agencies have

cooperated in an effort to upgrade the quality of the bank
reports so that less correction and revision are required;
clearly much more needs to be done.

While this process is

moving forward, both the OCC and the FDIC have had to modify
their analytical systems in order 'o utilize bank reports
that are sufficiently accurate for monitoring purposes.
b) Another essential of a monitoring system is a computer based
program that compiles individual bank ratios of balance sheet
and income and expense items and compares the ratios of each
bank with the same ratios for comparable banks.

Most moni-

toring systems use a technique known as "outlier analysis,"
flagging banks if its ratios deviate substantially from the
average of ratios for comparable banks.

The presumption is

that such analysis can provide clues as to banks with financial problems, current or prospective.
In a banking system as diverse as that in the U.S., differences
in operations among banks can be expected to be substantial.
A very large money market bank's ratios may appear to be unusual
or atypical of averages based upon ratios for all the banks,
large and small.

When its ratios are compared with those of

banks of comparable size, doing a comparable business, i.e.,
ratios of its "peers," such a bank may not be atypical or an
outlier.
III-35

APPENDIX III
APPENDIX III
Neither the OCC nor the FDIC could afford to wait for the
completion of definitive studies on how to sort bd'lks into
peer groups.

Such work is continuing on a theoretical level

as well as on an empirical level.

Currently, however, the

OCC G:as established peer groups on the basis of bank asset
size.
At FDIC, the effort has been made to allow the Regional Director
to specify the banks within his region that are "peers."

For

analytical purposes of the Washington Office staff, peer groups
have been defined primarily on the basis of asset size of bank
within Region or state.
With the large number of items in the Reports of Condition and
Income and the frequency with which such reports are filed, the
number of ratios that can be constructed for a particular period
or as measures of change between periods is extremely large.
Selection of the key indicator or indicators has consumed a
considerable amount of time at FDIC.

One approach, the Early

Warning System (EWS), examined literally hundreds of ratios to
determine which were the best discriminators between known problem
banks and control groups of banks with no known serious problems.
the result was a winnowing down to 7 ratios, 2 based upon income
and expense items and 5 based upon balance sheet items.
run annually to produce a list of

EWS is

danks whose seven ratios indicate

the similarity to banks with known problems.

A second approach

(JAWS) selected 6 ratios (plus an additional 2 .oI large banks)
III-36

APPENDIX III

APPENDIX III

which have proven to be indicators of basic changes in a bank's
operations.

These indicators have been incorporated into on-

line system available in the Regional Offices which flags banks
with ratios atypical of peer group averages, and displays five
important ratios based upon the latest report of examination
of each of these banks.
The OCC system includes certain ratios which have been designated
as "key indicators," i.e., that provide the best general measures
of unusual or changed circumstances in a bank.

The process is

sequential in that analysis of banks with atypical values for
key indicators is extended to additional financial ratios that
round out the picture of a bank's condition in the critical area.
c)

A third essential element of a monitoring system is the development of a method for evaluating its effectiveness or results.
The crux of the monitoring systems is the review of the output
of the computer based systems by trained financial analysts and the
FDIC has been using experienced examiners in this important function
who have flagged "watch lists" of banks which should be examined
earlier or more often than other banks.

In the final analysis,

however, no monitoring system has yet been developed which is
100% efficient in signaling banks with unusual problems.
Thus, some flagged banks turn out, on further analysis, to be
perfectly sound while some banks with serious problems are not
flagged.

Presently, the most any system does is suggest that a

bank examination should be scheduled and the aspect or aspects
of a bank's operation which requires special scrutiny.
III-37

APPENDIX III
d)

APPENDIX III

The fourth or final element of a monitoring system is implementation.

At the present time, the monitoring exercise leads

up to an examination of banks singled out by the financial
analysts.

Optimally, the examiner receives a profile of the

bank to be examined and a blueprint of the areas to be focused
on with the most care.
The GAO report

states, in essence, that the FDIC has recently

established trust examiner specialist positions.Although the FDIC
historically was the only one of the three federal bank regulatory
agencies that did not designate trust examiner specialists as such,
some FDIC bank examiners devoted a major portion of th<~ir examining
time to trust work.

However, it is correct that the FDIC has now

established 14 trust examiner specialist positions and is in the process
of filling these positions.

III-38

APPENDIX III

APPENDIX III

Recommendations (page 7-25)
We recommend that the Comptroller of the Currency invite FDIC and FRS
to jointly evaluate its new examination approach.

We further recom-

mend that, in the event of a favorable assessment of the new process,
the Board of Directors, FDIC, and the Board of Governors, FRS revise
their examination processes to incorporate the concepts of OCC's approaches.

FDIC Response
In light of the limited testing that has been conducted (10 banks) of
the OCC's new process, we believe it is premature to consider that
process a success either for large or small banks.

Representatives of

the OCC admitted that, while the new procedures are workable in banks
with assets between $50 million and $1 billion, they do not appear
feasible for banks with assets of less than $25 million.

We therefore

questicn the logic and wisdom of GAO's recommendation that FDIC adopt
such process, either for the large or small banks under our direct supervision, especially when it is recalled that 91% of the banks we directly supervise have assets of less than $50 million and 77% less than
$25 million.

Since the number of large banks directly supervised by

the FDIC has and continues to i;lcrease,

our examination process is

necessarily designed to handle small, medium and large-sized banks.
However, we shall follow closely OCC's experience with the new examination process as it undergoes further testing, and we remain receptive
to further revision in our own examination approach which will be beneficial to and improve our supervisory capabilities.
III-39

APPENDIX III

APPENDIX III

In our judgment, the discussion of changes in FDIC's examination approach does not reflect sufficiently the impact and significance of
those changes, especially with respect to our review of the management
policies and internal controls of a bank under examination.

We believe

that the changes made by the FDIC represent, at the present time, the
most logical, beneficial, and prudent improvements in the examination
process.

We have blended the proven techniques and practices with a

new approach which we feel should enable FDIC to focus more directly
on, and devote more time and effort to, problem and near-problem situations, and concomitantly less on healthy banks.

We refer to excerpts

from our General Memorandum #1, included with our general comments.

III-40

APPENDIX III

APPENDIX III

25
Recommendations (page 7-48)
We recommend that the Board of Directors, FDIC, the Board of Governors,
FRS, and the Comptroller of the Currency jointly staff a group to analyze shared national credits at State and national lead banks under Federal supervision and that the three agencies use the uniform classification of these loans when they examine the participating banks.
We recommend that the Board of Directors, FDIC, the Board of Governors,
FRS, and the Comptroller of the Currency work together to refine their
monitoring systems and their approaches to examining for compliance
with consumer credit laws.

FDIC Response
Although--as the GAO report points out--of the 183 participations in
shared national credits traced by GAO only 19 were to state nonmember
insured banks, the FDIC is now a participant in the Shared National
Credits Program.

We are, of course, in favor of the three federal bank regulatory agencies sharing and working together in the important area of consumer
credit compliance.

However, in many instances healthy competition in

the area of consumer credit compliance as well as in other areas of
banking supervision between the three federal bank regulatory agencies
can lead to a better system of supervision than complete uniformity.
Thus, the development of an independent approach by one or more of the
agencies may lead to a better end result.

II111-41

APPENDIX III

APPENDIX III
CHAPTE- 8

Our comments here cover the general theme of the chapter,
namely the supervisory and enforcement practices of the three
agencies and their "problem bank" criteria.

We feel discussion

of these vital subjects is desirable in order to place them in
their proper perspective.
The FDIC serves the dual roles of bank supervisor and insurer.
Accordingly, the FDIC "problem bank" definitions are based on those
banks which pose the greatest degree of financial risk to the
Corporation, with fine tuning of the designations into various
gradations of risk.

The three problem bank categories used are

analagous to the three ac.verse classifications of Substandard,
Doubtful, and Loss which the federal bank regulatory agencies
utilize to designate assets of greater than normal risk.
There are no simple mechanical formulae that can be universally
applied to determine whether or not an operating bank warrants FDIC
problem bank status.

Indeed, we believe a problem bank designation

should only be imposed on a case-by-case basis after a comprehensive,
Among the more important ele-

in-depth analysis of the entire bank.

ments requiring analysis and evaluation are asset quality and liquidity, the margin of capital protection, the degree of stability or
volatility in the bank's liability structure, the character and ability
of its management, the bank's earnings performance, and its adherence
to applicable laws and regulations.

These elements are closely inter-

related and, depending on the circumstances, each element may be

III-42

APPENDIX III
weighted differently.

APPENDIX III
Accordingly, the FDIC disseminates general

criteria--not specific guidelines--for the designation of problem
banks to our Regional Directors in order to encourage independent
judgment and provide some flexibility to meet the new areas of
regulatory concern as they arise.

The Washington Office of the

Division of Bank Supervision then applies more standardized analysis
and evaluation to the recommendations of our Regional Directors before
determining whether to add a particular bank to or delete it from
our list of problem banks.

The listing themselves are not subject

to approval by the Corporation's Board of Directors, although the
Board regularly receives extensive information about all problem
banks and may be directly involved in the imposition and enforcement of a corrective program with respect to particular banks.
FDIC also reviews examination reports of the FRS and the OCC,
assesses the risk exposure which the banks examined by those agencies
pose to the deposit insurance fund, and, where appropriate, designates
state member and national banks as Other Problem, Serious Problem or
Serious Problem-Potential Payoff.

Although FDIC does not directly

supervise these banks, we do follow closely the supervisory efforts
of the other agencies, largely because of our financial stake in the
outcome.
It should be noted that, with respect to banks under the direct
supervision of the FDIC, an

4 iferior

financial condition is not the

sole cause for more intense supervisory activity.

III-43

Causes for concern

APPENDIX III

APPENDIX III

may be reflected in violations of laws or regulations, marginal
management and policies, or a st'-par financial condition which had
not yet reached a level presentin6 an undue risk to the FDIC fund,
and thus, does not warrant a formal prolbem designation.

For example,

the various Regional Offices maintain informal listings of banks
which pose supervisory--but not financial--problems, an(' the Washington Office uses a computerized screening device which serves as an
additional test for uncovering financial as well as non-financial
supervisory problems.
The fundamental approach of FDIC to banks exhibiting supervisory problems or trends in that direction is to exercise preventive measures, that is to take necessary and appropriate measures
early enough to keep the bank from deteriorating to a level requirilLg the assignment of a formal problem designation.

As the GAO

report points out, informal methods are generally relied upon, and
experience indicates that these methods have largely been successful.

One of the more useful methods of informal supervision which

FDIC frequently employs has been effectively overlooked or ignored
in your report,

The method we refer to involves the use of the

so-called "Letter Agreement."

The Letter Agreement is used by our

Regional Directors following an examination to confirm with bank
directors a program which the Regional Director feels will, if adhered
to, correct the situation.

The Letter Agreement is not intended, and

is not used, as a substitute for a formal written agreement entered

III-44

APPENDIX III

APPENDIX III

into under Section 8 of the Federal Deposit Insurance Act or cease
and desist or termination of incurance proceedings, although the
letter agreement may serve as a basis for such subsequent action.
Its use is generally confined to corrective measures agreed to by a
bank's board of directors when a bank first shows problem or near
problem characteristics.
In addition to the foregoing, the GAO report discusses the use
of termination of insurance proceedings and states, in part, that
canceling "a bank's deposit insurance does not solve its problems."
While this statement is perhaps literally true, it could be misleading.

Termination of insurance authority has, through the years,

proven to be an effective and useful remedial enforcement tool.

The

threat of instituting and the institution of such a proceeding has,
in the vast majority of cases, been the vehicle for forcing a recalcitrant and/or poorly managed bank to take effective corrective
measures.
The table in the report dealing with the GAO sample of 54 FDICsupervised problem banks shows that a request for a formal response
to reported deficiencies was mrde -n 44

of the cases, that progress

reports were requested in 41%, that a meeting with the bank's directors
was requested in 30%, that there were written communications with 54%,
6% of the banks were visited, and that no credit is given for special
examinations.

We do not believe that the table presents an accurate

picture of FDIC supervisory efforts.

In point of fact, the Regional

Director transmits a letter to each bank, reiterating the problems
III-45

APPENDIX III

APPENDIX III

disclosed, and requesting appropriate corrective efforts.

Frequently,

in the letter to bank management, the Regional Directors request periodic progress reports which often lead to other exchanges of correspondence or meetings with respect to progress, or lack of it, shown in
the reports, a board meeting may be scheduled, or visitations or a
follow-up examination may be held.
We note that the table shows that meetings with directors were
held in only 30% of the banks.

It is the FDIC's policy to have a

board meeting in all problem situations.

As indicated previously,

in 1975, conferences were held with the management of banks on approximately 1,745 occasions, and 1,750 in 1976.
The GAO report, among other things, questions whether banks that
remain on the problem list for a period of time are indeed problems.
At year-end 1974, 76% of the FDIC supervised banks on our problem
list had been on the list for less than two years, and at year-end
1975, 82%.

In addition, at year-end 1975 only 16 banks had been on

the problem list in excess of three years (out of a total of 8,925
FDIC-supervised banks.)

To summarize, some form of formal supervisory

action: was taken in seven, or 44%, of the sixteen cases and informal
supervisory actions achieved improvements in another seven, or 44%,
of the sixteen cases.

Correction of the problems in the remaining

two banks is to a large degree dependent upon improvement in the
severely depressed economy of the banks' market area.

In calendar

year 1976, two of the sixteen banks were rehabilitated and, since they

III-46

APPENDIX III

APPENDIX III

no longer warranted problem designation, were removed from the FDIC
problem list.

In addition, two others were removed from the list--

one through merger into a healthy institution and the other was closed.
We also note in passing that a limited number of banks may present
financial and/or supervisory problems of a continuing nature which,
despite aggressive corrective efforts, do not lend themselves to a
permanent and wholly acceptable solution.

In such cases, the banks

are not in serious enough condition to warrant either termination of
their insured .status or of their charter.

It seems clear, however,

that these banks should be continued as problem banks and receive
special supervisory attention.

III-47

APPENDIX III
Recommendations

~~~~~~~APPENDIX

~APPENDIX
III

III

(page 8-")

We recommend that the Board of Directors, FDIC, the Board of Governors,
FRS; and the Comptroller of the Currency establish more aggressive policies for using formal actions.

Written guidelines should be developed

to identify the types and magnitude of problems that formal actions
could appropriately correct.

FDIC Response
Congress granted cease and desist powers in 1966 with the enactment of
Section 8(b) of the Federal Deposit Insurance Act.
thereafter, there was some

For several years

_luctance to utilize Section 8(b) powers

due mainly to a general misunderstanding of its purpose and usefulness.
Prior to enactment of Section 8(b), the FDIC's only experience with
formal administrative corrective measures was the termination of insurance proceedings, a severe form of action which could result in the
removal of the deposit insurance coverage of a bank.

Because of its

severity, the Section 8(a) proceeding was used judiciously and only
after all other means for accomplishing correction were exhausted.
Apparently, albeit erroneously, that same rationale was largely applied
to Section 8(b) proceedings.

In addition, there was to a lesser extent

an unwillingness to try something new.

Commencing in 1970, a program

to educate FDIC personnel as to the usefulness of Section 8(b) action
was begun.

The FDIC first used its cease and desist authority in 1971

and between 1971 and 1975 issued 38 cease and desist orders and three
formal written agreements.

In contrast, in a recent renewed effort to

III-48

APPENDIX III

APPENDIX III

foster the use and to test the effectiveness of cease and desist powers,
in calendar year 1976 alone FDIC issued 24 such orders and five emergency orders.

In addition, at year-end 1976, 18 more cease and desist ac-

tions were in various stages of process.

While cease and desist action is in most cases effective as a corrective measure, there are some instances where it may be of little or no
use and could perhaps be counterproductive.

For example, the recently

_xperienced worst economic period since the great depression caused
severe problems to the banking industry, many of which did not lend
themselves to correction through use of the cease and desist powers.
In short, it is not a panacea for the removal of all pru -ems

experi-

enced by the banking community.

The recommendation for ador+-U n of criteria for use in formal actions,
contained in the last sentence of the recommendation, is troublesome.
We would recommend against adopting formal criteria for use of Section 8, because the statutory criteria are adequate.

The facts and

circumstances of bank problems seem so varied, and the remedial actions
can differ so much according *:o the problem, it would be inhibiting to
have to work within the confines of additional written criteria.

The

adoption of such criteria could give the banks additional bases for
contesting Section 8 actions.

III-49

APPENDIX III

APPENDIX III
49
Recommendation (page 8-u)

We recommend that the Board of Directors, FDIC, the Board of Governors,
FRS, and the Comptroller of the Currency develop uniform criteria for
identifying problem banks.

FDIC Response
We believe our general comments clarify the posture of the three federal bank regulatory agencies with respect to problem banks, including
those which pose supervisory problems as well ab those which present
inordinate financial risk to the FDIC.

Moreover, we do not believe

there is confusion or wide disagreement among the bank regulatory agencies as to which banks shaould be accorded close surveillance and supervision by the respective agencies and that, except in a failing bank,
and to a limited extent in a bank holding company situation, there is
virtually no overlap of regulatory jurisdiction at the federal level.
Furthermore, the need to develop common criteria for problem banks is
not obvious and indeed may not be appropriate.

It is, we believe, appropriate and useful for the FDIC as an insurer
to view what constitutes a problem bank from a somewhat different perspective than the other two federal bank regulatory agencies.

In ad-

dition, the extent to which the three federal bank regulatory agencies
use somewhat different approaches to the issue of banks in need of increased and intensified supervision could foster a greater degree of
innovation in this area of supervisory endeavor and could serve as a

III-50

APPENDIX III

APPENDIX III

check and balance in the promotion of the widest coverage of such
banks.

Finally, the objectives and detached review process conducted

by FDIC of all types of examinations, in order to assess the degree of
financial exposure to the insurance fund, provides an overall review
of all banks without imposing across-the-board guidelines which may
not be suitable for the three agencies on an individual basis.

III-51

APPENDIX III

APPENDIX III
CHAPTER 9

The GAO report states:
"A recent FDIC study of 92 banks that failed between
1960 and September 1976 showed that 57.6 percent were
caused by improper loans to officers, directors or owners, or by loans to out-of-territory borrowers...."
The experience noted above led to the issuance by FDIC in
1976 of a regulation entitled "§337.3 Insider Transactions," as
part of the Corporation's "Unsafe and Unsound Banking Practices"
regulations.

III-52

APPENDIX III

APPENDIX III
CHAPTER 10

The following is a brief summary of the operation of the FDIC
Training Facility:
The Division of Bank Supervision (DBS) Training Center located in
Rosslyn, Virginia, was established in February 1970 and presently has a
permanent staff complement of seventeen.

The training programs which it

conducts are directed toward achieving professional proficiency and the
maintenance of a highly qualified bank examination staff.
Career training is accomplished, in balance with field examination
experience, primarily through our bank examination schools which are
comprised cf seven different schools or courses of study.
has a duration of two to three weeks.
examination schools:

Each school

Subject schools include the basic

School for Assistant Examiners, designed for newly

employed examining personnel;

the School for Senior Assistant Examiners,

which provides training in accounting, EDP, and consumer protection laws;
the School for Examiners, which is designed for the development of the
commissioned examiner; and the Basic Trust School, which deals with the
basics of trust department examination.

In addition, more senior train-

ing is provided through the Advanced Trust School, the Course in Examining
a Computerized Bank, and the School for Commissioned Examiners.

Subject

matter within the various bank examination schools is well structured
both with respect to material to be covered in the daily preson-;.tions
as well as the pre-course study expected.

Students ordinarily spend

eight hours a day Monday through Friday in classroom and related work.

III-53

APPENDIX III

APPENDIX III

During the six-year period 1970 through 1975, 189 school sessions
were held involving nearly 5,100 students.

For the 1976 school year,

we held 46 school sessions with approximately 1,200 students attending.
Training is directed primarily towards FDIC personnel.

However, during

the period 1970 through 1975, training was provided for 549 state bank
examiners, 28 students nominated by foreign government banking authorities,
and 13 FRB examiners.

The related figures for 1976 are 157, 22, and 7,

respectively.
An additional and important operation of the Training Center is the
Progress Evaluation Program for senior assistant examiners who are being
considered for career advancement to the status of commissioned examiner.
This program assesses a candidate's knowledge and proficiency in rules,
regulations, and policies; loan analysis; and development of conclusions
and recommendations after review of a repoit of examination.
includes both written and oral portions.

The program

Findings of the progress evalu-

ation are weighed as one of several factors in considering a senior
assistant examiner for promotion to commissioned examiner status.
Between 100 and 200 such candidates are evaluated annually.

The eval-

uation utilizes a three-member panel of examiners over a three-day period
for each candidate.

II11-54

APPENDIX III

APPENDIX III
Recommendation (page 10-6)

We recommend that where feasible the Comptroller of the Currency; the
Board of Directors, FDIC; and the Board of Governors, FRS, combine
their examiner schools and standardize their curriculums.

FDIC Respcnse
Although we find the comments and rrormnendations contained in the
report on examiner training provocative, on balance we believe they
did not afford sufficient treatment or depth to the various examiner
training and educational programs offered by the FDIC.

We are especially dismayed by the fact that the GAO study largely ignores the operation of the FDIC Division of Bank Supervision (DBS)
Training Center.

The FDIC Training Center is undoubtedly the best

bank examiner training program in the country.

Nevertheless, because

of our burgeoning training needs, the FDIC is considering constructing
its own, larger facility with resident dormitory quarters.

The FDIC

has approached the FRS and the OCC to join with the Corporation in a
cooperative training facility.

Both the FRS and the OCC have evidenced

interest in this project and discussions on a cooperative training effort are going forward.

A brief summary of the operation of the FDIC

DBS Training Center is included with our general comments.

III-55

APPENDIX III

APPENDIX III
Recommendations

(page 10-10)

We recommend that the Board of Governors, FRS,

(1) establish a full-

time training of ice to operate its examiner training program and (2)
carry out the revision of examiner school curriculums which it has
recognized as needed for sometime.

We also recommend that the CoLptroller of the Currency; the ,3oard of
Directors, FDIC; and the Board of Governors, FRS; increase their training in EDP, law, and accounting, as desired by their examiners.

FDIC Response
We plan to give further attention to this apparent need.

It is worth

pointing out in passing, however, that, at least with respect to EDP
training, in addition to the regular basic EDP courses (Course in
Examin../go a Computerized Bank (CECB) I and II), an advanced eight-week
technical EDP school, known as Field Examiner Advanced Automation
Training (FEAAT), is presently offered to examiners who have a desire
to become highly proficient technically in EDP matters.

III-56

APPENDIX III

APPENDIX III
CHAPTER 11

The GAO report indicates that a cooperative effort among the
federal bank regulatory agencies in the development of monitoring
systems may have "speeded development" and mentions the need for
continued coordination.
Each agency is in the process of developing or has developed
monitoring systems and each has learned from its own experiences.
We agree that there should be coordination among the agencies in
these efforts and would point out that a significant amount of
sharing and exchange of concepts and ideas has already been effected.
However, as pointed out previously, there is also merit to the three
age.icies developing systems independent of one another.

Innovation

is fostered and a healthy competition to have the best system available could be beneficial to all the agencies.

In addition, although

the major objectives of the three federal bank regulatory agencies are
s4milar, there are unique characteristics of each which may render the
development of a system common to all inappropriate.

While we do not

presume to comment on the unique needs of the OCC and FRS, central to
the FDIC's needs is the development of a system to cope with the substantial number of small and medium sized as well as a significant
number of large sized banks under our direct supervision.

Thus, in

the case of FDIC, a system that is sufficiently flexible to meet the
needs of supervising large sophisticated banks, as well as smaller less
complicated banks, is apparently what is required.
III-57

APPENDIX III

APPENDIX III

Of course, adequate staffing and gathering accurate data on a
timely basis are two vital elements in the development, implementation, and maintenance of any monitoring system.

The FDIC is moving

forward in its efforts to satisfy these essential elements.

Finally,

FDIC has, in the main, completed the testing phase of our monitoring
systems and is in the process of integrating them into our examination process.
The GAO report states that an OCC official indicated that the
interagency system for processing bank data was inadequate because
banks were not meeting established reporting deadlines and FDIC
took approximately four months to keypunch and computer-edit the
system.
The FDIC does maintain the bank reported financial data for all
insured banks supervised by the OCC, FRS and FDIC.

Data submitted by

the national and state member banks are initially processed by the FRS
and submitted to the FDIC for edit testing and acceptance into the
finalized data base from which all of these agencies draw information.
The OCC i3 correct in asserting that the FDIC has taken up to four
months to

,rocess all of the reports from some 15,000 insured banks

and to produce a final data base.

However, delays in receipt of

correction data from thie OCC and FRS where edit tests have failed
on banks under their respective supervision Ehee been a major factor
in the finalization of the data base.

III-58

Efforts are being made to

APPENDIX III

APPENDIX III

obtain agreement among the agencies on edit-check criteria so that
corrections can be made on a more timely basis.
In order to meet both the monitoring and other needs dependent
on bank reported financial data, it would, of course, be to the
benefit of all agencies to derive a set of editing criteria which
would produce an acceptable financial data. base with greater alacrity.

III-59

APPENDIX III
Recommendation

APPENDIX III
(page 11-8)

We recommend that either (1) the Board of Directors, FDIC; the Board
of Governors, FRS; and the Comptroller of the Currency jointly establish a more effective mechanism to combine their forces in undertaking
significant initiatives to improve the bank supervisory process or in
attacking and resolving common problems; or (2) the Congress enact
legislation to establish a mechanism for more effective coordination.
We would be glad to assist the committees in drafting appropriate
legislation.

FDIC Response
We recognize the merit of resolving common problems of the three agencies through closer coordination and cooperation.

Indeed, there is at

the present time a substantial exchange of information between the
agencies' headquarters as well as at the field levels.

However, if

there is any merit to the concept of separate federal supervisory agencies, and to a dual banking system with state and federal supervision
of banks, the benefit would seem to be the opportunity to try different
approaches and to have a diversity of examination and supervisory procedures.

The possibility of useful innovation and improvement in the

bank examination and supervisory processes is greater if there are
several agencies trying different approaches than if every change in
examination methodology required approval of all the agencies.

Neverthe-

less, the possibility of establishing a particular vehicle for the agencies to resolve common problems and take joint efforts in new initiatives
will receive serious consideration.

III-60

APPENDIX III

APPENDIX III
FDIC ADDENDUM

We note that the draft GAO report is silent with respect to
the planning and modernization efforts undertaken by FDIC in recent years to keep our supervisory activities abreast of economic,
technical, and social developments.

We have attached a digest of

our planning and implementation of those planning efforts.

III-61

APPENDIX III

APPENDIX III

In 1965, an exhaustive analysis of the examination and supervision functions of the Corporation, similar in many respects to the Haskins and Sells
study of the OCC, was undertaken by a committee of three experienced field
examiners who were detailed to the Washington Office.

Some of the recom-

mendations flowing from that study were:
- Increased emphasis on examination-by-exception techniques with at
least an annual visitation to each bank under our supervision
- Mutual interchange of (non-confidential; data with state banking
departments
- The establishment of effective guidelines for the volume of loans
which should be analyzed it
-Development

given bank

of a program of procedural audits of certain banks and

furnishing audit assistance by the Corporation to some banks upon
request
- Revision of the report of examination to make it more usable to
bank managements and the supervisory functions of the Corporation
- Adoption of recommended policies by the Corporation in regard to
asset reserves, common capital stock, classification of assets, and
utilization of termination of insurance proceedings
- Publication of the Corporation's policies
- Limitation of field investigations of statutory applications to
those which are of significance in respect to competitive and
bank soundness considerations
- Development of more efficient application aid investigation forms

III-62

APPENDIX III
- Streamlining,

APPENDIX III
J expediting of application processing within the

Corporation
- Delegation of authority to the Regional Directors for acting on
certain statutory applications
- Restructuring of the geographic and managerial composition of the
Regions
- Internal revisions designed to follow a specialized, functional
approach promoting better communications and training
- Utilization of automated systems to aid in scheduling examinations,
the review of examination reports, and gathering information in connection with Statutory applications
- Expansion of EDP training programs, and the selection and training
of examiner personnel in the managerial aspects of computer
operations
- Strengthening of requirements for commissioned examiners as well as
revisions of the centralized evaluating process
- Expansion and intensification of training of examiner personnel,
including the executive levels
- Conducting periodic staff meetings to include both Regional and
Washington Office senior personnel
- The interchange of senior examining personnel with other Federal
agencies for short periods of time
- The revision and enhancement of expense allowances for travel and
relocations.

III-63

APPEINDIX III

APPENDIX III

Long range planning programs have been continuous since the 1963 study.
The Proiects and Planning Branch of the Division of Bank Supervision was
established in 1971, and the Board of Directors created the Office of
Corporate Planning in 1974.

Developments at the Corporation within the

past five years or so, flowing from planning efforts, and paralleling
recomrmendations in the Haskins and Sells study, include:
- Implementation of completely revised examination report formats
for commercial banks (late 1969) and mutual savings banks (late
.972)
- Development in 1970 of an extensive training center for our
personnel as well as those of the Federal Reserve, Comptroller,
Sta.e Banking Departments, and some foreign studerts
- Reorganization of the Washington Office of the Division of Bank
Supervision along functional

lines and the addition of a legal

counsel to our Regional structure in 1971
- Extensive revision of the Manual of Examination Policies was
begun in 1972
- New forms for filing and investigating statutory applications were
developed and implemented between 1971 and 1973
- Delegation in 1973 of specifically defined authority to the Regional
Directors for approval of all statutory applications except those
involving mergers and the granting of deposit insurance
- Limitation on actual field investigations of statutory applications
to those situations where Lhe competitive or overall bank soundness
considerations made them necessary

III-64

APPEWDIX III

APPENDIX III

- Dissemination of the Corporation's policy statements and decision
guidelines was begun in 1970
- The development of automated early warning, trend analysis,
consumer loan evaluation, and review-by-exception systems was
initiated in 1971
- Emphasis was substantially increased on training programs and
specialization, particularly in the areas of automation, trust,
and international activities (although FDIC has limited direct
involvement in the international field)
-The

development of guidelines end the initiation of experimentation

with an examination-by-exception program which emphasizes the
evaluation of management and systems
- Study, experimentation, and implementation of statistical sampling
,s part of the examination process
- More widespread application of disclosure requirements in
connection with securities offerings by banks
- The development of a new examination report for trust departments
and a complete revision of the Manual of Examination Policies
relating to trust activities, and selection of Trust Specialists
in order to provide more expertise in this complex area of bank
examination.
Other FDIC planning efforts include reviewing considerations of overlapping regulatory functions resulting in the development of a recommendation for regulatory reform, the experi-aental Selective Withdrawal

III-65

APPENDIX III
APPENDIX III
separate
conducting
in
from Examination Program, and an experiment
compliance examinations, aimed largely at measuring adherence by banks
with consumer-oriented

laws, regulations and policies.

A separate Office of Bank Customer Affairs was created in early 1975 to
oversee a variety of depositor and consumer-oriented functions.

In

addition, a Consumer Affairs Unit within the Division of Bank Supervision
was established in 1971 and continues in operation.

Additionally, our examination staff has been expanded from about 900 in
1960 to approximately 2,000 at year-end 1975, and we expect to add
approximately 150 more examination personnel annually during the next few
years, spaced so as to allow efficient assimilation into our examination
corps,

Considerable effort has been expended on the development of information
systems,

and data contained in Call and Income and Dividend Reports have

been available to the pubiic since 1972.

Along with the development of early warning systems, the Co'ooration has
increased its emphasis on the potential risks to t.Le insurance fund flowing from larger banks, liquidity, earnings performance as an indicator of
overall bank soundness, and failure to use or untimely use of ,!nforcement
measures.

Farly and more detailed review of problem and near-problem

situations at the Board level has led to an expanded review staff, and our
experiences in problem situations prompted issuance of a regulation
gcverning insider transactions ;n bank3 under our direct supervision.
ILI-66