The
principal objectives of a Deposit Insurance
Scheme (DIS) are to contribute to the
stability of a country’s financial
system and protect less financially sophisticated
depositors from the loss of their deposits
when banks and financial institutions
(BFIs) fail. The DIS alone cannot establish
stability of the financial system; it
can only complement a strong regulatory
and supervisory system. Effective bank
supervision and regulation is still the
key for financial stability.
The
DIS is preferable to an implicit protection
if it can meet depositors’ claims
at agreed levels. In addition, DIS limits
the scope for discretionary decisions
that may result in arbitrary action by
governments and regulatory authorities.
By their very nature, DIS systems entail
moral hazard. To minimize moral hazard
and to be credible, DIS systems need to
be properly designed, well implemented
and understood by the public.
There
is a case for establishing a DIS in Sri
Lanka. Due to lack of adequate membership,
the current voluntary deposit insurance
scheme (VDIS) does not have a sizeable
deposit insurance fund to meet any contingencies
if such a need arises. Given the unsuccessful
experience of the VDIS, the proposed DIS
should be made mandatory to deposit taking
financial institutions that come under
the purview of the Central Bank of Sri
Lanka (CBSL), with possible extension
to other deposit taking institutions in
the future.
The
CBSL should take the lead in designing
a mandatory deposit insurance scheme (MDIS),
but it should not be involved in implementing
the system. CBSL’s involvement would
result in an overlap of supervisory and
regulatory procedures and further enhance
moral hazard of the member institutions.
The CBSL, government, deposit taking institutions
and all other stakeholders should start
a dialogue by addressing policy and operational
issues, such as the organizational structure,
mandates, powers, funding and coverage
etc., prior to the setting up of the insurance
scheme. It is also critical to address
explicitly inter-relationship issues among
regulators and other deposit protectors
by specifying areas of work and agreeing
to exchange necessary information. Given
the fact that Sri Lanka has a mixed banking
structure of state, private and foreign-owned
banks, it is important to assess the risks
of different BFIs and attempt to formulate
a risk based premium system for the proposed
DIS. If it takes time to establish a risk
based premium system, the authority should
begin with a flat rate premium and move
towards a risk based premium system after
a few years of operation. The characteristics
of the system, its benefits and limitations
should be published regularly to maintain
the credibility of the system. It takes
some time to set up a DIS, given the need
to enacting laws. In the meantime, a CBSL-led
team together with other stakeholders
could start preparatory work.
The DIS can be justified on public policy
grounds as providing protection to those
least able to protect themselves in a
society. The public policy takes into
consideration that the less affluent in
society should be looked after because
their savings represent the best hope
for their future well being. To inculcate
the habit of saving among the less affluent,
they should be assured that the institution
to which they hand over their savings
is safe and sound. It is also considered
that the less affluent and poorer segments
of society are not in a position to obtain
information or assess the financial condition
of deposit soliciting institutions by
themselves. Private sector institutions
are not keen to provide deposit protection.
Experience, the world over, has proved
that DIS has been effective in preventing
a large number of bank failures and losses
of savings of the public. The number of
explicit DIS schemes implemented by various
countries has risen from 12 in 1974 to
approximately 75 by 2002.
In
addition to public policy, the safeguarding
of the less affluent saver would bring
in public benefits. First, the DIS provides
incentives to savers. Secondly, they replace
ad hoc approaches to banking problems,
crisis resolutions and discriminatory
practices that are adopted by central
banks and governments. Thirdly, DIS would
enhance public acceptance of and overall
confidence in, the financial system. These
public policy benefits do not come automatically.
The authorities should prepare a conducive
environment in which DIS could work effectively
and it is dependent upon the establishment
of healthy and competitive banking practices,
well established licensing and supervisory
oversight by the regulatory authorities.
A few central banks are involved in DIS.
They take part in setting up DIS; have
shareholdings; and appoint directors to
DIS boards. However, many central banks
do not implement DIS due to a variety
of reasons. The central banks, on the
other hand, should be concerned with the
large volume of liquid funds that could
go back to the public as and when insurance
claims are settled. Central banks should
have their own methods of mopping up this
liquidity from the system, which is a
monetary policy concern.
This
paper consists of 5 sections. The introduction
provides briefly the objectives and uses
of a DIS. Section 1 deals with the financial
landscape in Sri Lanka and the present
supervisory and regulatory system. Section
2 deals with the existing VDIS, while
Section 3 sets out the main objectives,
key features and the key principles of
the proposed DIS. This section draws heavily
on the US Federal Deposit Insurance Corporation
(FDIC), discusses its relevance to Sri
Lanka, and highlights common problems
in the design of DIS. Section 4 sets out
the preparatory work required in setting
up a DIS. The concluding section summarizes
the findings of the paper.
The
licensed commercial banks and specialized
banks dominate the financial sector in
Sri Lanka, accounting for about two thirds
of the total financial assets. Annex Table
1 summarizes the size of the deposit taking
institutions. At the end of 2002, there
were 23 commercial banks (i.e.12 foreign
banks and 11 private commercial banks),
13 specialized banks, including the National
Savings Bank (NSB), two large development
banks (i.e. the National Development Bank
and DFCC Bank) and 6 regional development
banks. There were also 26 finance companies
registered with the CBSL. All these institutions
are supervised by the CBSL, while the
merchant banks, investment banks, insurance
companies, venture capital companies,
unit trusts and housing finance institutions
are not under the supervision of the CBSL.
In addition, there are a large number
of micro finance institutions, which solicit
public deposits in various forms. While
the Insurance Board and the Securities
and Exchange Commission are overseeing
some of these institutions, many others
are not effectively supervised by any
organization. The Cooperative Commissioner
registers some of the micro finance institutions
but does not conduct any effective supervision.
Having had a crisis in the 1980s, the
CBSL is keeping a close eye on the registered
finance companies which are operating
on a relatively small capital compared
to the commercial and specialized banks.
The Bank of Ceylon (BOC), the Peoples
Bank (PB*), NSB and the State Mortgage
& Investment Bank (SMIB) are wholly
owned public sector banks. The state banks
accounted for 30 percent of all deposit
taking institutions and 34 percent of
GDP at end of 2002. These public sector
banks together with three retirement benefit
funds and mortgage institutions accounted
for 55 percent of the financial system’s
assets and 62 percent of GDP in 2002 (annexed
Table 2). Despite considerable progress
during the past two decades to reduce
the oligopolistic characteristics, the
two state banks still dominate the banking
sector while state sector institutions
dominate the financial sector.
The
banking sector was adversely affected
by the worldwide economic down turn in
2001. The reported levels of non-performing
loans (NPL) of banks increased in the
recent years and currently, the industry
average of NPL in total loans and advances
is around 17%. On average, the provisions
cover around 40% of total gross NPLs of
all banks, but the private banks have
managed to keep them at an average of
16%. The foreign banks have relatively
low ratios of NPLs. Several banks continue
to operate with low levels of provisions.
In terms of prudential ratios, not only
state banks, but also some of the foreign
banks have to bring up their prudential
ratios up to accepted levels. The average
NPLs of finance companies was around 12%.
The financial sector has strengthened
significantly over the last two decades.
At present, the CBSL is the sole authority
of supervising and regulating BFIs. The
supervisory mechanism has been upgraded
largely in conformity with the Basle core
principles of bank supervision. The Bank
Supervision Department (BSD) of the CBSL
is dedicated to carry out regulatory and
supervisory responsibilities of banks
within the provisions of the Banking Act
and the Monetary Law Act. The Non-Bank
Supervision Department of the CBSL oversees
the operations of the finance companies
under the Finance Companies Act.
The
supervisory and regulatory functions of
CBSL encompass the issue of prudential
regulations such as capital adequacy,
single borrower limits, lending and investment
concentration, liquidity ratios, provisioning
for bad and doubtful debt, audit criteria
and directions, etc. BSD also conducts
on-site and off-site examinations. Most
off-site examination reports are based
on comparative compliance of analysis
of CAMEL components while on-site examinations
look into the details of CAMEL ratios
and compliance with other prudential requirements.
At present, most banks adhere to the minimum
risk weighted capital adequacy ratio of
8 percent but for prudence, the CBSL has
announced that banks should increase capital
adequacy to 10 percent from 2003 onwards.
The CBSL is also the authority for issuing
bank licences with the approval of the
Minister of Finance. The minimum capital
for commercial banks is Rs 500 million
and Rs 200 million for specialized banks.
The required capital of a finance company
is Rs. 100 million. The ownership of any
bank by one individual or group is limited
to 10 percent of its shares, unless an
exemption is granted in accordance with
the provisions of the Banking Act. The
CBSL also manages the VDIS and has taken
the initiative to enhance public awareness
of the banking facilities and their interest
rates, charges and commissions by publishing
such information in a consolidated form.
15 finance companies failed in the 1980s.
Some were liquidated and others were merged
or acquired. Pay out to depositors has
been uneven and the liquidation process
has not been completed on one of the large
finance companies. In recent times there
have been two bank failures in Sri Lanka
– one was the Colombo Branch of
the BCCI. In this instance, there was
no loss to depositors. The other was the
Pramuka Savings & Development Bank.
Its licence was cancelled by the CBSL
due to fraud, mismanagement and negative
net worth. Being a licensed specialized
bank, Pramuka has not passed any systemic
threat to the banking system but hardship
to depositors.
Despite these, public confidence in the
banking system is still high partly because
of the government ownership of up to 55
percent of deposits taking and retirement
benefit institutions. The NSB Act of 1971
exclusively protects deposits of the NSB,
while the state banks have an implicit
guarantee on deposits.
The existing DIS in Sri Lanka is a voluntary
scheme, which has not achieved sufficient
membership and scope of coverage to make
it a viable programme. The present DIS
was introduced in March 1987 under the
auspices of the CBSL, the purpose of which
was to provide some protection to small
savers and account holders. Section 32A
of the Monetary Law Act (MLA) specifies
that every insured bank or a cooperative
society shall be liable to pay a premium
to CBSL and that premium shall not exceed
15 cents per annum for every 100 Rupees
of the total amount of deposits in that
bank/cooperative society. Accordingly,
the CBSL has set rules and regulations,
definitions, fees and the basic entry
and exit criteria. From the inception,
the scheme was open to a variety of deposit-taking
entities, namely, banks and deposit-taking
cooperative societies. The individual
deposit insurance coverage was set at
Rs.100, 000/-.
One of the drawbacks of the VDIS was that
it lacks a sizeable deposit insurance
fund to meet contingencies and the ability
to pass on significant costs, should the
insurance fund be called upon to meet
claims by depositors. The DIS commenced
operations without a contribution of initial
startup funds to form an insurance reserve.
Although it meant to build up the insurance
fund over time, this has not materialized
mainly due to inadequate membership.
Initially, 13 financial institutions joined
the VDIS, but the number has declined
over time. As at present, there are only
3 multi-purpose cooperative societies
in the DIS and this membership is insufficient
to generate adequate financing to build
up a viable insurance fund capable of
meeting insurance obligations. By way
of premium, the insurance fund had Rs.1.4
million and the size of the VDIS was Rs.140
million at the end of 2002, which includes
Rs.50 million of equity by CBSL. The members
who joined the VDIS were relatively small
deposit-taking institutions. All big banks,
including the two state banks, which account
for 21% of the financial system are out
of the VDIS. This highlighted the fact
that state banks ride on the government
ownership, which has led to the public
perception of an implicit guarantee of
their deposits by the government.
Given
the need to establish safety nets for
depositors and restore public confidence
close on the heels of a collapse of a
specialized bank, it is considered necessary
that a workable mandatory DI scheme be
introduced. When a Mandatory Deposit Insurance
Scheme (MDIS) is introduced, by returning
simultaneously, the funds to the government
and the CBSL, the existing VDIS can be
dissolved. Alternatively, the insurance
reserve of approximately Rs.80 million
can be transferred to the new MDIS.
The basic insurance concept should be
simple and direct, if it is to provide
safe, secure liquid assets to individuals
and small savers. The DIS should provide
an assurance that it will return at least
a part of depositors’ funds if a
depository institution were to fail, become
illiquid or is closed. For the system
to work successfully, there should be
the widespread public belief that funds
would be available to satisfy the insurance
scheme. In this regard, it may be worthwhile
for Sri Lanka to take a close look at
the oldest and well-known US DIS system,
operated by the Federal Deposit Insurance
Corporation (FDIC) and learn from its
design, operations and drawbacks, and
assess its applicability to Sri Lanka.
The following sections would draw heavily
on the US DIS system while advocating
the adoption of the scheme, wherever possible.
The US introduced the Federal Deposit
Insurance Scheme in 1934 with the broad
objectives of: (1) enhancing macroeconomic
and financial stability by mitigating
or preventing bank runs; (2) protecting
the small depositors from loss if their
banks fail. The same objectives would
be applicable to Sri Lanka.
Since the advent of the US Federal Deposit
Insurance in 1934, bank failures have
become much less frequent and widespread
banking panic in the US has disappeared.
Although there have been other important
factors that have helped the US macro
economy to be more stable, DIS in general
has created beneficial economic externality
as the financial system and the macro
economy have been less volatile. During
2002, 11 FDIC insured institutions failed,
of which 10 with combined assets of $2.5
billion were insured by the Bank Insurance
Fund (BIF). Losses for the failed institutions
are estimated at $ 630 million. Over 90
percent of financial institutions pay
essentially nothing for deposit insurance.
This compares well with failed and assisted
1,617 banks through 1980-1994 for which
FDIC has effected straight deposit payoffs.
FDIC had 135 problematic institutions
to handle as at the end of 2002 (Annexes
3a & 3b).
Sri Lanka’s record of bank failures
has been insignificant compared to most
other countries. Although the two state
banks’ efficiency ratios are far
from satisfactory, their prudential ratios
have improved following the 1993 and 1996
large scale restructuring. For better
management and efficiency reasons, options
for rehabilitating the Peoples Bank are
under consideration. More recently, the
CBSL’s decision to suspend the licence
and liquidate the Pramuka Savings Development
Bank - a licensed specialized bank has
been a surprise to depositors who have
filed several cases in court challenging
the CBSL’s decision. There have
been several finance company failures
in the 80s of which the CBSL met part
of the deposit liabilities in an ad hoc
manner. Although the licensed banks and
registered finance companies are under
close supervision of the CBSL, there are
underlying vulnerabilities that could
surface from time to time causing threats
to the stability and confidence of the
financial system. Therefore establishing
a MDIS as a precautionary measure would
be desirable.
This goal can be realized in two ways
(a): by guaranteeing depositors so that
they will not suffer losses even if their
banks fail; and (b) by reducing contagion.
The existence of a comprehensive DIS should
avoid bank runs even if the depositor
learns that another bank has failed. A
transparent scheme whereby the MDIS would
meet depositor liabilities up to an agreed
level would no doubt make the public and
other stakeholders aware of what each
party would get in case of a collapse
of a bank or a financial institution.
The
sharp reduction in bank runs would also
reduce systemic risks. This was proved
in the US when bank failures in the early
1990s did not precipitate many bank runs.
Banks have failed because they were insolvent
and not because they became illiquid in
the midst of panic runs. The important
principle that needs to be well established
is that the insurance product is only
a guarantee on deposits but not a guarantee
of troubled financial institutions.
Any DIS would create some degree of moral
hazard. The DIS should however be designed
to minimize exhaustive risk taking, i.e.
additional risk taking induced by DIS
itself. Similarly, if the DIS should not
favor one type of bank to another; one
category of bank deposit for another;
or one type of risk taking behavior for
another. Although, in practice, it is
difficult to observe these, the scheme
should be as neutral as possible. At the
start of the scheme, it should be made
clear the differentiation, if any, to
enable depositors to make their choice.
The government or the DIS authority should
neither tax the industry by charging too
much for DIS nor subsidise it by charging
too little. In the US, the FDIC observes
this tradition. DISs, in general, make
the economy less risky for businesses
and this justifies a subsidy. The element
of subsidy, however, would result in the
government selling DIS at a loss and it
should be minimized as far as possible.
If all member institutions pay a steady
premium over a long period, it would result
in the insurance fund fluctuating in response
to insurance losses. The fund will be
dependent on government or other financial
institutions’ support as and when
funds are inadequate, which would multiply
the subsidy element. A periodic change
of the premium should be included in the
system. This is a critical policy area
that requires extensive discussions by
the Government, CBSL and BFIs, as the
MDIS should not be an undue burden either
on the Government or banks. Having empowered
a central bank or regulatory authority
to effectively supervise BFIs, some governments
can be reluctant to be involved in the
ultimate bailouts of depositors. Yet,
governments, in the national interest,
would be responsible for ensuring financial
system stability and for that, they should
take part in DI schemes. The initial amount
of funding of the DIS is debatable in
Sri Lanka given the present budgetary
conditions. To avoid an unnecessary burden
on the budget, all relevant stakeholders
should discuss policy, operational and
funding issues at an early stage. It may
be worthwhile to explore the possibility
of using some of the concessionary foreign
funds, if available, or discuss with international
institutional investors such as the ADB
and IFC for equity participation.
In a publicly run DIS, the government
uses taxpayers’ funds which serve
as the ultimate re-insurer. For example,
when the savings and loans (S&L) debacle
emerged in the US in the 1980s, the taxpayer
was presented with a multi-billion dollar
bill. The FDIC changed this structure
and arranged for banks to recapitalise
the insurance fund whenever it fell below
1.25 percent of insured deposits thereby
avoiding the liability of insurance loss
for the taxpayer. However, in reality,
one cannot avoid a small residual risk
falling on the taxpayer. Attempts should
be made to remove this potential exposure
as far as possible, except in situations
where natural disasters, health disasters,
etc. take place, which are taken as axiomatic
as the government will in such instances
serve as the ultimate backstop. Further
reducing taxpayer exposure to zero de
facto, as opposed to de jure, is probably
unrealistic. If government were to provide
equity capital and to continually fund
the MDIS, then it will become a government
institution and this will have serious
implications on the taxpayer.
Although it is argued that the taxpayer
should monitor his bank and withdraw funds
from unsafe institutions, in reality it
is not possible in most instances. If
the depositor were to do his job well,
it is necessary for the taxpayer/depositor
to get first-hand information on the viability
of BFIs. The central banks or other regulators
would be reluctant to issue information
relating to viability of individual BFIs.
If DIS were meant for depositors and not
for troubled financial institutions, then
regulators would have to ensure that they
control and supervise BFIs to establish
public confidence in the financial system.
Moreover, in developing countries, neither
depositors nor the general public have
the knowledge or the ability to supervise
or monitor the institutions in which they
place their funds. The set-up costs of
monitoring a bank by an individual are
not economical either. Against this background,
the authorities may not be able to hold
depositors responsible for monitoring
their own banks. An effective scheme of
government-funded DI would relieve small
depositors of this burden.
The S&L debacle that took place in
the 1980s was a significant test for DIS.
Why so many of them actually failed was
not very clear, although the common belief
is that inadequate supervision and regulation
of savings and loans was the main reason.
It is also not clear whether all S&L
had to be rehabilitated by the FDIC and
the US government. Currently, all depositors
in the US are accustomed to seeing FDIC
guaranteeing that their bank deposits
are safe. A lesson from the US S&L
would be that the regulators should act
swiftly as symptoms of possible failure
of BFIs surface rather than allowing problems
to escalate compelling the DIS to rehabilitate
them. The S&L of the US have a distinct
focus from that of banks as they are permitted
to provide home mortgage credit, which
is a more risky long-term investment.
The
FDIC acts as the receiver or liquidating
agent for failed insured deposit institutions.
In its role as a receiver, it has fiduciary
obligations to all creditors of receivership
and to stockholders of the fund to maximize
the amounts as quickly as possible. To
include enabling provisions for the DIS
authority to take receivership should
be discussed at the designing stage, as
Sri Lanka may or may not need such a provision.
The Sri Lankan equivalent of US S&L
is the finance companies registered with
the CBSL. There are however, many savings
mobilizing institutions, which are not
under any authority’s effective
supervision. The demarcation of the type
of institutions will be an issue in the
design of a MDIS but it is important that
adequate consideration be given to the
risks of bankruptcy of these unregulated
institutions and their implications for
the financial system as a whole and the
small and unsophisticated depositor, in
particular.
Any insurance should be priced as accurately
as possible to reflect expected losses
and cover risks. The FDIC provides protection
to banks, S&Ls and credit unions.
The marginal pricing could deviate from
expected marginal costs depending on the
extent that moral hazard problems will
lead to special optimal business decisions
by managers. Unfortunately, the current
US DIS which is supposed to feature risk
based prices for individual institutions
is prevented from doing so because the
way the FDIC treats the designated reserve
ratio (DRR) for the insurance funds. The
DRRs for both the Banks’ Insurance
Fund (BIF) and the Savings Association
Insurance Fund (SAIF) have been exceeded
for several years now. For example, during
2002, deposits insured by BIF rose by
4.9 percent to $ 2.5 trillion. As shown
in Appendix Tables 3a-3b, in 2002, the
BIF exceeded the reserve ratio marginally
to 1.27 percent above the target ratio
of 1.25 percent of estimated insured deposits,
while the SAIF reached a ratio of 1.37
percent. During 2002, 11 FDIC insured
institutions failed, of which 10 with
combined assets of $2.5 billion were insured
by BIF. Losses for the failed institutions
were estimated at $630 million. As it
is, over 90 percent of financial institutions
pay essentially nothing for deposit insurance
and this indicates that premiums are not
based on risk. This raises the issue whether
FDIC is playing the role it should, in
reducing moral hazards and giving proper
pricing signals to the market place.
This
problem has led to a number of market
distortions. For example, a large bank
with a huge deposit base and extremely
complex risks may pay the same amount
of premium (zero) for insurance as a small
community bank. Even banks in different
risk categories (the top two CAMEL categories)
pay the same for insurance. When DI premiums
are effectively zero for most institutions,
the incentive for bank managers to improve
their operations and in particular to
worry about the risks they take on, is
insignificant. In a dynamic market place
where new financial intuitions are being
created (some are growing rapidly than
others), such undifferentiated premiums
can lead to moral hazards as well as problems
of unfairness. This would mean that many
banks would get a free ride on DIS.
If
Sri Lanka were to design a flat rate DIS,
the treatment of DRRs should be carefully
thought through, as “pay nothing”
formula would lead to a bonus to some
and a disincentive to others. “Pay
Nothing” or “Zero Premium”
DIS schemes would also lead to institutions
not fully exploiting all available information
in determining the risk. The current DIS
in the US does not have to deal with the
“too big to fail” issue. Although
the Federal Deposit Insurance Corporation
Improvement Act of 1991(FDICIA) significantly
tightened restrictions on bailouts of
such institutions, it seems likely that
the largest institutions will get extra
benefits in the form of potential bailouts
for uninsured depositors, etc. In Sri
Lanka where the two state banks still
occupy more than 50% of the banking sector
and 30% of the financial system, and given
their high risk portfolios, they should
not be allowed to have a free ride on
others. Some institutions may take advantage
of the fact that non-deposit taking activities
are also covered under the scheme. Giving
institutions extra protection without
a cost may cause them to engage in more
risky behaviour than they otherwise would.
In principle, the premiums that banks
pay should reflect the cost they would
impose on the insurance fund (elementary
efficiency condition). But it is reported
that FDICIA, through its prompt corrective
action (PCA) provisions, which mandate
regulatory intervention well in advance
of insolvency, has reduced these expected
costs. If PCA operates properly, the FDIC
should not suffer a loss except in case
of frauds. The marginal costs would therefore
be seen as zero. Up to now, Sri Lanka
has not engaged in PCA but the proposed
amendments to the Banking Act may provide
adequate powers to CBSL to engage in such
action in the future.
If
break even premiums in the future should
be lower than they have been in the past
because of PCA, it is not fair to argue
that it should have zero premiums, because
perfection is not normally achievable
in bank supervision and there can be unforeseen
circumstances. Also, there can be occasional
large adverse macro shocks leading to
insolvency in institutions before corrective
action could be taken (good loans can
turn to bad loans during slumps). Even
if future losses to the FDIC are caused
by macro shocks (no fault of the banks),
it remains true that banks differ in the
degree of risk they impose on the FDIC.
This means that banks with strong balance
sheets are more likely to survive even
in an acute macro shock and hence are
less likely to cause a loss to the FDIC.
The weaker banks, on the other hand, would
pose greater risk to the fund. Therefore,
risk-based premiums designed to reflect
expected loses are still valid for FDIC,
although the system was not designed originally
to function on a risk-based model.
The Canadian Deposit Insurance Corporation
(CDIC), on the other hand, assesses risk
exposure through an early warning system
(EWS) for its individual member financial
institutions, drawing on its experience
of having handled the failure of 43 member
institutions during the past 3 decades.
CDIC emphasizes that a well-designed EWS
can be effective at the early detection
of problem institutions. The three key
EWS emphasized in the CDIC System are:
the macroeconomic; macro prudential; and
advance warning of individual financial
institutions, which are relevant to the
FDIC as well, but the basic purpose of
the US and the Canadian systems differs
in terms of their target population model
and the structure of DIS.
In
principle, DIS should be risk-based, linked
to the capital of the deposit taking institutions.
But, given the absence of comprehensive
information on institutional risks, it
may be difficult to set out a risk based
premium formula at the start. The alternative
would be to start with a second best solution
of a flat rate premium and ensure that
the risk for better managed financial
institutions are minimized through effective
PCA by the regulatory authorities. The
options for introducing a risk-based MDIS
should be kept open for future consideration.
Given the most important goals of DIS
being enhancing macroeconomic and financial
stability, the current US system strives
to maintain a DI fund equal to 1.25 percent
of insured deposits. Hence it stops collecting
significant insurance premiums once the
DRR target has been attained. Instead,
if losses stemming from bank failures
drag the fund below the 1.25 percent DRR,
the FDIC now imposes an ex-post settling
mechanism that could cause banks to pay
high premiums during periods of financial
distress. This feature of the current
DIS in the US would cost financial institutions
a large sum of money as extra premiums
due to high cost of deposit gathering
during business cycle downturns. This
raises the issue whether the system targeted
reserve ratio was based on a long run
expected loss formula which is less procyclical,
but have superior operating characteristics.
Although a procyclical premium system
would be more logical, it may be difficult
to operate such a system in a country
like Sri Lanka as the financial sector
is still used to “fixed rates”,
“fixed installments” and “fixed
premiums”. Also it would be unfair
on the good banks, which are called upon
to subside weaker banks to pay according
to business cyclical changes.
The current US DIS determines a deposit
coverage limit that is fair and transparent,
protects small depositors, allows for
sound financial planning and yet does
not exacerbate moral hazard risk. Choosing
such a coverage limit obviously represents
a trade off among competing goals. However,
the current US coverage limit is not indexed
which appears to be at variance with other
government programmes that benefit citizens
(social security payments, medicare benefits
and personal tax exemptions) which are
all indexed for inflation. The current
system also suffers on the transparency/economic
efficiency front because of the multiple
registrations allowable for coverage.
Up to now, in Sri Lanka, most social benefit
schemes are not inflation-indexed. If
the real return on deposits and investments
can be maintained at all times, the inflation
indexation may not arise. If not, an inflation
linked system may have to be considered
at a future date.
At
present, in the US, account balances up
to US$100,000/- are insured. A customer
with only one individual account with
a certificate of deposit of $ 80,000 with
$ 15000 in accrued interest ($95000 in
total) would be paid the full $95000.
A customer with only one individual account
with a certificate of deposit of $ 90,000
with $ 15000 in accrued interest ($ 105000
in total) will be paid only $ 100,000
because of the insurance limit. Because
the current deposit insurance laws permit
separate insurance coverage for each right
and capacity in which an individual holds
an account, the $100,000/- limit does
not appear to be very effective. In general,
a family taken together could maintain
a much higher amount in their account
in a single institution. There is a strong
case, therefore, for imposing the insurance
limit on the individual depositor and
not on the account. However, this may
be an important policy issue.
The unconventional principle that small
depositors should not be expected to monitor
the safety and soundness of their banks
and that their deposits would be 100 percent
unsafe is that the bank supervisory agency
would be solely responsible for effective
supervision, which is not achievable at
all times. What constitutes a “small”
deposit would imply the coverage limit
for FDIC insurance of which the current
limit of US$100,000/-. This was announced
in 1980 and is not considered to be excessive
even today. But the real value of the
US$100,000/- limit has been significantly
eroded by inflation since 1980. There
are some options to consider in deciding
the potential level of coverage.
Whatever
the initial limit set should be adjusted
over time to reflect the growth in nominal
income/wealth, not just the increase in
the price level. The recommendation arising
out of a critical review of the DIS in
the US was (a) increasing the 100,000/-
coverage limit substantially; (b) indexing
it; and (c) legal simplification that
applies the limit to all the accounts
held in whole or in part by a single individual
in a single institution. In summary, the
idea is to set up a “tuned up”
DIS in which banks pay risk-based premiums.
In the US, the “user fee and “mutual
model” may be seen as complementary.
FDIC is seen as selling an insurance product
to the banks and it ought to charge actuarially
fair premiums for that service. In that
sense, FDIC premiums are treated as user
fees. In particular, the marginal dollar
cost should be positive on insurance deposits.
If a private monopoly in the US sets its
price too high, entry by new competitors
will erode its monopoly profits. Even
if the FDIC sets its premiums too high,
the assets will build up within the insurance
funds as has happened in recent years.
If FDIC sets premiums too low, exit is
not an option. Instead, a flow of claims
will gradually deplete the fund. Given
the theme that DIS should be designed
to neither tax nor subsidise the banking
system in the long run, then the FDIC
should strive to operate on an approximately
break-even basis, which implies a mutual
insurance arrangement. The user fee and
mutual model can be reconciled by allowing
for a two-way flow of revenue between
the proposed authority for Insurance and
the banks. The authority should charge
some positive marginal cost for the insurance
it provides (user fee element). If these
premiums lead to excess profits and unnecessary
build up of the fund, the authority should
rebate monies back to the banks in proportion
to past premiums paid (the mutual element)
or an appropriate formulae.
The DIS can be just “pay box”
type, geared to pay claims as and when
they arise after a failure of a bank or
a financial institution or “risk
minimiser” type, which will have
more broader powers and responsibilities
to control entry and exit to and from
DIS and ability to assess and manage its
own risks. The “pay box” type
DIS does not normally have powers for
supervision of financial institutions
or intervention, but it should have adequate
funding and information relating to deposits.
From 1980 through 1994, the FDIC managed
120 straight deposit pay-offs out of 1617
failed and assisted banks, or 7.4% of
all closings. In 1983 the FDIC created
the insured deposit transfer (IDT) to
the straight pay-off. In an IDT, the insured
deposits and secured liabilities of a
failed bank or thrift society are transferred
to a healthy institution (the agent institution)
and the FDIC makes a matching payment
of cash and /or assets to the institution.
The agent institution pays customers with
the dues to the insured deposits or if
the customer request it, to an account
opened by him in the agent institution.
The IDT minimizes the inconveniences.
DISs designed to have broader powers have
the ability to control entry and exit
from the deposit insurance system, assess
its own risks, conduct examinations of
financial institutions, provide financial
assistance to failed institutions, enforce
resolution activities and adopt risk minimization
methods in discharging their functions.
The mandate and powers should enable the
proposed DIS to transfer at a future date
the dues to accounts or institutions chosen
by the depositor.
The governance practices of DIS should
be on strategic planning, risk management
processes and general internal controls.
The institution set up for DIS should
be subject to corporate governance principles,
in particular in subjecting appointments
to “fit & proper” tests.
The institutions should be accountable
to a higher authority or authorities and
ensure transparency in all its operations.
DIS should also be manned by skilled staff
to ensure its success.
Banks have an explicit claim on FDIC revenues
– a claim that might be valued as
an asset on he banks’ balance sheets.
The FDIC is not a mutual insurance company
owned by banks; rather it belongs to the
taxpayers. For example, if the Congress
were to decide that more coverage is needed
(higher than US$100,000/-) and that a
larger insurance fund is therefore appropriate,
rebates to banks should be cut even though
they may have overpaid for insurance in
the past.
The following policy and operational issues
should be resolved prior to establishing
a DIS for Sri Lanka. All stakeholders
should be involved and be clear of the
issues and participate at discussions.
The DIS function could be assigned to
an existing organization or a separate
entity. If a separate authority is to
be established, issues such as who will
provide the initial capital and who will
be the lender of last resort facilities
should be decided. In general, governments
provide initial capital and are also the
lenders of last resort in a systemic failure.
As experienced elsewhere, the industry
undertakes to repay the government as
soon as the systemic threat is over or
when payments to depositors are completed.
Should it be only banks (commercial and
specialized banks)? Or, banks and finance
companies which are supervised by the
CBSL or should it be extended to other
deposit taking institutions, such as cooperative
rural banks, thrift and credit cooperative
societies and others which are not effectively
supervised by any authority.
The present VDIS in Sri Lanka limits insurance
coverage to Rs 100,000 per individual.
Is it adequate in the present context?
Does it adequately cover small deposits?
How small is small? Should it reflect
some median of small deposits (say, upto
Rs 50,000 or less)? Or should it be only
very small deposits? Should coverage limit
be per deposit or per depositor? The CBSL,
government and the commercial banks should
collectively decide on the appropriate
coverage or the limit. It is also necessary
to decide whether the amount is for one
account or for the aggregate; the scheme
should prevent depositors opening multiple
accounts below the insured limit. Perhaps
the insured amount could be based per
depositor and per bank. Should the coverage
be per individual or family, or for joint
accounts? If it were the aggregate, then
what would be the cut off payment?
What should be the ratio of the deposit
insurance fund to insured deposits (cut
off level)? What action should be taken
when the cut off limit is reached? Should
banks pay above the cut off or should
insured institutions be allowed to enjoy
the benefits of the already paid premium?
Should the premium be a steady rate (flat
rate) over a long period or should it
be a variable rate? Should it be designed
to maintain a target ratio?
Should low risk banks pay for part of
the risks of high risk banks under a flat
rate premium?
What are the categories of deposits to
be included in the DIS? Should all deposits
be included? Should the insurance coverage
be limited to domestic rupee deposits?
Given the importance of foreign currency
accounts in Sri Lanka’s economy,
shouldn’t NRFC and RFC deposits
be included? If so, should premium be
charged in Rupees or foreign currency?
This would make the designing more complex
as one needs to decide the mode of premium
payments and a separate cover for foreign
exchange deposits, the Central Bank having
to ensure that banks have adequate controls
in the management of their foreign exchange.
Who covers the exchange risks if depositors
have to be paid in foreign currency? Should
certificates of deposits be included in
the DIS? (they may disappear when the
money laundering legislation is passed).
The effectiveness of a DIS and maintenance
of public confidence would depend on the
source and the availability of funds.
Usually, governments jointly with member
institutions fund DISs and they also borrow
from the market. Funding should be available
to reimburse promptly the depositors’
claims. Member banks should pay to build
up the DIS, as their clients would be
protected from loss. The DIS Authority
should provide for its own ongoing financing
by placing a deposit insurance premium
assessment on individual depositories
to maintain and if necessary, to restore
the insurance fund to a pre-determined
minimum level.
At the outset it is important to decide
whether the proposed DIS is only a “pay
box” type or whether it should be
empowered with some supervisory and intervention
powers (risk minimiser). Given the enhancement
of powers of the CBSL under the proposed
amendments to the Banking Act, the duplication
of functions between CBSL and DIS authority
should be avoided.
Like in the US, the proposed DIS in Sri
Lanka should be vested in an authority
outside the CBSL to alleviate moral hazards
of banks as well as the public as both
parties would assume that the CBSL will
be there to protect depositors. In addition
to banks, which are regulated by the Office
of the Comptroller of Currency (OCC),
the FDIC consists of Federal Savings Association,
which is regulated by the Office of Thrift
Supervision (OTS). The Federal Reserve
Board regulates their respective holding
companies.
The proposed DIS authority should: enhance
market confidence; be able to have boards
to dispense with claims that arise from
insolvent banks and financial institutions
as quickly as possible; be autonomous
in order to obtain services of highly
skilled personnel at market prices; and
help to maintain the integrity and stability
of the nation’s financial system.
The DIS and CBSL should complement each
other’s areas of functionality and
coordinate their work to avoid duplication.
There is no unique model of a DIS applicable
to all countries and markets. In general,
the DIS legislation should provide for
the following: -
