and Schwartz (1963) argued that the banking crises that sparred the Great
Depression in 1929, was due to non-fundamental factors as there were no
significant fluctuations in macroeconomic factors. Friedman and Schwartz
concluded that the banking crisis was a result of weak fundamentals or
self-fulfilling forecasts rather than actual deteriorations in fundamental
Since the introduction
of Sunspot variables by the Diamond and Dybvig (1983) in their model, most research
papers focused on analyzing the effect of such variables on financial
stability. Several academic papers have also attempted to examine the
relationship between Sunspot variables and the probability of run on banks. The
empirical literature usually ignored the identification of the sunspot variable
and instead, attempted to prove its effect on bank runs.
One of the banks most essential roles is regulating
funds and capital holdings of depositors, which affects the maturity transformation
of the economy. In the past, scholars who studied the cases of banks runs
discovered that bank runs theoretically and evidently occur due to either fundamental
factors as argued by Mitchel (1941), or the spread of panic among depositors, the
so-called “Mass hysteria” or “Mob psychology” (Kindleberger, 1978).
Kindleberger (1978) suggests that causes
behind the sudden and unexpected early withdrawal decision of depositors was the
“Mob psychology”; resulting in mass panic among consumers who suddenly tend to
queue at banks to withdraw their money. The believers of such a view describe
bank runs as a panic caused by false beliefs or rumors which are spread among
masses and push consumers to react accordingly (John Dury, 2015). Describing
the consumer’s reaction as a panic is relevant because such behavior is
self-defeating and irrational, in other words they would not tend to take this
decision in normal times.
As an example, in 2015
The European Central Bank (referred to as ECB, hereafter) decided to end the
financial support granted to Greek banks which has allowed them to
function despite the massive money
withdrawals that lasted for months. This resulted in
a bank run that forced the Greek central bank to announce publically the
decision of terminating Greek banks businesses until the referendum is
conducted. The referendum
accepted the bailout conditions in the country’s debt-crisis proposed
jointly by International Monetary Fund (IMF), ECB, and European commission (EC).
Following the Bryant (1980) model on
banking crises, Diamond and Dybvig (1983) introduced the second model analyzing
bank runs, in which they derived the possibility of the non-fundamental factors
resulting in a bank run. This theory introduced a new approach in studying the
bank failures and banking crises which is the non-monetary structure. The
theory’s main discovery is the strategic complementarity- depositors who come
first, are served first- due to the high cost of liquidation of assets. The
theory also showed how multiple equilibria arise, including a crisis one, which
occurs when all consumers withdraw their funds from the bank, despite their
actual need for money. Diamond and Dybvig’s vital contribution to the study of
bank runs and financial crisis and the new trend in modeling bank runs that was
established, was well acknowledged by researchers in the same field as it was a
turning point in bank runs modeling; to the extent that bank runs literature was
divided into pre and post Diamond and Dybvig’s model.
The seminal paper by Diamond and Dybvig first
introduced the presence of Sunspot variables which drive consumers to change
their consumption decisions in the form of early withdrawal of deposits from
banks, resulting in the depletion of bank’s capital and the reduction of the
liquidity available for other consumers who come later. The Sunspot variables
are usually viewed as a series of random public announcements involving banks
activities or the economy’s well-being. This assumption is very realistic and
evident in the real world, as the psychological studies proved that people tend
to move in groups than to act individually. Such variables are supposed to
affect the preferences of the consumers which have no direct impact on
fundamental factors, hence why they are not mentioned in most papers analyzing
the effect of fundamental factors on bank runs.
of Diamond and Dybvig’s theory believe in the existence of uncertainty among
consumers, but they distinguished between the intrinsic uncertainty caused by
local shocks in banks or the market (Rolnick and Weber, 1986) and the extrinsic
uncertainty which is resulted by widespread deteriorations in fundamental
factors in the economy. Meanwhile Cone (1983) and Jacklin (1987) followed the
Diamond and Dybvig’s theory of bank runs, by focusing on the assumptions underlining
the panic based runs, and pointed out that in order to avoid the run on banks,
there is a necessity of depositors to have restricted trade opportunities or
banks would have to compete with financial institutions which can eliminate the
deposit insurance that they were going to offer.
Carlsson and Van Damme (1993)
explained how the availability of some asymmetric information about fundamental
factors in the economy could eliminate the multiple equilibria recognised by
past scholars. The reliability of this argument can be seen by its applicability
in real life scenarios. Moreover, in a relevant trial, Cass and Shell (1983)
has put the Sunspot equilibria analysis into formal literature related to Macroeconomics,
finance and other fields by proving that some of the random market outcomes are
directly caused by fundamental shocks.
Franklin Allen and Elena
Carletti (2000) criticize Diamond and Dybvig’s theory for not elaborating on their
idea that considers Sunspot variables as a coordination device as it is in fact
true, given that it coordinates depositor’s actions towards the withdrawal of
deposits from banks. While Cooper and Ross (1998) criticize Diamond and Dybvig’s
model for noting that in the ex-ante contracting stage Sunspot variables are
modelled as a cause of a bank run, but never reflecting the repercussions of
the presence of Sunspot variables in their design of deposit contracts.
Obstfeld (1996) believes
that the mechanism showed in Diamond and Dybvig’s theory on bank runs is more relevant
to other financial crises such as a currency crisis, debt run or repo run. Some
contributions indicate that bank runs are naturally grown out of the weak
fundamentals arising in the business cycle in the economy; Gorton’s (1988)
empirical study found evidence that proves that banking panics can be predicted
by examining and analyzing business cycles. During his study of different views
on causes of bank runs, he criticized the Diamond and Dybvig’s model for being
untestable for not showing the formation and change in depositor’s beliefs.
difficulties within Diamond and Dybvig’s model have paved the way for other
researchers to dig deeper into the constraints that aroused the panic and
resulted in massive funds withdrawals. Meanwhile, the model developed by Postlewaite
and Vives (1987), has derived a unique equilibrium without the inclusion of Sunspot
variables in their analysis. That, and the fact that there is not an overall
satisfaction with the assumption proposed by the Diamond and Dybvig’s model
about the Sunspots resulting in a bank run, the literature analyzing the
relationship between fundamental factors and bank runs has to be explored and
3.1 Fundamentals affecting the run
As banks clearly
participate in economic growth of any country, any trouble in the banking
sector will have dramatic repercussions on the economy through affecting fluctuations
in important factors such as liquidity, investment rates, unemployment rates,
and the confidence of consumers and investors in the economy. Banking crises as
per the Arrow model (1953) could occur due to insecurities and doubts about the
economic fundamentals such as exchange rates, stock market prices, etc.
Afterwards, Arrow (1964) extended his model to explain how the instability of
fundamentals is conveyed to different sectors of the economy.
to Mitchel’s (1941) contribution that proved bank crises occur due to
fundamental factors related to business cycle, building on the traditional view
on what causes bank runs, Jacklin and Bhattacharya (1988) believe that bank runs are
naturally grown from fluctuations in the business cycle curves. Such an attempt
interpreted bank crises as an outcome of sufficient data that indicates
negative economic performance, rather than a random event. Various researchers
anticipated bank crises and related them to a downturn within the economy,
through which depositors tend to withdraw their funds suddenly out of the
belief that banks could not be able to fulfill their commitments.
The Calomiris and
Mason’s (2003) model explores the failure of individual banks during the Great
Depression and argues it was because of fundamental factors and exogenous
local, regional, and national economic shocks. In the context, of the previous
model, a bank’s failure is caused by reasons other than the spread of panic,
for example a nationwide decline of banking activities.
The Chari and Jagannathan (1988) model has revealed that
bank crises transpire when depositors receive negative signals about the future
of banks liquidity and when the liquidation of capital is “expensive”. In other
words, this model claims that the public announcements give bad indications
concerning the fundamental factors in the economy. Such model elaborated that
the Sunspot variables that could affect the banking system should be related to
the fundamental factors in the market. The model’s credibility is mirrored in
its relevance to real life situations, where there is a connection between
sunspot variables and fundamental factors in the economy.
Following Gorton’s (1988) footsteps, Allen
and Gale (1998) developed a model which clarified that the origins of bank
crises are related to business cycles, considering the information as the main
reason of bank runs. Despite the fact that the model claimed that bank runs are
useful as financial crisis can ensure efficiency, it did not give any empirical
evidence to support such assumption. In
a similar trial, Calomiris and Kahn (1991) tried to prove through developing a model
that in a certain situation, bank runs are beneficial for the bankers, by
showing the benefits the demandable debts provide to bankers.
According to Calomiris and Gorton’s
(1991) model, there is no evidence that supports the assumption that Sunspot
variables result in banking runs, but more that they occur due to a series of
random events. This assumption makes sense as it did not neglect the impact or
the availability of Sunspots in the market, although it stated that most of bank
failures that occurred worldwide are due to fundamentals factors. In addition,
it emphasized the ability to reduce the costs of the crisis when banks create a
partnership to ensure standing to face forthcoming panics.
Therefore, Arifovic, Jiang and Xu (2013)
assumed that away from the Sunspot variable, the absence of coordination
between banks is the reason behind bank failures. We can expect that depositors
tend to react more strongly, in times of uncertainty, to public announcements- especially
when it contains information about the conditions of the economy.
From the afore-mentioned papers
concerning fundamental factors affecting the run on banks, we are provided with
an implication that there is no major agreement on any factor as the main cause
of bank runs. This paper is building upon the idea, proposed by some scholars,
that Sunspot variables are always present in the market, assuming that bank
failure is caused by factors linked to the market and economy. As a further
step to the pre-mentioned assumption, we will examine policies tackling bank
runs, in an attempt to avoid the run on banks and meditate the effects of a
bank failure, which ensures a smoother path towards financial stability.
4.1 Policies tackling bank runs
Due to the major consequences of bank
failures on financial stability, and out of the realization that a banking
crisis is inevitably repetitive, governments decided to take precautionary
measures in order to avoid the occurrence of bank runs. These measures
considered the studies that explored the different reasons behind the bank runs.
This paper will inspect policies tackling banking crises which act as
safeguards to the banking system and the economy as a whole.
In response to the Bank turmoil of
the 1930s, government officials and monetary policymakers implemented certain
policies in order to avoid bank runs and consequently lessen the risks accompanying
them. Applying counter policies to escape the negative repercussions of bank
runs is at the root of capital sufficiency requirements and deposit insurance.
Federal deposit insurance was created
to avoid bank runs, and subsequently a large number of banks during the Great Depression
in United States urged the federal government to implement this, believing that
depositors under insurance would react less to mass runs than those without. However,
the insurance is only partially effective. According to Jacklin (1987), banks
should offer insurance in order to compete in financial markets; this piece of literature,
and these types of policies, gave the depositors a way to protect their
deposits and investments.
Understanding the origins of bank
failures and the sudden deposit withdrawal is considered the first and most
crucial phase when applying the policy which counters bank failures. For
example, if bank failures are caused by fundamental factors, then the policymakers
should work on supporting the financial regulation and supervision of a banking
system acting as a monitoring system for bank failures. Meanwhile, the Madiès
(2006) model emphasizes the efficiency of applying an equivalent preventive
method to avoid bank runs, which is relevant to bank failures that results from
depositor’s confusion in the face of asset value shocks. This model was the
first to offer an experimental study of mis-coordination based bank runs within
Diamond and Dybvig’s model.
Building on the previous literature,
and the case we are studying, we will move on to examining the capital
regulation policy in order to find out how it protects banks and governments from
a potential bank run.
4.2 The Basel Accord on Capital
The Committee of Banking Regulations
and Supervisory Practices was established
in 1974 by the central bank Governors of the group of 10 countries as a
response to serious disturbances in international currency and banking markets;
they formed the Basel Committee on Banking Supervision. This committee was
working under the supervision of the bank of international settlements in
Basel, Switzerland in order to improve financial stability by enhancing the
quality of banking supervision worldwide, serving as a platform for regular
cooperation among its members.
It introduced the committee’s first
accord on capital regulation in 1988, ‘Basel I’ was considered a benchmark
agreement in the field of regulating capitals, as it standardised the bank
capital regulations for the first time among 12 countries. Apparently the Basel
Accord had set goals that were jointly designed to reduce the risk of international
banking system and to maximize competitiveness within the banking market.
In addition, the committee expanded its membership up to 45 institutions
and 28 jurisdictions since its foundation. Starting with the first accord 1975,
the Committee has established a series of international standards for bank
regulation to ensure efficiency and competition. Furthermore, the Basel accord
spread the use of risk-based capital ratios which granted banks the ability to
measure risk capital requirements in a quantitative and qualitative manner for
the first time in an internal model
4.3 Capital regulation policy
Building on Diamond and Dybvig’s (1983) model
that assumes banks are valuable for providing consumers liquidity services
through offering liquidity insurance, there would be no need for financial
intermediaries as banks will play this role. Banks are supposed to lend to
investors and depositors attain efficient risk allocation at banks. The
literature on capital regulation by A. C. Santos (2001) explains the role of
this policy in eliminating the possibility of run on banks, depending on the
inclusion of the deposit insurance method. Applying the deposit insurance
policy essentially protects depositor’s investments, which in turn will deter
them from withdrawing their deposits before their expected maturity time.
Away from assuming that
the proposal resulted in the recent progress in research on banking, researchers
could not yet find the optimal approach to regulating bank’s capital. Following
this, it is clear that there are various opinions concerning the capital
regulation policy design which is relevant to the narrow scope of studying the
implications of such regulation.
assumption, Rochet developed a model demonstrating that banks would be risk-lovers,
if they are undercapitalized. Rochet’s approach raised many concerns among
scholars, who focused on the method used in measuring risk, not opposing the
principle of relating capital-holding to risk and how subjective the process of
setting minimum capital levels is. The major problem lies in the need to
classify banks’ assets into a limited number of categories, each of which
requires a standard capital holding, which is obviously somewhat unrealistic.
According to Gehrig
(1995), a critic of the accord on credit risk to the exclusion of market risk, argues
that risk-based capital adequacy standards might in some situations create
incentives that would increase the vulnerability of the banking system.
Despite the level of
progress reached concerning modeling credit risk, some researchers believed
that capital regulation would end up with the migration of business lines to
other market segments which are not very strictly regulated, in order to capture
the gains that come with a transition to a less regulated market.