Description
The weakness of the Japanese banking industry, suffering from acute problem of non-performing loans, prevents Japan from restoring sound growth rates despite having undertaken structural reforms and substantial fiscal policy efforts, and, through impairing transmission channels of monetary policy, it has also made ineffective efforts to stimulate the economy through “zero interest rates” and quantitative easing policy.
Change and Crisis
in the Japanese
Banking Industry
Mariusz K. Krawczyk
HWWA DISCUSSION PAPER
277
Hamburgisches Welt-Wirtschafts-Archiv (HWWA)
Hamburg Institute of International Economics
2004
ISSN 1616-4814
Hamburgisches Welt-Wirtschafts-Archiv (HWWA)
Hamburg Institute of International Economics
Neuer Jungfernstieg 21 - 20347 Hamburg, Germany
Telefon: 040/428 34 355
Telefax: 040/428 34 451
e-mail: [email protected]
Internet:http://www.hwwa.de
The HWWA is a member of:
• Wissenschaftsgemeinschaft Gottfried Wilhelm Leibniz (WGL)
• Arbeitsgemeinschaft deutscher wirtschaftswissenschaftlicher Forschungsinstitute
(ARGE)
• Association d’Instituts Européens de Conjoncture Economique (AIECE)
HWWA Discussion Paper
Change and Crisis in the
Japanese Banking Industry
Mariusz K. Krawczyk
HWWA Discussion Paper 277http://www.hwwa.de
Hamburg Institute of International Economics (HWWA)
Neuer Jungfernstieg 21 - 20347 Hamburg, Germany
e-mail: [email protected]
This Discussion Paper has been prepared in collaboration with the HWWW-Research
Programme “Development und Integration”, while the author was a guest at the
HWWA.
Edited by the Department World Economy
Head: PD Dr. Carsten Hefeker
HWWA DISCUSSION PAPER 277
May 2004
Change and Crisis in the
Japanese Banking Industry
Abstract
The weakness of the Japanese banking industry, suffering from acute problem of
non-performing loans, prevents Japan from restoring sound growth rates despite having
undertaken structural reforms and substantial fiscal policy efforts, and, through impair-
ing transmission channels of monetary policy, it has also made ineffective efforts to
stimulate the economy through “zero interest rates” and quantitative easing policy. Mis-
understanding the roots of the banking crisis contributed greatly to its exceptional length
and depth and prevented its early solution. Poor coordination and sequencing of liberali-
sation of financial services together with macroeconomic policy mistakes have been re-
sponsible for the crisis. But the origins of those mistakes can be traced to the bureau-
cratic character of the country’s policy-making process often lacking a comprehensive
analysis of macroeconomic effects of a particular policy decision. Simple solution to the
banking crisis in the form of eliminating non-performing loans burden, bailing out ail-
ing financial institutions etc., if not accompanied by decisive steps restoring rationality
to the public spending and resolving the problems arising from the presence of the gov-
ernment sponsored financial institutions, may not be sufficient to achieve sound growth
and prevent similar crises in the future.
JEL-Classification:.G21, G28
Keywords: Japanese economy, banking crisis, financial liberalisation
Mariusz K. Krawczyk
Faculty of Economics
Fukuoka University, 8-19-1 Nanakuma, Jonan-ku, Fukuoka 814-0180
Japan
E-mail: [email protected]
1
For more than 10 years Japan has not been able to overcome the
repeating sequence of mild recessions followed by periods of stagnation. An
average real growth rate of approximately 1 percent per year between 1991 and
2001 was the lowest among advanced industrial countries. Negative growth rates,
falling individual consumption, historically high unemployment, and fast growing
public debt stand in such a sharp contrast with relatively stable performance
before 1980s, that Japanese often call the 1990s their “lost decade”.
The depth and length of the Japan’s stagnation - the costliest recession
suffered by any advanced industrial economy after the World War II - has led
some to predict that, till now very conservative, Japanese savers may even flee
into cash, gold, and offshore accounts (equal to thing until now unimaginable; to
capital flight from Japan)
1
. Although, there are numerous causes why the
Japanese economic performance continues to disappoint, the weakness of the
country’s banking industry is one of the main factors behind the inability to break
this prolonged cycle of recessions. For a large part of the last decade Japan has
experienced a steady deterioration of its banking sector condition. The decline of
the Japanese banking industry began at the end of the 1980s asset bubble and
turned into a full-blown crisis in the second of half of 1997 when a number of
high profile financial institutions collapsed. There is no significant improvement
since then, despite of enormous amounts of capital allocated for strengthening
and revitalising the financial system (approximately 60 trillion yen spent in last
five years what amounts to 12% of the country’s GDP). This contrasts sharply to
the S&L (savings and loan association) crisis of the late 1980s in the USA or to the
Nordic Banking Crisis of the early 1990s, where the recovery started 2-3 years
after the crises had begun. Ailing banking industry prevents Japan from restoring
sound growth rates despite having undertaken structural reforms and substantial
fiscal policy efforts, and, through impairing transmission channels of monetary
policy, it has been also one of the reasons why the Bank of Japan (BOJ) policy of
quantitative easing has failed to stimulate the economy
2
.
Banking crisis is not a phenomenon unique to Japan. On the contrary,
the basic aspects of the Japanese banking crisis are almost typical of banking
crises in general (Mayes et al., 2001). Its roots can be traced to major regulatory
changes towards liberalisation that was poorly coordinated between sectors of the
economy, a period of rapid economic growth accompanied by an excessive asset
expansion, unsustainable macroeconomic policies, and a substantial adverse
shock. It is argued here that misunderstanding the origins of the crisis, lack of
sense of urgency, weak corporate governance, poor supervision and regulatory
forbearance when the system came under stress led to the worsening of the crisis
and its exceptional length. The solution to the crisis should not be viewed in
isolation from the processes taking place in the rest of the economy and must not
be reduced to a mere resolving the non-performing debt problem.
The remaining of the paper is organised as follows. Section 1 provides a
brief overview of the banking crisis and sums up the main factors that contributed
to the crisis, e.g. the regulatory framework as well as banks’ and authorities’ poor
management of the crisis. Section 2 presents the current condition and the main
problems the banking industry is facing. Section 3 defines the conditions
necessary for the recovery of the banking industry against the background of
economic conditions prevailing in Japan. Finally, concluding section provides
some more general lessons that can be drawn from the experience of the Japanese
banking crisis.
2
1. A Brief Overview of the Japanese Banking Crisis
1.1 The Prelude: Asset Inflated Economy
The emergence of the bubble economy can be attributed to a combination
of external pressure, ongoing domestic liberalisation, and policy mistakes made by
the central authorities. Growing consensus that the US dollar was unrealistically
overvalued against the other currencies (especially the Japanese yen) resulted in
coordinated policy effort to devalue the US currency known as Plaza Accord
(September 22, 1985). Consequently, the dollar lost 45% of its value between
February 1985 (when it reached a peak of 260 yen/dollar) and May 1987 (141
yen/dollar). The move, primarily aimed at putting cap on the US trade deficit with
Japan, imposed a heavy toll on the Japanese economy known as the “High-Yen
Recession”. The real economic growth dropped to –0.5%, its first negative rate
since the oil shock and the total unemployment rate reached historically high level
of 3.1% in May 1987. Concerned with already growing at that time public debt
(50.3% of the GNP), the authorities chose monetary expansion with reduction of
the official discount rates hoping for a wealth effect that would increase domestic
consumption and investment and, in the end, boost the real economic growth.
The scenario worked as expected. In February 1987, the BOJ reduced its
discount rate to a historically low 2.5% level and kept it there for two years till the
mid-1989. The result was an increase in the money supply (defined as M2 plus
CD) that reached 10.8% in 1987 and remained on average 11% until 1990. As
shown in Figure 1, the policy of easy money supported investment and asset
prices increase. The stock prices increased threefold (Nikkei 225 increased from
13,024 in January 1983 to 38,916 in December 1989) and land prices increased
fourfold during the same time. Meanwhile, the BOJ concentrated its anti-
inflationary vigilance on the consumer price movements, which, thanks to the yen
appreciation, were becoming increasingly distorted by the decreasing prices of
tradable goods. The Bank’s policies largely ignored the surge in asset prices. The
unprecedented monetary policy easing coincided with massive privatisation
programmes performed by the Nakasone administration (Japan National Railway
and Nippon Telephone and Telegraph) and partial liberalisation of financial
markets. These two factors too contributed to the asset bubble getting out of
control.
Figure 1: Asset Prices in Japan (1983 = 100)
0
100
200
300
400
500
600
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
Nominal GDP Nikkei 225 Land prices
Source: Japan Institute for Social and Economic Affairs
3
The asset bubble was abruptly brought to an end in the second half of
1989 when the BOJ started increasing its discount rate. The share prices
collapsed in summer 1990 (Nikkei 225 dropped from 38,915 in December 1989 to
20,984 in September 1990). Meanwhile, to contain the rise in land prices, the
Ministry of Finance (MOF) introduced limits on total amounts of loans to real
estate. The limits resulted in sudden contraction of lending to real estate and were
lifted in January 1992, after the land prices had started to decline. The overheated
economy gradually slowed down.
1.2 Banks’ Response to the Asset Bubble
The banks actively participated in the exacerbation of asset bubble by
credit creation. The rapid increase in land prices and real estate related
businesses shares led to a lending boom towards this kind of businesses. On the
contrary, lending to large manufacturing corporations was on decline. While the
share of lending to big manufacturing companies to total lending decreased from
almost 60% in 1985 to less than 50% in 1990 and 40% in 1991, the share of
construction and real estate lending increased from 14% to 18% during the same
period (Fujiwara, 1998). The lending to real estate related businesses was based
on misguided expectation that land prices would increase indefinitely (as they did
during all post war years except for 1975). The sharp decline of land prices
following the end of the asset bubble caused the quality of loans to the real estate
industry to deteriorate rapidly. Banks themselves were to blame for lack of proper
risk evaluation. Lending offices were using a simple “direct capital comparison
method” e.g. comparing the published land prices of neighbouring properties (only
later they began calculating the value of land collateral based on future income
Interest Rate Controls in Japan
The Temporary Interest Rate Adjustment Law was introduced in 1947.
It allowed the Bank of Japan (under the Ministry of Finance supervision) to
establish detailed interest rate ceilings on deposits and on short-term lending.
The Law applied to all kind of banks and credit cooperatives (however the
Shinkin small- and medium-sized businesses credit associations, agricultural
cooperatives, and the credit cooperatives were exempt from the restrictions on
lending rates). Government financial institutions (including postal savings)
were not subject to the regulations. As a result all financial institutions applied
almost the same interest rates. The variation of the interest rates remained
very small even after gradual relaxation of the controls (a significant part of the
controls on lending rates was removed in 1959) as banks tended to apply a
uniform rates within the Federation of Bankers Association. The same
happened with long-term lending rates despite of the fact that formally the
long-term rates were not a subject to interest rate controls.
During 1980s the controls on deposit interest rates were gradually
relaxed. In 1979, certificates of deposits (CD) became exempt from the interest
rate controls. Large deposits (more than 1 billion yen) became exempt from
controls in 1988 and their minimum amount was later reduced and finally
abolished. Not until 1971 were the ordinary banks allowed to accept deposits
of maturity longer than one year (in 1981 they were allowed to accept deposits
of maturity three years). Only since October 1993, floating rates and maturity
longer than three years have been permitted. Before that, long-term credit
banks had a monopoly on long-term deposit activities. The interest rate
controls were finally abolished in September 1994.
4
flow), they extended their lending disproportionately to the value of collateral, their
lax policies allowed for using the same collateral for multiple borrowing etc.
Table 1: Credit Ratings of Major Japanese Banks
Bank 3/90 3/95 3/96 3/97 3/98 3/99 3/00 3/01 3/02
Industrial Bank of J.
AAA A+ A A A- BBB BBB+ BBB+ A
Long Term Credit B.
1 AA A- BBB+ BBB+ BBB- BB- BBB- BBB- BBB-
Nippon Credit Bank
2 AA+ A BBB- BB+ BB+ BB- BB- BB BBB-
Dai-Ichi Kangyo
AA A+ A A BBB+ A A A BBB
Sakura
AA+ AA- A- A- A- BBB BBB A- A-
Fuji
AA A+ A- A- BBB+ A A+ A+ BBB
Mitsubishi
3 AA+ AA- A+ A+ A A- A- A- BBB+
Asahi
AA+ A A A A BBB BBB BB+ BB+
Sanwa
AA AA- A A A BBB+ BBB+ BBB+ BBB
Sumitomo
AA+ A+ A A A- BBB BBB BBB+ BBB
Daiwa
AA+ A- BBB+ BBB+ BBB- BB+ BB+ BB+ BB+
Tokai
AAA A A A A BBB- BBB- BBB A
Hokkaido
Takushoku
AA A BBB- BBB- BBB-
Tokyo
3 AA+ AA+ AA+ AA+
Notes:
1
since 1999 Shinsei Bank,
2
since 1999 Aozora Bank,
3
in 1996 merged into Bank of Tokyo Mitsubishi.
For mergers after 1999 see Table 2.
Source: Miyajima and Yafeh (2003).
Large increase in lending to real estate is often viewed as a primary cause
of the non-performing loans problems that emerged later in 1990s. However it
would be a great oversimplification to attribute all the problems of the banking
industry only to the irrational lending behaviour during the late 1980s. On the
contrary, it is possible to argue that Japanese banks were left with no choice but
to undertake risky lending activities in order to survive. Their very unusual
lending practices can be linked to two characteristics of the Japanese banking
industry at that time. First, partially due to the pressure from the US government,
the Japanese authorities embarked on the protracted process of liberalisation of
the country’s financial system. By mid-1980s the process was only half finished
with gradual removal of restrictions on access to the corporate bond market and
creation of the commercial paper market on the one hand but still numerous
restrictions on financial institutions activities in place (see box on liberalisation of
the interest rates controls). These developments had serious consequences.
Incipient liberalisation triggered competition for customers that further cut down
already low profit margins of Japanese banks (Shimizu, 1999). At the same time
large corporations, until then the banks’ main borrowers used their newly
acquired freedom of access to corporate bond markets. Enjoying very high credit
ratings, the corporations were able to borrow directly from the market at
significantly lower cost. Japanese corporations raised approximately $519 billion
in first five years after lifting their bond market restriction (Taniguchi, 1999). The
long-term credit banks, which enjoyed most protection under the interest rate
controls, were among hardest hit by the new reality. Facing declining profits and
loosing their traditional corporate customers, the banks had no choice but to look
for substitute lending markets. They increased their loan activity not only for the
5
real estate industry, but also for small and medium size enterprises, individuals,
finance industry, and overseas investment (Shimizu, 1999, 2000).
Second, the Japanese banking industry suffered from huge overcapacity
problems. The industry was designed to provide financing for the Japanese
economy that was growing on average 9.7% between 1960 and 1973. At the same
time the average growth in Europe and the USA was 4.8% and 3.9% respectively.
Slowing growth (on average 4.4% between 1975 and 1985) and decreasing
demand for funds in the second half of 1970s and in 1980s was not accompanied
by the reduction of the banking sector capacity. Fierce competition for lending
under heavy overcapacity in the banking industry and banks’ special attention to
their market shares
3
caused a kind of cognitive dissonance e.g. in order to survive
banks had to compete for borrowers even if it meant willingly loosening credit
standards. The myopic character of the lending patterns reached almost farcical
dimensions with banks engaging in phoney realtor activities in exchange for kick
backs or lending more than $1.8 billion to a small restaurant owner. Fully rational
change in lending patterns, which resulted from changes in regulatory
environment, reached at that stage a state of irrational exuberance.
It was however still 5 years before the banking sector soundness came to
question. During that time the amount of non-performing loans was steadily
increasing. Banks were extremely reluctant to raise their loan-loss provisions.
This might have happened either due to the banks’ concern that increase in loan-
loss provisions would be providing markets with negative expectations about
future increases of non-performing loans, or because individual banks did not
want to draw attention to themselves by braking ranks with other banks. Banks
could not write-off their non-performing loans due to restrictive tax regulations
that allowed for write-offs only in cases of bankruptcy or foreclosure proceedings.
Given the average speed of legal proceedings in Japan it substantially slowed
down the writing-off non-performing loans. Close relationship between banks and
their borrowers, known as the main bank system might have been yet another
important reason why banks were slow in cleaning their balance sheets of non-
performing loans. The main bank may be reluctant to identify troubled borrower
because this in turn may raise questions about its own monitoring abilities and
because the main bank would be usually required to absorb the main bulk of the
losses incurred by creditors. When the main bank itself comes under stress it may
result in bank becoming a captives of its own borrowers. Finally, poor corporate
governance of Japanese banks might have been one of the main factors behind
the growing amount of non-performing loans. The weak corporate governance of
Japanese banks originates from the fact that banks’ major shareholders have little
incentive to confront the management. Main groups of shareholders have been
either dependent on management (banks’ employees and banks’ corporate
borrowers) or had close relationship with management (other banks and
insurance companies due to cross shareholding). Members of boards of directors
are usually promoted from the ranks of employees at the end of their careers and
resign after their terms expire in order to be replaced by junior employees. This,
combined with weak internal and external audits, creates incentives for concealing
problems rather than taking decisive steps during manager’s tenure. During 5
years following the collapse of the bubble economy, Japanese banks chose not to
make determined effort to solve mounting non-performing loans problem. Instead,
they preferred to wait for stock and land prices to recover to the bubble economy
level.
6
1.3 Banking Crisis
1.3.1 Instability Begins (early 1995 – Fall 1997)
The instability in Japanese banking industry came into public view in
early 1995. Two insolvent credit cooperatives (Tokyo Kyowa and Anzen) collapsed
at the end of 1994 and further two (Cosmo and Kizu) followed by August 1995. In
August 1995 Hyogo Bank, one of regional banks, was ordered to suspend its
operations. The closure of insolvent financial institutions marked the end of policy
not allowing depository institutions to fail. Adding insult to injury, the loss
concealing scandal forced Daiwa Bank to close its operations in the US later in the
year. Those events coincided with the controversy surrounding the Jusen (the
housing loan) companies.
The Jusen companies, established in 1970s by banks and other financial
institutions for lending in home mortgage market, engaged in real estate lending
in late 1980s and at the beginning of the 1990s. With funds from agricultural
cooperatives the Jusen companies increased their lending to real estate
businesses filling the gap left by commercial banks that had to follow restrictions
on lending to real estate imposed by the MOF
4
. Already by early 1992 the Jusen
companies lending quality raised serious concerns but it was only in 1995 when
authorities decided to step in. By that time, three quarters of Jusen loans were
already non-performing.
Soon the Jusen problem became a major political issue. It was the first
time that the government attempted to use public funds for solving problems of
weak financial institutions. For that reason, the discussion surrounding the Jusen
companies set an important precedent in dealing with such problems. Facing a
clear public dislike to the idea of using taxpayers’ money for rescuing banks, the
authorities reached a consensus that problems of financial institutions should be
solved possibly without direct involvement of the government. In Jusen case, 55%
(or 3.5 trillion yen) of their losses had to be covered by their founding banks, 27%
(1.7 trillion yen) by the lending banks, 8% (530 billion yen) by the agricultural
financial institutions, and the remaining 10% (680 billion yen) by the public
funds
5
.
The collapse of insolvent financial institutions together with the Jusen
confusion heavily damaged the image of Japanese banking industry. It was then
that the “Japan Premium” (premium on lending to Japanese institutions)
appeared for the first time on international financial markets and international
rating agencies, already downgrading Japanese banks since 1989, radically
reduced the credit ratings for 1996 (while there was no major Japanese bank with
B ratings in 1995, a year later four of them were classified as BBB and only one
reached AA rating). Despite of that, inaction and regulatory forbearance of
financial authorities continued until 1997, substantially undermining the public
confidence in country’s banking industry.
1.3.2 Crisis (Fall 1997 – mid 1999)
The increase of consumption tax rate in April 1997 marked the beginning
of another recession. The economic conditions deteriorated and by November 1997
Japanese financial industry faced several spectacular bankruptcies. On November
3, 1997, Sanyo Securities, a second tier security company defaulted on the inter-
bank loan market. It was followed by the collapse of Yamaichi Securities, one of
the country’s big four securities companies, and by Hokkaido Takushoku Bank,
one of the, then 12, city banks. Those events resulted in serious disruptions on
the inter-bank market and sell-off of banks’ shares on the Tokyo Stock Exchange.
Unable to continue its non-interference policy, the government established, under
newly enacted Financial Stabilisation Law, a 30 trillion yen fund for capital
7
injection into troubled banks. The Law was an important novelty in authorities’
response to banking problems. Loosely modelled after the US Prompt Corrective
Action regulations, the Law provided introduction of well-defined self-assessment
procedures and their external audit for banks and, no less important, it clearly
defined the capital threshold ratios for authorities to intervene in an individual
bank’s management affairs (restrictions on dividend payments, management
bonuses, closure of branches, suspending operations etc.). It reduced the scope
for regulatory inaction so often exercised before 1997.
In March 1998, four months after the crisis had begun, the newly created
Financial Crisis Management Committee supervised capital injection of 1.8 trillion
yen for 21 major banks in the form of subordinate debt. The injection of public
funds was intended to help the banks to meet their capital requirements. In order
to avoid drawing attention to weaker banks, all banks applied for the same
amount (100 billion yen) of capital. Calm did not last however long. The collapse of
the Long Term Credit Bank, the largest bank failure in post-war Japan, brought
new waves of instability. In October 1998, the Financial Function Early
Strengthening Law replaced the earlier Financial Stabilisation Law. The new Law
was designed to help banks with insufficient capital. The Diet authorised the use
of 25 trillion yen for this purpose. 15 major banks received 7.4 trillion yen already
at the beginning of 1999. Contrary to the 1998 public fund injection, this time
banks were required to submit rehabilitation plans and the amount of public
capital varied between banks. The capital injection took form of preferred stock
that could be converted into common stock and used for exerting pressure on
bank’s management if the implementation of a rehabilitation plan was not
satisfactory. Of major banks, only the Bank of Tokyo Mitsubishi, the biggest and
soundest of Japanese banks at that time, turned down the offer of public funds.
The second law introduced in October 1998, the Financial Reconstruction
Law, was designed to deal with failed banks. Under this law, 17 trillion yen was
set aside on special account to become a deposit guarantee for failed banks, and
another 18 trillion was held for purchasing shares (nationalisation) of failed banks
and for supporting the Resolution and Collection Corporation (RCC) to purchase
non-performing loans. Finally, admitting that banks’ supervision had not been
sufficient, the Financial Supervisory Agency (FSA) was established and took over
supervision of banks from the MOF
6
in June 1998.
Under the new legal framework, the Financial Reconstruction Committee
decided to place under public administration (that means effective nationalisation)
the Long-Term Credit Bank and the Nippon Credit Bank (late 1998), two banks
that used to be indispensable participants of Japan’s rapid economic growth in
the early post-war period.
1.3.3 Merger Wave (1999-2003)
Capital injection under the Financial Function Early Strengthening Law
required receiving banks to clean up their balance sheets of the non-performing
loans and strengthen their capital positions. Restructuring and cost cutting was
an obvious way to become profitable and be able to return public funds as soon as
possible. This triggered a series of mergers. 23 large banks that existed in 1985
have been regrouped into four groups (Table 2)
7
. The number of regional banks
also slowly decreased from 133 to 118 (mainly in the second half of 1990s).
The long awaited liberalisation of financial services (the Japanese “Big
Bang” Plan) finally started being implemented. Under the new rules, the
restrictions that once separated banking, securities, and insurance businesses
have been lifted. In order to survive in a new environment, banks had to look for
new partners.
8
Table 2: Consolidation of the Banking Industry
Date of Merger Merging Parties New Financial Group
September
2000
Dai-Ichi Kangyo Bank, Fuji Bank, Industrial
Bank of Japan
Mizuho Holdings
April 2001 Sakura Bank, Sumitomo Bank
Sumitomo Mitsui
Banking Corp. (SMBC)
April 2001
Bank of Tokyo Mitsubishi, Mitsubishi Trust
and Banking, Nippon Trust Bank, Tokyo Trust
Bank
Mitsubishi Tokyo
Financial
Group, Inc.
April 2001
Sanwa Bank, Tokai Bank, Toyo Trust and
Banking
UFJ (United Financial of
Japan) Group
March 2003 Asahi Bank, Daiwa Bank Resona Group
1.4 What Caused the Crisis?
As the above story of unfolding banking crisis in Japan shows there is no
single cause that triggered the meltdown of the industry. Instead it is possible to
identify a set of factors that contributed to the worsening of the crisis.
Undoubtedly, monetary policy mistakes by the BOJ in the second half of the
1980s generated asset inflation and provided excessive liquidity to the financial
system that was unprepared for absorbing it
8
. It is also possible to argue, as
Jinushi et al. (2000) do, that too tight monetary policy at the beginning of the
1990s further aggravated the crisis that was already well under way.
Second, the Japanese financial system was an instrument of industrial
policy aimed at supporting reindustrialisation, supporting investment and export-
led growth, protecting domestic businesses from international competition and
providing liquidity at possibly low cost. The set of objectives produced a highly
bank-dependent very rigid system full of administrative controls and guidance.
The system fully met its objectives and served well during the period of high
growth of Japanese economy in the 1960s. The authorities were very slow however
to modify the system once it accomplished its mission (Cargill, 2000, Shimizu,
2000). Japan, Germany, and France are usually given as examples of bank-
oriented financial systems. But while the rate of bank deposits in Germany
decreased from 67.9% in 1971 to 39.3% in 1999 as shown in Figure 2, the
Japan’s ratio of bank deposits to total financial assets remained almost
unchanged (62.9% in 1998). As a result, at the time the asset inflated bubble
economy arrived, Japan still relied on its highly bank-dependent but only partially
liberalised and lacking transparency financial system. This leads to the third
factor, extremely poor supervision and regulatory forbearance of Japanese
financial authorities. There might have been various reasons of the authorities
poor performance ranging from the lack of political leadership, to cosy relationship
between regulators and regulated that originated from amakudari
9
, to simply poor
quality of staff working on banking issues (see conclusions at the end of the
paper), or to the fourth factor; the opposition from the taxpaying public to use
public funds for dealing with ailing financial institutions (the opposition that
became even firmer when series of scandals at the MOF and at the BOJ was
revealed).
9
Figure 2: Breakdown of Financial Assets in the Household Sector
0%
20%
40%
60%
80%
100%
Germany France UK USA J apan
country
stock
trust banking
insurance andpension
bonds
bank deposit
Source: Comparative Economic and Financial Statistics, BOJ
Perhaps each of the factors partially contributed to the depth and the
length of the crisis. These factors can fully explain the unfolding of the banking
crisis in Japan and there is no need for “revisionist” theories claiming that the
Japanese bureaucrats intentionally planned and implemented the crisis as for
instance Lee (1999) does.
2. The Condition of the Japanese Banking Industry
2.1 Increasing Non-performing Loans
Japanese banks accumulated loss of approximately 88.2 trillion yen due
to disposing of non-performing loans. Despite of this enormous loss they still have
more than 38 trillion yen of disclosed non-performing loans (Table 3). This
amounts to more than 5% of their portfolio, yet there are some analysts who argue
that the disclosed amount of non-performing loans is rather underestimated. For
instance, Fukao (2002) points out that the rules of the FSA Bank Examination
Manual contain too narrow definition of non-performing loans. Others, like Horie
(2003), argue that there is a bias towards big companies in estimating their
financial position and therefore also overestimation of the performance of funds
loaned to these companies.
Table 3: Loss on Disposal of Bad Loans in All Banks (\ billion)
Source: the FSA
10
The prolonged recession is partially to blame for the increasing amount of
non-performing loans since more and more firms face financial distress. Their
loans increase the amount of non-performing loans. On the other hand banks
tend to willingly underestimate their non-performing loans positions. Peculiarities
of Japanese corporate governance mentioned in the preceding section aside,
banks have been under heavy external and internal pressure. Within the keiretsu
(industrial group) framework, which big Japanese firms still tend to prefer,
classifying a certain loan as risky, puts not only a particular borrower but also a
group as a whole in a difficult position. And bank, itself being a member of such a
group can hardly afford to alienate its other members. Next, disposing of a bad
loan and placing a borrower under bankruptcy means a reduction of tax income
for local authorities. This, combined with the presence of “politically well-
connected firms”, results in an external pressure on banks not to classify loans as
non-performing. The internal pressure originates from the banks’ top
management, which tends to avoid public fund injections associated with
restructuring conditions (not least because it may reduce their own retirement
bonuses).
Japanese banks maintain not sufficient loan-loss reserves. This happens
due to the lenient reserving policy of the FSA (Fukao, 2002) and restrictive tax
deduction guidelines on loan-loss specific provisions as well as unrealistically low
limit on loan-loss general provisions (Kanaya and Woo, 2000). Comparing to the
US banks, which maintain loan-loss reserve to bad loan ratio of more than 160%,
Japanese banks’ ratio has been in range of 40-60% since 1994.
2.2 Weak Capital Position
Under lenient supervision that has prevailed despite reorganisation, the
BIS capital rules often have been manipulated. First, lower reserves for non-
performing loans increase the banks’ capital base. Second, banks are allowed to
keep large amounts of deferred taxes as a part of their capital. However, in order
to get the deferred taxes back, banks should be profitable. Since Japanese banks
have been losing money during the post-bubble period, the deferred taxes should
not appear in their capital base. It is said that a stricter application of this rule
contributed to the failure of Resona Group in May 2003
10
. Finally, cross holding of
shares between banks and insurance companies is shown as capital although the
shares are nothing more than mutually held debt. Fukao (2002) made an
estimation of the Japanese banks capital base with strict application of the BIS
rules. According to his estimation, in 2001 four out of fifteen major banks had
negative capital base, while the capital base at the remaining eleven banks was
less than 2%.
Historically, Japanese banks faced no limits on the number of corporate
share they could hold. Large share holdings are very sensitive to stock price
fluctuations. The situation became very serious when the Nikkei 225 fell below
10,000 yen in 2002 and fluctuated in the range of 8,000 yen till summer 2003.
Banks having difficulties with maintaining the BIS capital rules reduce their
lending creating a credit crunch that results in financial distress for non-financial
corporations, further reduces the value of shares held by banks and causes even
more difficulties for banks themselves.
2.3 Low Profitability
The results released in March 2003 show that most of the banks have
deficit
11
. Most of Japanese banks have been unprofitable for the last ten years.
They have been surviving mainly by realising capital gain from shares and real
estate sales. There are a few reasons for the banks’ current weak profitability. One
11
is that Japanese banks did not undertake serious restructuring until late 1990s.
Not only the number of banks did not decreased significantly despite heavy
overcapacity in banking industry (contrary to the USA where the number of banks
decreased by 15%, Norway by 10%, and Sweden by 80% during first two years
after the banking crises began), but the employment increased by 3% (the USA –
5%, Norway –25%, and Sweden –8%), and the personnel costs increased by more
than 15% in the first half of 1990s (the USA and Norway –15%, Sweden –12%,
Iwatsubo and Hino, 1998). The intensive cost reducing at the end of 1990s
reduced the operating costs but the pace of improvement is certainly
unsustainable because of long overdue investment in infrastructure, without
which banks will further lose their market share to convenience stores and foreign
financial institutions
12
.
Second, deposit interest rate controls that existed until the early 1990s
allowed banks to maintain relatively high lending margins. However after the
controls had been lifted, the deteriorating market conditions prevented banks
from significant increase of their lending rates. While real interest rates (lending
rates adjusted for inflation rates) remain high for borrowers due to deflationary
environment in Japanese economy, they have been hardly covering banks’
operating costs.
Another reason for banks’ poor profitability is the fact that they have to
compete with government sponsored financial institutions. Those institutions hold
almost 40% of a market. It is hard to compete with housing loan programmes that
offer longer maturity and accept prepayment without penalties. In the deposit
market, the postal savings system has more than 24,000 offices around the
country (against 600 of the largest banking group Mizuho), does not charge
account maintenance fees, and offers deposit interest rates similar to those of
private financial institutions. Needless to say, the government sponsored financial
institutions enjoy authorities’ full guarantee.
Finally, technological change resulted in emergence of new types of banks.
Those include, for instance, the IY Bank, which specializes in payment and
settlement services for individual customers. Established in May 2001 by the
largest Japanese supermarket chain Ito Yokado, the bank operates by using ATMs
installed in the chain’s convenience stores. Others, like Sony Bank, eBank or
Japan Net Bank, established by trading companies, electronic makers,
information services companies etc. offer bank services via the Internet or portable
phone networks. This makes the traditional banks’ competitive position even more
dramatic. The environment of the Japanese financial market prevents banks from
restoring their profitability.
2.4 Macroeconomic Effects of the Banking Crisis
Banking crisis is responsible for serious macroeconomic difficulties the
Japanese economy has been facing. Struggling to improve their capital base and
profitability, banks reduce their lending and contribute to further deterioration of
economic conditions through a credit crunch. Instead of lending to private
businesses, banks choose the purchase of government bonds and effectively
prevent passing through the effects of the BOJ expansionary monetary policies (it
is the government, not private businesses that end holding the increased money
supply). Despite of growing deposits, banks’ lending was declining on average
3.4% between June 2001 and June 2003. This makes a bulk of advice on the use
of expansionary monetary policies or inflation targeting in Japan at least
impractical (for instance Bernanke, 2003). Such policies cannot succeed without
healthy banking industry, unless the BOJ directly lends to private businesses
what is not a commonly accepted practice for a central bank.
12
Except for impairing the effectiveness of the BOJ monetary policies, the
banking crisis has also other effects. First, the currency to bank deposit ratio has
been growing steadily
13
indicating lack of confidence in banking system. At the
same time the share of postal deposits to total deposits almost doubled. Given
poorer effectiveness of the investment of postal deposits (Iwatsubo and Hino,
1998, estimate that postal deposits are at least 20% less effective in generating
growth) this may contribute substantially to the severity of the current recession.
Second, banks’ attitude impairs other policies aimed at revitalising businesses as
for instance the Industrial Reconstruction Agency launched in May 2003 to buy-
back some of non-performing loans of small and medium enterprises in local
economies.
On the other hand, Miyajima and Yafeh (2003) see some positive aspects
of the banking crisis in Japan. They argue that credit crunch resulting from the
crisis affects mainly small low profit companies in low-tech sectors with limited
access to bond markets. On the contrary, companies with high R&D investment in
high-technology sectors can easily find their way to direct financing. In this way
the banking crisis contributes to the “natural selection” of firms and banks and
therefore helps to transform the Japan’s heavily bank oriented financial markets
into a more direct financing system.
3. Solution to the Crisis
The solution to the crisis should include two groups of measures. One is
strictly addressed to banks and should include tackling non-performing loans
problem and improving bank supervision. The other should comprise of
macroeconomic policy measures aiming at revitalising the economy.
Unfortunately, despite of highly publicised policy steps not very much has been
changed in relations between banks and their regulators. Even now, under the
FSA supervision banks are allowed for under-disclosure of the non-performing
loans and poor profitability. Only in July 2003, the FSA issued the profitability
warnings to several banks for the first time ever. The implicit policy of not
interfering into the internal management of the banks that have received public
funds adopted by the FSA results in management of the banks not being held
responsible for their actions. Announcements of not interfering into the
rehabilitation of de facto nationalised Resona Group seems to confirm this policy
trend. But such attitude has been further undermining public confidence in
financial sector and may result in even stronger opposition to using public funds
in future. The government must stringently asses the real amount of non-
performing loans, make banks to put sufficient loan-loss reserves, make public
funds injections without the banks’ management consent, and under the
conditions of effective nationalisation, rehabilitate them
14
. This is how Ryutaro
Komiya once described the role of the main bank in the economy. Now the banks
themselves need the same service.
But cleaning the banks sheets of non-performing loans alone is not
sufficient. Decisive actions must be undertaken in order to restore a sound
economic growth without which the banks cannot improve their profitability. The
Japanese government tried to revitalise the economy through such policies as
promoting information technology-led growth (called an “IT revolution”),
liberalisation and opening of the Japanese economy (including the highly
publicised “Japanese Big Bang”), stimulating stock market prices in order to
support the value of banks’ shareholdings, and massive public spending. All of
them ultimately failed to live to the expectations because seemingly none of these
13
attempts was actually a consumer-oriented policy. Most of regulatory changes as
well as fiscal stimuli are strictly business-oriented measures aimed at providing
relief not even to the economy as a whole but to particular industries. Most of
them do not improve an individual consumer’s welfare. The same can be argued
about the changes to the pension system, health insurance, social security,
environment protection regulations and so on. In reality, the policy measures had
rather adverse effects on consumer spending, while not providing enough relief to
businesses.
Increased spending on public works projects failed to stimulate individual
consumption because Japanese public is aware of negative fiscal effects of such
projects and about their rather limited rationality. Numerous scandals of bid
rigging, corruption among bureaucrats and politicians involved in public work
projects, poor quality of construction work, and involvement of organised crime
further undermine the confidence of Japanese public which seems to view those
spending as a waste of public funds rather than genuine steps leading to an
economic recovery. This, not less than deflation itself, discourages consumer
spending. Anecdotal evidence says that, for the same reason as consumers, the
companies participating in public works projects use leased machinery and
equipment and hire poor quality part-time labour force instead of making capital
investment and hiring full time employees as the primary objective of the public
spending ought to be.
In order to restore public confidence, it is necessary to make substantial
changes restoring rationality to the national pension system, health insurance,
education and other areas of public spending. At the same time, in order to make
a one-time boost to consumer spending, 0% consumption tax rate on certain
goods (e.g. housing, vehicles, electric and electronic appliances and so on) for a
limited period of time (for instance one year) could be introduced. Because the CPI
deflation rate is still lower than the consumption tax rate, it could result in a
substantial increase in consumer spending. On the contrary, the increase of
consumption tax rate, as advocated by some of the business leaders, may rather
result in 1997-like recession with similar results for the economy in general and
banking system in particular. Given rapidly dissipating public confidence and
much worse government’s fiscal position, the recovery may be even more difficult
than last time. However, reducing its transfers for thousands of government and
semi-government agencies of questionable usefulness, that receive hefty subsidies,
or at least do not pay taxes, the Japanese government has still enough room for a
substantial fiscal improvement. Therefore the adverse fiscal effects of such a
consumption tax reduction need not to be catastrophic.
No less important is resolving the problems arising from the presence of
the government sponsored financial institutions. This especially applies to the
postal saving services, which currently rule over more than 20% of total fund-
raising and is effectively crowding banks out from their primary deposit market.
But privatisation of postal services, promoted as a trademark for structural
reforms, is not likely to provide a sound solution to the problem. If simply
privatised, as the current administration views it, the postal savings would
become a near monopoly player on the country’s deposit market, several times
bigger than the biggest of banking groups. Additionally, transforming postal
services into ordinary banking business would require finding solutions to its
insufficient capital base (currently its own capital is less than one percent of its
total assets, the remaining part would have to be provided by injection of public
funds) and building its lending facilities from naught. As the experience teaches
this is not a simple task. Needless to say, this would require a substantial fiscal
effort while creating at the same time a lot of confusion on the financial markets.
14
Without addressing this issue, a fragile recovery in the banking industry may be
seriously damaged.
4. Conclusion
Except for its extraordinary length, the banking crisis in Japan is very
typical in all aspects. Therefore the lessons that can be drawn from its experience
are also very typical, although quite often overlooked.
1/ Though it is often viewed as a crisis that triggered a change, in fact the banking
crisis in Japan is the product of a change that resulted in a crisis. Poor
coordination and sequencing of liberalisation allowed for “regulatory arbitrage”
(unequal treatment of different institutions engaging in similar activities) while
failure to address the overcapacity in the banking industry led to “survival
myopia” (engaging in increasingly risky activities in order to survive on the
market). Similarly, a specific Japanese corporate governance system (including
the presence of a main bank, weak corporate management, low profitability etc.)
that worked almost flawlessly in highly regulated environment of a post war
growth era had to fail if confronted with new conditions of (partially) liberalised
economy. The origins of this failure can be traced to the simplistic view of a
market mechanism as a simple demand-supply game. On the contrary, viewing
market as an internally coherent set of institutions that have to be created could
result in much better prepared reforms.
2/ It is very difficult to perform economic policies in isolation. Although perhaps
not directly responsible for the banking crisis in Japan, the Plaza Accord led to
series of events that ultimately triggered the crisis and aggravated it depth. The
ability to foresee what a particular policy may actually result in is a very
important but often neglected element of policy making. Making short-term ad
hoc international policy commitments without undertaking proper adjustment
has continued also later as liberalisation of imports (often blamed for the
ongoing consumer price deflation) within the WTO framework proves.
3/ Many policy mistakes made by authorities may be attributed to the specific
character of policy making in Japan. The policy making process is under control
of bureaucrats and therefore the quality of bureaucratic cadres is essential to its
effectiveness. The central bureaucracy is largely comprised of graduates from
undergraduate schools of law (it should be noted that in Japanese education
system it means approximately two years of general education and two years of
rather general studies of law). Bureaucracy is generally closed to economists
(Heizo Takenaka’s appointment as Minister for Finance and Economy is a very
rare exception) and for that reason economics, finance and economic policy are
self-taught subjects at the highest level of Japanese decision-making process
(Cargill, 2000). Involving country’s economists in decision-making process, as it
happens in other countries, might have spared Japan some of its current
problems.
15
NOTES
1 Posen (2002).
2 For instance Bayoumi (1998), Sekine (1999), Woo (1999) and others.
3 Japanese banks’ obsession with their market share has been in a sense a
remnant of the overly regulated financial system. Under the interest rate
controls, the banks’ lending spreads were more or less fixed and the size of their
net income was determined by the sheer size of their outstanding loans (Kanaya
and Woo, 2000).
4 The agricultural cooperatives were under the supervision of the Ministry of
Agriculture and the limits on lending to real estate did not apply to them.
5 The largest creditor group of the Jusen companies, the agricultural cooperatives
were charged with only minor part of the bail out costs due to the political
pressure from the Ministry of Agriculture (Kanaya and Woo, 2000).
6 The FSA took over the supervision of banks, insurance and security companies,
and non-bank financial institutions from the MOF, Shinkin credit associations
from Regional Financial Bureaus and credit cooperatives from local
governments. As one of the first changes, the FSA started investigating the non-
performing loans on its own; until then banks calculated them themselves. In
July 2000, the Financial Supervision Agency was renamed as the Financial
Services Agency and from January 2001 took over the duties of the Financial
Reconstruction Commission.
7 The fifth one, the Resona Group was launched in March 2003 only to file for
protection two months later.
8 Much of the criticism directed at the BOJ policies during the bubble economy is
based on our current knowledge of the events that took place later and our
much-improved modern analytical tools. It is however very hard to estimate how
aware of the problem were the officials at that time. The discussion on the
responsibility for creating the asset bubble economy in Japan is far from being
closed, just to mention recent papers such as Bank of Japan (2000), Kosai, Ito
and Arioka (2000), Shiratsuka (2000), Bernanke and Gertler (2001), Okina and
Shiratsuka (2001) and so on.
9 The practice, where top bureaucrats assume management posts in private
businesses after retirement contributes greatly to the supervision failures (it
happens that government controllers have to control their own mentors) and
policy mistakes. Although the practice of amakudari seems to have subsided in
recent years, still between 5.7% (regional banks) and 9.6% (second tier regional
banks) of top management posts is occupied by former bureaucrats (Asahi
Shinbun, September 13, 2003).
10 In March 2003 (end of fiscal year 2002), deferred taxes accounted for 60.8% of
Mizuho Group own capital (in September the share decreased to 43.6%), for
58.7% (51.5% in September) in Mitsui Sumitomo, 41.6% (26.8% in September)
in Mitsubishi Tokyo, 59.4% (51.6% in September) in UFJ, and 99.3% (12.6% in
September) in Resona. The September low figures for Resona are due to public
funds injection (Asahi Shinbun, Nov. 26, 2003).
16
11 Although the half-year results released in November 2003 show substantial
improvement (except for Resona Group) there remain certain doubts about
sustainability of the recovery in the banking industry (Asahi Shinbun, Nov. 26,
2003).
12 Because of outdated fund transfer system, unable to accept Chinese characters
banks cannot compete with 24-hours bar code reading terminals at convenience
store chains.
13 It has grown from 7.1% in the first quarter of 1993 to 9.1% in the first quarter
of 2002 (Iwatsubo and Hino, 1998, Posen, 2002).
14 This is more or less exactly what happened in Korea after the 1997 crisis. See
Nakai (2003) for comparison of Japanese and Korean approaches to banking
crises.
REFERENCES
Bank of Japan (2000): “A Symposium on the Monetary Policy During the Bubble
Period”, Institute for Monetary and Economic Studies (IMES) Discussion Paper
(in Japanese).
Bayoumi, Tamim (1998): “The Morning After: Explaining the Slowdown in
Japanese Growth in the 1990s”, IMF Working Paper.
Bernanke, Ben (2003): “Some Thoughts on Monetary Policy in Japan”, a speech
delivered before the 60
th
Anniversary Meeting of the JSME, Tokyo, May 31,
2003.
Bernanke, Ben, and Mark Gertler (2001): “Should Central Banks Respond to
Movements in Asset Prices?” American Economic Review, 91 (2), pp. 253-257.
Cargill, Thomas F. (2000): “What Caused Japan’s Banking Crisis?” in Hoshi and
Patrick (2000).
Fujiwara, Kenya (1998): “The Fragility of Banking System – Issues of
Rehabilitation”, in Ishigaki and Hino (1998).
Fukao, Mitsuhiro (2002): “Barriers to Financial Restructuring: Japanese Banking
and Life-Insurance Industries”, paper presented at the Conference on “East
Asian Monetary and Financial Co-operation; Concepts, Policy Prospects and the
Role of the Yen”, Hamburg Institute for International Economics, May 29, 2002.
Horie, Yasuteru (2003): “Loan Classification and the Scale of Non-Performing
Loans”, paper presented at the annual conference of the JSME, Tokyo May 31,
2003 (in Japanese).
Hoshi, Takeo and Hugh Patrick (eds.) (2000): Crisis and Change in the Japanese
Financial System, Kluwer Academic Publishers.
Ishigaki, Ken-ichi and Hiroyuki Hino (eds.) (1998): “Restoring Sound Banking
System in Japan”, Kobe University Research Institute for Economics and
Business Administration (in Japanese).
Iwatsubo, Kamon and Hiroyuki Hino (1998): “Banking Crisis and Macroeconomy”,
in Ishigaki and Hino (1998).
17
Jinushi, Toshiki, Yoshihiro Kuroki, and Ryuzo Miyao (2000): “Monetary Policy in
Japan Since the Late 1980s: Delayed Policy Actions and Some Explanations”,
in: Mikitani and Posen (2000).
Kanaya, Akihiro and David Woo (2000): “The Japanese Banking Crisis of the
1990s: Sources and Lessons”, IMF Working Paper.
Kosai, Yutaka, Osamu Ito, and Ritsuko Arioka (2000): “Reconsidering the
Monetary Policy During the Bubble Period”, IMES Discussion Paper (in
Japanese).
Lee, Jongsoo (1999): “The ‘crisis’ of non-performing loans”, in Tsutsui (1999).
Mayes, David, Liisa Halme and Aarno Liuksila (2001): Improving Banking
Supervision, Palgrave.
Mikitani, Ryoichi and Adam S. Posen (2000): Japan’s Financial Crisis and its
Parallels to US Experience”, Institute for International Economics, Washington,
DC.
Miyajima, Hideaki and Yishay Yafeh (2003): “Japan’s Banking Crisis: Who has the
Most to Lose?” Waseda University Financial Paper no. 03/02.
Nakai, Hiroyuki (2003): “Dealing with Bad Loans in Japan: A Comparative
Assessment”, paper presented at the annual conference of the JSME, Tokyo May
31, 2003 (in Japanese).
Okina, Kunio and Shigenori Shiratsuka (2001): “Asset Price Bubbles, Price
Stability, and Monetary Policy; Japan’s Experience”, IMES Discussion Paper.
Posen, Adam (2002): “The Looming Japanese Crisis”, International Economics
Policy Brief 02-5, Institute for International Economics.
Sekine, Toshitaka (1999): “Firm Investment and Balance Sheet Problems in
Japan”, IMF Working Paper.
Shimizu, Yoshinori (1999): “Problems in the Japanese Financial System in the
Early 1990s”, in William M. Tsutsui (1999).
Shimizu, Yoshinori (2000): “Convoy Regulation, Bank Management, and the
Financial Crisis in Japan”, in: Mikitani and Posen (2000).
Shiratsuka, Shigenori (2000): “Asset Prices, Financial Stability, and Monetary
Policy: Based on Japan’s Experience of the Asset Price Bubble”, IMES
Discussion Paper.
Tsutsui, William M. (ed.) (1999): Banking in Japan, Routledge, London.
Taniguchi, Tomohiko (1999): “Japan’s Banks and the “Bubble Economy” of the
late 1980s”, in William M. Tsutsui (1999).
Woo, David (1999): In Search of Credit Crunch: Supply Factors Behind the
Slowdown in Japan”, IMF Working Paper.
doc_409007546.pdf
The weakness of the Japanese banking industry, suffering from acute problem of non-performing loans, prevents Japan from restoring sound growth rates despite having undertaken structural reforms and substantial fiscal policy efforts, and, through impairing transmission channels of monetary policy, it has also made ineffective efforts to stimulate the economy through “zero interest rates” and quantitative easing policy.
Change and Crisis
in the Japanese
Banking Industry
Mariusz K. Krawczyk
HWWA DISCUSSION PAPER
277
Hamburgisches Welt-Wirtschafts-Archiv (HWWA)
Hamburg Institute of International Economics
2004
ISSN 1616-4814
Hamburgisches Welt-Wirtschafts-Archiv (HWWA)
Hamburg Institute of International Economics
Neuer Jungfernstieg 21 - 20347 Hamburg, Germany
Telefon: 040/428 34 355
Telefax: 040/428 34 451
e-mail: [email protected]
Internet:http://www.hwwa.de
The HWWA is a member of:
• Wissenschaftsgemeinschaft Gottfried Wilhelm Leibniz (WGL)
• Arbeitsgemeinschaft deutscher wirtschaftswissenschaftlicher Forschungsinstitute
(ARGE)
• Association d’Instituts Européens de Conjoncture Economique (AIECE)
HWWA Discussion Paper
Change and Crisis in the
Japanese Banking Industry
Mariusz K. Krawczyk
HWWA Discussion Paper 277http://www.hwwa.de
Hamburg Institute of International Economics (HWWA)
Neuer Jungfernstieg 21 - 20347 Hamburg, Germany
e-mail: [email protected]
This Discussion Paper has been prepared in collaboration with the HWWW-Research
Programme “Development und Integration”, while the author was a guest at the
HWWA.
Edited by the Department World Economy
Head: PD Dr. Carsten Hefeker
HWWA DISCUSSION PAPER 277
May 2004
Change and Crisis in the
Japanese Banking Industry
Abstract
The weakness of the Japanese banking industry, suffering from acute problem of
non-performing loans, prevents Japan from restoring sound growth rates despite having
undertaken structural reforms and substantial fiscal policy efforts, and, through impair-
ing transmission channels of monetary policy, it has also made ineffective efforts to
stimulate the economy through “zero interest rates” and quantitative easing policy. Mis-
understanding the roots of the banking crisis contributed greatly to its exceptional length
and depth and prevented its early solution. Poor coordination and sequencing of liberali-
sation of financial services together with macroeconomic policy mistakes have been re-
sponsible for the crisis. But the origins of those mistakes can be traced to the bureau-
cratic character of the country’s policy-making process often lacking a comprehensive
analysis of macroeconomic effects of a particular policy decision. Simple solution to the
banking crisis in the form of eliminating non-performing loans burden, bailing out ail-
ing financial institutions etc., if not accompanied by decisive steps restoring rationality
to the public spending and resolving the problems arising from the presence of the gov-
ernment sponsored financial institutions, may not be sufficient to achieve sound growth
and prevent similar crises in the future.
JEL-Classification:.G21, G28
Keywords: Japanese economy, banking crisis, financial liberalisation
Mariusz K. Krawczyk
Faculty of Economics
Fukuoka University, 8-19-1 Nanakuma, Jonan-ku, Fukuoka 814-0180
Japan
E-mail: [email protected]
1
For more than 10 years Japan has not been able to overcome the
repeating sequence of mild recessions followed by periods of stagnation. An
average real growth rate of approximately 1 percent per year between 1991 and
2001 was the lowest among advanced industrial countries. Negative growth rates,
falling individual consumption, historically high unemployment, and fast growing
public debt stand in such a sharp contrast with relatively stable performance
before 1980s, that Japanese often call the 1990s their “lost decade”.
The depth and length of the Japan’s stagnation - the costliest recession
suffered by any advanced industrial economy after the World War II - has led
some to predict that, till now very conservative, Japanese savers may even flee
into cash, gold, and offshore accounts (equal to thing until now unimaginable; to
capital flight from Japan)
1
. Although, there are numerous causes why the
Japanese economic performance continues to disappoint, the weakness of the
country’s banking industry is one of the main factors behind the inability to break
this prolonged cycle of recessions. For a large part of the last decade Japan has
experienced a steady deterioration of its banking sector condition. The decline of
the Japanese banking industry began at the end of the 1980s asset bubble and
turned into a full-blown crisis in the second of half of 1997 when a number of
high profile financial institutions collapsed. There is no significant improvement
since then, despite of enormous amounts of capital allocated for strengthening
and revitalising the financial system (approximately 60 trillion yen spent in last
five years what amounts to 12% of the country’s GDP). This contrasts sharply to
the S&L (savings and loan association) crisis of the late 1980s in the USA or to the
Nordic Banking Crisis of the early 1990s, where the recovery started 2-3 years
after the crises had begun. Ailing banking industry prevents Japan from restoring
sound growth rates despite having undertaken structural reforms and substantial
fiscal policy efforts, and, through impairing transmission channels of monetary
policy, it has been also one of the reasons why the Bank of Japan (BOJ) policy of
quantitative easing has failed to stimulate the economy
2
.
Banking crisis is not a phenomenon unique to Japan. On the contrary,
the basic aspects of the Japanese banking crisis are almost typical of banking
crises in general (Mayes et al., 2001). Its roots can be traced to major regulatory
changes towards liberalisation that was poorly coordinated between sectors of the
economy, a period of rapid economic growth accompanied by an excessive asset
expansion, unsustainable macroeconomic policies, and a substantial adverse
shock. It is argued here that misunderstanding the origins of the crisis, lack of
sense of urgency, weak corporate governance, poor supervision and regulatory
forbearance when the system came under stress led to the worsening of the crisis
and its exceptional length. The solution to the crisis should not be viewed in
isolation from the processes taking place in the rest of the economy and must not
be reduced to a mere resolving the non-performing debt problem.
The remaining of the paper is organised as follows. Section 1 provides a
brief overview of the banking crisis and sums up the main factors that contributed
to the crisis, e.g. the regulatory framework as well as banks’ and authorities’ poor
management of the crisis. Section 2 presents the current condition and the main
problems the banking industry is facing. Section 3 defines the conditions
necessary for the recovery of the banking industry against the background of
economic conditions prevailing in Japan. Finally, concluding section provides
some more general lessons that can be drawn from the experience of the Japanese
banking crisis.
2
1. A Brief Overview of the Japanese Banking Crisis
1.1 The Prelude: Asset Inflated Economy
The emergence of the bubble economy can be attributed to a combination
of external pressure, ongoing domestic liberalisation, and policy mistakes made by
the central authorities. Growing consensus that the US dollar was unrealistically
overvalued against the other currencies (especially the Japanese yen) resulted in
coordinated policy effort to devalue the US currency known as Plaza Accord
(September 22, 1985). Consequently, the dollar lost 45% of its value between
February 1985 (when it reached a peak of 260 yen/dollar) and May 1987 (141
yen/dollar). The move, primarily aimed at putting cap on the US trade deficit with
Japan, imposed a heavy toll on the Japanese economy known as the “High-Yen
Recession”. The real economic growth dropped to –0.5%, its first negative rate
since the oil shock and the total unemployment rate reached historically high level
of 3.1% in May 1987. Concerned with already growing at that time public debt
(50.3% of the GNP), the authorities chose monetary expansion with reduction of
the official discount rates hoping for a wealth effect that would increase domestic
consumption and investment and, in the end, boost the real economic growth.
The scenario worked as expected. In February 1987, the BOJ reduced its
discount rate to a historically low 2.5% level and kept it there for two years till the
mid-1989. The result was an increase in the money supply (defined as M2 plus
CD) that reached 10.8% in 1987 and remained on average 11% until 1990. As
shown in Figure 1, the policy of easy money supported investment and asset
prices increase. The stock prices increased threefold (Nikkei 225 increased from
13,024 in January 1983 to 38,916 in December 1989) and land prices increased
fourfold during the same time. Meanwhile, the BOJ concentrated its anti-
inflationary vigilance on the consumer price movements, which, thanks to the yen
appreciation, were becoming increasingly distorted by the decreasing prices of
tradable goods. The Bank’s policies largely ignored the surge in asset prices. The
unprecedented monetary policy easing coincided with massive privatisation
programmes performed by the Nakasone administration (Japan National Railway
and Nippon Telephone and Telegraph) and partial liberalisation of financial
markets. These two factors too contributed to the asset bubble getting out of
control.
Figure 1: Asset Prices in Japan (1983 = 100)
0
100
200
300
400
500
600
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
Nominal GDP Nikkei 225 Land prices
Source: Japan Institute for Social and Economic Affairs
3
The asset bubble was abruptly brought to an end in the second half of
1989 when the BOJ started increasing its discount rate. The share prices
collapsed in summer 1990 (Nikkei 225 dropped from 38,915 in December 1989 to
20,984 in September 1990). Meanwhile, to contain the rise in land prices, the
Ministry of Finance (MOF) introduced limits on total amounts of loans to real
estate. The limits resulted in sudden contraction of lending to real estate and were
lifted in January 1992, after the land prices had started to decline. The overheated
economy gradually slowed down.
1.2 Banks’ Response to the Asset Bubble
The banks actively participated in the exacerbation of asset bubble by
credit creation. The rapid increase in land prices and real estate related
businesses shares led to a lending boom towards this kind of businesses. On the
contrary, lending to large manufacturing corporations was on decline. While the
share of lending to big manufacturing companies to total lending decreased from
almost 60% in 1985 to less than 50% in 1990 and 40% in 1991, the share of
construction and real estate lending increased from 14% to 18% during the same
period (Fujiwara, 1998). The lending to real estate related businesses was based
on misguided expectation that land prices would increase indefinitely (as they did
during all post war years except for 1975). The sharp decline of land prices
following the end of the asset bubble caused the quality of loans to the real estate
industry to deteriorate rapidly. Banks themselves were to blame for lack of proper
risk evaluation. Lending offices were using a simple “direct capital comparison
method” e.g. comparing the published land prices of neighbouring properties (only
later they began calculating the value of land collateral based on future income
Interest Rate Controls in Japan
The Temporary Interest Rate Adjustment Law was introduced in 1947.
It allowed the Bank of Japan (under the Ministry of Finance supervision) to
establish detailed interest rate ceilings on deposits and on short-term lending.
The Law applied to all kind of banks and credit cooperatives (however the
Shinkin small- and medium-sized businesses credit associations, agricultural
cooperatives, and the credit cooperatives were exempt from the restrictions on
lending rates). Government financial institutions (including postal savings)
were not subject to the regulations. As a result all financial institutions applied
almost the same interest rates. The variation of the interest rates remained
very small even after gradual relaxation of the controls (a significant part of the
controls on lending rates was removed in 1959) as banks tended to apply a
uniform rates within the Federation of Bankers Association. The same
happened with long-term lending rates despite of the fact that formally the
long-term rates were not a subject to interest rate controls.
During 1980s the controls on deposit interest rates were gradually
relaxed. In 1979, certificates of deposits (CD) became exempt from the interest
rate controls. Large deposits (more than 1 billion yen) became exempt from
controls in 1988 and their minimum amount was later reduced and finally
abolished. Not until 1971 were the ordinary banks allowed to accept deposits
of maturity longer than one year (in 1981 they were allowed to accept deposits
of maturity three years). Only since October 1993, floating rates and maturity
longer than three years have been permitted. Before that, long-term credit
banks had a monopoly on long-term deposit activities. The interest rate
controls were finally abolished in September 1994.
4
flow), they extended their lending disproportionately to the value of collateral, their
lax policies allowed for using the same collateral for multiple borrowing etc.
Table 1: Credit Ratings of Major Japanese Banks
Bank 3/90 3/95 3/96 3/97 3/98 3/99 3/00 3/01 3/02
Industrial Bank of J.
AAA A+ A A A- BBB BBB+ BBB+ A
Long Term Credit B.
1 AA A- BBB+ BBB+ BBB- BB- BBB- BBB- BBB-
Nippon Credit Bank
2 AA+ A BBB- BB+ BB+ BB- BB- BB BBB-
Dai-Ichi Kangyo
AA A+ A A BBB+ A A A BBB
Sakura
AA+ AA- A- A- A- BBB BBB A- A-
Fuji
AA A+ A- A- BBB+ A A+ A+ BBB
Mitsubishi
3 AA+ AA- A+ A+ A A- A- A- BBB+
Asahi
AA+ A A A A BBB BBB BB+ BB+
Sanwa
AA AA- A A A BBB+ BBB+ BBB+ BBB
Sumitomo
AA+ A+ A A A- BBB BBB BBB+ BBB
Daiwa
AA+ A- BBB+ BBB+ BBB- BB+ BB+ BB+ BB+
Tokai
AAA A A A A BBB- BBB- BBB A
Hokkaido
Takushoku
AA A BBB- BBB- BBB-
Tokyo
3 AA+ AA+ AA+ AA+
Notes:
1
since 1999 Shinsei Bank,
2
since 1999 Aozora Bank,
3
in 1996 merged into Bank of Tokyo Mitsubishi.
For mergers after 1999 see Table 2.
Source: Miyajima and Yafeh (2003).
Large increase in lending to real estate is often viewed as a primary cause
of the non-performing loans problems that emerged later in 1990s. However it
would be a great oversimplification to attribute all the problems of the banking
industry only to the irrational lending behaviour during the late 1980s. On the
contrary, it is possible to argue that Japanese banks were left with no choice but
to undertake risky lending activities in order to survive. Their very unusual
lending practices can be linked to two characteristics of the Japanese banking
industry at that time. First, partially due to the pressure from the US government,
the Japanese authorities embarked on the protracted process of liberalisation of
the country’s financial system. By mid-1980s the process was only half finished
with gradual removal of restrictions on access to the corporate bond market and
creation of the commercial paper market on the one hand but still numerous
restrictions on financial institutions activities in place (see box on liberalisation of
the interest rates controls). These developments had serious consequences.
Incipient liberalisation triggered competition for customers that further cut down
already low profit margins of Japanese banks (Shimizu, 1999). At the same time
large corporations, until then the banks’ main borrowers used their newly
acquired freedom of access to corporate bond markets. Enjoying very high credit
ratings, the corporations were able to borrow directly from the market at
significantly lower cost. Japanese corporations raised approximately $519 billion
in first five years after lifting their bond market restriction (Taniguchi, 1999). The
long-term credit banks, which enjoyed most protection under the interest rate
controls, were among hardest hit by the new reality. Facing declining profits and
loosing their traditional corporate customers, the banks had no choice but to look
for substitute lending markets. They increased their loan activity not only for the
5
real estate industry, but also for small and medium size enterprises, individuals,
finance industry, and overseas investment (Shimizu, 1999, 2000).
Second, the Japanese banking industry suffered from huge overcapacity
problems. The industry was designed to provide financing for the Japanese
economy that was growing on average 9.7% between 1960 and 1973. At the same
time the average growth in Europe and the USA was 4.8% and 3.9% respectively.
Slowing growth (on average 4.4% between 1975 and 1985) and decreasing
demand for funds in the second half of 1970s and in 1980s was not accompanied
by the reduction of the banking sector capacity. Fierce competition for lending
under heavy overcapacity in the banking industry and banks’ special attention to
their market shares
3
caused a kind of cognitive dissonance e.g. in order to survive
banks had to compete for borrowers even if it meant willingly loosening credit
standards. The myopic character of the lending patterns reached almost farcical
dimensions with banks engaging in phoney realtor activities in exchange for kick
backs or lending more than $1.8 billion to a small restaurant owner. Fully rational
change in lending patterns, which resulted from changes in regulatory
environment, reached at that stage a state of irrational exuberance.
It was however still 5 years before the banking sector soundness came to
question. During that time the amount of non-performing loans was steadily
increasing. Banks were extremely reluctant to raise their loan-loss provisions.
This might have happened either due to the banks’ concern that increase in loan-
loss provisions would be providing markets with negative expectations about
future increases of non-performing loans, or because individual banks did not
want to draw attention to themselves by braking ranks with other banks. Banks
could not write-off their non-performing loans due to restrictive tax regulations
that allowed for write-offs only in cases of bankruptcy or foreclosure proceedings.
Given the average speed of legal proceedings in Japan it substantially slowed
down the writing-off non-performing loans. Close relationship between banks and
their borrowers, known as the main bank system might have been yet another
important reason why banks were slow in cleaning their balance sheets of non-
performing loans. The main bank may be reluctant to identify troubled borrower
because this in turn may raise questions about its own monitoring abilities and
because the main bank would be usually required to absorb the main bulk of the
losses incurred by creditors. When the main bank itself comes under stress it may
result in bank becoming a captives of its own borrowers. Finally, poor corporate
governance of Japanese banks might have been one of the main factors behind
the growing amount of non-performing loans. The weak corporate governance of
Japanese banks originates from the fact that banks’ major shareholders have little
incentive to confront the management. Main groups of shareholders have been
either dependent on management (banks’ employees and banks’ corporate
borrowers) or had close relationship with management (other banks and
insurance companies due to cross shareholding). Members of boards of directors
are usually promoted from the ranks of employees at the end of their careers and
resign after their terms expire in order to be replaced by junior employees. This,
combined with weak internal and external audits, creates incentives for concealing
problems rather than taking decisive steps during manager’s tenure. During 5
years following the collapse of the bubble economy, Japanese banks chose not to
make determined effort to solve mounting non-performing loans problem. Instead,
they preferred to wait for stock and land prices to recover to the bubble economy
level.
6
1.3 Banking Crisis
1.3.1 Instability Begins (early 1995 – Fall 1997)
The instability in Japanese banking industry came into public view in
early 1995. Two insolvent credit cooperatives (Tokyo Kyowa and Anzen) collapsed
at the end of 1994 and further two (Cosmo and Kizu) followed by August 1995. In
August 1995 Hyogo Bank, one of regional banks, was ordered to suspend its
operations. The closure of insolvent financial institutions marked the end of policy
not allowing depository institutions to fail. Adding insult to injury, the loss
concealing scandal forced Daiwa Bank to close its operations in the US later in the
year. Those events coincided with the controversy surrounding the Jusen (the
housing loan) companies.
The Jusen companies, established in 1970s by banks and other financial
institutions for lending in home mortgage market, engaged in real estate lending
in late 1980s and at the beginning of the 1990s. With funds from agricultural
cooperatives the Jusen companies increased their lending to real estate
businesses filling the gap left by commercial banks that had to follow restrictions
on lending to real estate imposed by the MOF
4
. Already by early 1992 the Jusen
companies lending quality raised serious concerns but it was only in 1995 when
authorities decided to step in. By that time, three quarters of Jusen loans were
already non-performing.
Soon the Jusen problem became a major political issue. It was the first
time that the government attempted to use public funds for solving problems of
weak financial institutions. For that reason, the discussion surrounding the Jusen
companies set an important precedent in dealing with such problems. Facing a
clear public dislike to the idea of using taxpayers’ money for rescuing banks, the
authorities reached a consensus that problems of financial institutions should be
solved possibly without direct involvement of the government. In Jusen case, 55%
(or 3.5 trillion yen) of their losses had to be covered by their founding banks, 27%
(1.7 trillion yen) by the lending banks, 8% (530 billion yen) by the agricultural
financial institutions, and the remaining 10% (680 billion yen) by the public
funds
5
.
The collapse of insolvent financial institutions together with the Jusen
confusion heavily damaged the image of Japanese banking industry. It was then
that the “Japan Premium” (premium on lending to Japanese institutions)
appeared for the first time on international financial markets and international
rating agencies, already downgrading Japanese banks since 1989, radically
reduced the credit ratings for 1996 (while there was no major Japanese bank with
B ratings in 1995, a year later four of them were classified as BBB and only one
reached AA rating). Despite of that, inaction and regulatory forbearance of
financial authorities continued until 1997, substantially undermining the public
confidence in country’s banking industry.
1.3.2 Crisis (Fall 1997 – mid 1999)
The increase of consumption tax rate in April 1997 marked the beginning
of another recession. The economic conditions deteriorated and by November 1997
Japanese financial industry faced several spectacular bankruptcies. On November
3, 1997, Sanyo Securities, a second tier security company defaulted on the inter-
bank loan market. It was followed by the collapse of Yamaichi Securities, one of
the country’s big four securities companies, and by Hokkaido Takushoku Bank,
one of the, then 12, city banks. Those events resulted in serious disruptions on
the inter-bank market and sell-off of banks’ shares on the Tokyo Stock Exchange.
Unable to continue its non-interference policy, the government established, under
newly enacted Financial Stabilisation Law, a 30 trillion yen fund for capital
7
injection into troubled banks. The Law was an important novelty in authorities’
response to banking problems. Loosely modelled after the US Prompt Corrective
Action regulations, the Law provided introduction of well-defined self-assessment
procedures and their external audit for banks and, no less important, it clearly
defined the capital threshold ratios for authorities to intervene in an individual
bank’s management affairs (restrictions on dividend payments, management
bonuses, closure of branches, suspending operations etc.). It reduced the scope
for regulatory inaction so often exercised before 1997.
In March 1998, four months after the crisis had begun, the newly created
Financial Crisis Management Committee supervised capital injection of 1.8 trillion
yen for 21 major banks in the form of subordinate debt. The injection of public
funds was intended to help the banks to meet their capital requirements. In order
to avoid drawing attention to weaker banks, all banks applied for the same
amount (100 billion yen) of capital. Calm did not last however long. The collapse of
the Long Term Credit Bank, the largest bank failure in post-war Japan, brought
new waves of instability. In October 1998, the Financial Function Early
Strengthening Law replaced the earlier Financial Stabilisation Law. The new Law
was designed to help banks with insufficient capital. The Diet authorised the use
of 25 trillion yen for this purpose. 15 major banks received 7.4 trillion yen already
at the beginning of 1999. Contrary to the 1998 public fund injection, this time
banks were required to submit rehabilitation plans and the amount of public
capital varied between banks. The capital injection took form of preferred stock
that could be converted into common stock and used for exerting pressure on
bank’s management if the implementation of a rehabilitation plan was not
satisfactory. Of major banks, only the Bank of Tokyo Mitsubishi, the biggest and
soundest of Japanese banks at that time, turned down the offer of public funds.
The second law introduced in October 1998, the Financial Reconstruction
Law, was designed to deal with failed banks. Under this law, 17 trillion yen was
set aside on special account to become a deposit guarantee for failed banks, and
another 18 trillion was held for purchasing shares (nationalisation) of failed banks
and for supporting the Resolution and Collection Corporation (RCC) to purchase
non-performing loans. Finally, admitting that banks’ supervision had not been
sufficient, the Financial Supervisory Agency (FSA) was established and took over
supervision of banks from the MOF
6
in June 1998.
Under the new legal framework, the Financial Reconstruction Committee
decided to place under public administration (that means effective nationalisation)
the Long-Term Credit Bank and the Nippon Credit Bank (late 1998), two banks
that used to be indispensable participants of Japan’s rapid economic growth in
the early post-war period.
1.3.3 Merger Wave (1999-2003)
Capital injection under the Financial Function Early Strengthening Law
required receiving banks to clean up their balance sheets of the non-performing
loans and strengthen their capital positions. Restructuring and cost cutting was
an obvious way to become profitable and be able to return public funds as soon as
possible. This triggered a series of mergers. 23 large banks that existed in 1985
have been regrouped into four groups (Table 2)
7
. The number of regional banks
also slowly decreased from 133 to 118 (mainly in the second half of 1990s).
The long awaited liberalisation of financial services (the Japanese “Big
Bang” Plan) finally started being implemented. Under the new rules, the
restrictions that once separated banking, securities, and insurance businesses
have been lifted. In order to survive in a new environment, banks had to look for
new partners.
8
Table 2: Consolidation of the Banking Industry
Date of Merger Merging Parties New Financial Group
September
2000
Dai-Ichi Kangyo Bank, Fuji Bank, Industrial
Bank of Japan
Mizuho Holdings
April 2001 Sakura Bank, Sumitomo Bank
Sumitomo Mitsui
Banking Corp. (SMBC)
April 2001
Bank of Tokyo Mitsubishi, Mitsubishi Trust
and Banking, Nippon Trust Bank, Tokyo Trust
Bank
Mitsubishi Tokyo
Financial
Group, Inc.
April 2001
Sanwa Bank, Tokai Bank, Toyo Trust and
Banking
UFJ (United Financial of
Japan) Group
March 2003 Asahi Bank, Daiwa Bank Resona Group
1.4 What Caused the Crisis?
As the above story of unfolding banking crisis in Japan shows there is no
single cause that triggered the meltdown of the industry. Instead it is possible to
identify a set of factors that contributed to the worsening of the crisis.
Undoubtedly, monetary policy mistakes by the BOJ in the second half of the
1980s generated asset inflation and provided excessive liquidity to the financial
system that was unprepared for absorbing it
8
. It is also possible to argue, as
Jinushi et al. (2000) do, that too tight monetary policy at the beginning of the
1990s further aggravated the crisis that was already well under way.
Second, the Japanese financial system was an instrument of industrial
policy aimed at supporting reindustrialisation, supporting investment and export-
led growth, protecting domestic businesses from international competition and
providing liquidity at possibly low cost. The set of objectives produced a highly
bank-dependent very rigid system full of administrative controls and guidance.
The system fully met its objectives and served well during the period of high
growth of Japanese economy in the 1960s. The authorities were very slow however
to modify the system once it accomplished its mission (Cargill, 2000, Shimizu,
2000). Japan, Germany, and France are usually given as examples of bank-
oriented financial systems. But while the rate of bank deposits in Germany
decreased from 67.9% in 1971 to 39.3% in 1999 as shown in Figure 2, the
Japan’s ratio of bank deposits to total financial assets remained almost
unchanged (62.9% in 1998). As a result, at the time the asset inflated bubble
economy arrived, Japan still relied on its highly bank-dependent but only partially
liberalised and lacking transparency financial system. This leads to the third
factor, extremely poor supervision and regulatory forbearance of Japanese
financial authorities. There might have been various reasons of the authorities
poor performance ranging from the lack of political leadership, to cosy relationship
between regulators and regulated that originated from amakudari
9
, to simply poor
quality of staff working on banking issues (see conclusions at the end of the
paper), or to the fourth factor; the opposition from the taxpaying public to use
public funds for dealing with ailing financial institutions (the opposition that
became even firmer when series of scandals at the MOF and at the BOJ was
revealed).
9
Figure 2: Breakdown of Financial Assets in the Household Sector
0%
20%
40%
60%
80%
100%
Germany France UK USA J apan
country
stock
trust banking
insurance andpension
bonds
bank deposit
Source: Comparative Economic and Financial Statistics, BOJ
Perhaps each of the factors partially contributed to the depth and the
length of the crisis. These factors can fully explain the unfolding of the banking
crisis in Japan and there is no need for “revisionist” theories claiming that the
Japanese bureaucrats intentionally planned and implemented the crisis as for
instance Lee (1999) does.
2. The Condition of the Japanese Banking Industry
2.1 Increasing Non-performing Loans
Japanese banks accumulated loss of approximately 88.2 trillion yen due
to disposing of non-performing loans. Despite of this enormous loss they still have
more than 38 trillion yen of disclosed non-performing loans (Table 3). This
amounts to more than 5% of their portfolio, yet there are some analysts who argue
that the disclosed amount of non-performing loans is rather underestimated. For
instance, Fukao (2002) points out that the rules of the FSA Bank Examination
Manual contain too narrow definition of non-performing loans. Others, like Horie
(2003), argue that there is a bias towards big companies in estimating their
financial position and therefore also overestimation of the performance of funds
loaned to these companies.
Table 3: Loss on Disposal of Bad Loans in All Banks (\ billion)
Source: the FSA
10
The prolonged recession is partially to blame for the increasing amount of
non-performing loans since more and more firms face financial distress. Their
loans increase the amount of non-performing loans. On the other hand banks
tend to willingly underestimate their non-performing loans positions. Peculiarities
of Japanese corporate governance mentioned in the preceding section aside,
banks have been under heavy external and internal pressure. Within the keiretsu
(industrial group) framework, which big Japanese firms still tend to prefer,
classifying a certain loan as risky, puts not only a particular borrower but also a
group as a whole in a difficult position. And bank, itself being a member of such a
group can hardly afford to alienate its other members. Next, disposing of a bad
loan and placing a borrower under bankruptcy means a reduction of tax income
for local authorities. This, combined with the presence of “politically well-
connected firms”, results in an external pressure on banks not to classify loans as
non-performing. The internal pressure originates from the banks’ top
management, which tends to avoid public fund injections associated with
restructuring conditions (not least because it may reduce their own retirement
bonuses).
Japanese banks maintain not sufficient loan-loss reserves. This happens
due to the lenient reserving policy of the FSA (Fukao, 2002) and restrictive tax
deduction guidelines on loan-loss specific provisions as well as unrealistically low
limit on loan-loss general provisions (Kanaya and Woo, 2000). Comparing to the
US banks, which maintain loan-loss reserve to bad loan ratio of more than 160%,
Japanese banks’ ratio has been in range of 40-60% since 1994.
2.2 Weak Capital Position
Under lenient supervision that has prevailed despite reorganisation, the
BIS capital rules often have been manipulated. First, lower reserves for non-
performing loans increase the banks’ capital base. Second, banks are allowed to
keep large amounts of deferred taxes as a part of their capital. However, in order
to get the deferred taxes back, banks should be profitable. Since Japanese banks
have been losing money during the post-bubble period, the deferred taxes should
not appear in their capital base. It is said that a stricter application of this rule
contributed to the failure of Resona Group in May 2003
10
. Finally, cross holding of
shares between banks and insurance companies is shown as capital although the
shares are nothing more than mutually held debt. Fukao (2002) made an
estimation of the Japanese banks capital base with strict application of the BIS
rules. According to his estimation, in 2001 four out of fifteen major banks had
negative capital base, while the capital base at the remaining eleven banks was
less than 2%.
Historically, Japanese banks faced no limits on the number of corporate
share they could hold. Large share holdings are very sensitive to stock price
fluctuations. The situation became very serious when the Nikkei 225 fell below
10,000 yen in 2002 and fluctuated in the range of 8,000 yen till summer 2003.
Banks having difficulties with maintaining the BIS capital rules reduce their
lending creating a credit crunch that results in financial distress for non-financial
corporations, further reduces the value of shares held by banks and causes even
more difficulties for banks themselves.
2.3 Low Profitability
The results released in March 2003 show that most of the banks have
deficit
11
. Most of Japanese banks have been unprofitable for the last ten years.
They have been surviving mainly by realising capital gain from shares and real
estate sales. There are a few reasons for the banks’ current weak profitability. One
11
is that Japanese banks did not undertake serious restructuring until late 1990s.
Not only the number of banks did not decreased significantly despite heavy
overcapacity in banking industry (contrary to the USA where the number of banks
decreased by 15%, Norway by 10%, and Sweden by 80% during first two years
after the banking crises began), but the employment increased by 3% (the USA –
5%, Norway –25%, and Sweden –8%), and the personnel costs increased by more
than 15% in the first half of 1990s (the USA and Norway –15%, Sweden –12%,
Iwatsubo and Hino, 1998). The intensive cost reducing at the end of 1990s
reduced the operating costs but the pace of improvement is certainly
unsustainable because of long overdue investment in infrastructure, without
which banks will further lose their market share to convenience stores and foreign
financial institutions
12
.
Second, deposit interest rate controls that existed until the early 1990s
allowed banks to maintain relatively high lending margins. However after the
controls had been lifted, the deteriorating market conditions prevented banks
from significant increase of their lending rates. While real interest rates (lending
rates adjusted for inflation rates) remain high for borrowers due to deflationary
environment in Japanese economy, they have been hardly covering banks’
operating costs.
Another reason for banks’ poor profitability is the fact that they have to
compete with government sponsored financial institutions. Those institutions hold
almost 40% of a market. It is hard to compete with housing loan programmes that
offer longer maturity and accept prepayment without penalties. In the deposit
market, the postal savings system has more than 24,000 offices around the
country (against 600 of the largest banking group Mizuho), does not charge
account maintenance fees, and offers deposit interest rates similar to those of
private financial institutions. Needless to say, the government sponsored financial
institutions enjoy authorities’ full guarantee.
Finally, technological change resulted in emergence of new types of banks.
Those include, for instance, the IY Bank, which specializes in payment and
settlement services for individual customers. Established in May 2001 by the
largest Japanese supermarket chain Ito Yokado, the bank operates by using ATMs
installed in the chain’s convenience stores. Others, like Sony Bank, eBank or
Japan Net Bank, established by trading companies, electronic makers,
information services companies etc. offer bank services via the Internet or portable
phone networks. This makes the traditional banks’ competitive position even more
dramatic. The environment of the Japanese financial market prevents banks from
restoring their profitability.
2.4 Macroeconomic Effects of the Banking Crisis
Banking crisis is responsible for serious macroeconomic difficulties the
Japanese economy has been facing. Struggling to improve their capital base and
profitability, banks reduce their lending and contribute to further deterioration of
economic conditions through a credit crunch. Instead of lending to private
businesses, banks choose the purchase of government bonds and effectively
prevent passing through the effects of the BOJ expansionary monetary policies (it
is the government, not private businesses that end holding the increased money
supply). Despite of growing deposits, banks’ lending was declining on average
3.4% between June 2001 and June 2003. This makes a bulk of advice on the use
of expansionary monetary policies or inflation targeting in Japan at least
impractical (for instance Bernanke, 2003). Such policies cannot succeed without
healthy banking industry, unless the BOJ directly lends to private businesses
what is not a commonly accepted practice for a central bank.
12
Except for impairing the effectiveness of the BOJ monetary policies, the
banking crisis has also other effects. First, the currency to bank deposit ratio has
been growing steadily
13
indicating lack of confidence in banking system. At the
same time the share of postal deposits to total deposits almost doubled. Given
poorer effectiveness of the investment of postal deposits (Iwatsubo and Hino,
1998, estimate that postal deposits are at least 20% less effective in generating
growth) this may contribute substantially to the severity of the current recession.
Second, banks’ attitude impairs other policies aimed at revitalising businesses as
for instance the Industrial Reconstruction Agency launched in May 2003 to buy-
back some of non-performing loans of small and medium enterprises in local
economies.
On the other hand, Miyajima and Yafeh (2003) see some positive aspects
of the banking crisis in Japan. They argue that credit crunch resulting from the
crisis affects mainly small low profit companies in low-tech sectors with limited
access to bond markets. On the contrary, companies with high R&D investment in
high-technology sectors can easily find their way to direct financing. In this way
the banking crisis contributes to the “natural selection” of firms and banks and
therefore helps to transform the Japan’s heavily bank oriented financial markets
into a more direct financing system.
3. Solution to the Crisis
The solution to the crisis should include two groups of measures. One is
strictly addressed to banks and should include tackling non-performing loans
problem and improving bank supervision. The other should comprise of
macroeconomic policy measures aiming at revitalising the economy.
Unfortunately, despite of highly publicised policy steps not very much has been
changed in relations between banks and their regulators. Even now, under the
FSA supervision banks are allowed for under-disclosure of the non-performing
loans and poor profitability. Only in July 2003, the FSA issued the profitability
warnings to several banks for the first time ever. The implicit policy of not
interfering into the internal management of the banks that have received public
funds adopted by the FSA results in management of the banks not being held
responsible for their actions. Announcements of not interfering into the
rehabilitation of de facto nationalised Resona Group seems to confirm this policy
trend. But such attitude has been further undermining public confidence in
financial sector and may result in even stronger opposition to using public funds
in future. The government must stringently asses the real amount of non-
performing loans, make banks to put sufficient loan-loss reserves, make public
funds injections without the banks’ management consent, and under the
conditions of effective nationalisation, rehabilitate them
14
. This is how Ryutaro
Komiya once described the role of the main bank in the economy. Now the banks
themselves need the same service.
But cleaning the banks sheets of non-performing loans alone is not
sufficient. Decisive actions must be undertaken in order to restore a sound
economic growth without which the banks cannot improve their profitability. The
Japanese government tried to revitalise the economy through such policies as
promoting information technology-led growth (called an “IT revolution”),
liberalisation and opening of the Japanese economy (including the highly
publicised “Japanese Big Bang”), stimulating stock market prices in order to
support the value of banks’ shareholdings, and massive public spending. All of
them ultimately failed to live to the expectations because seemingly none of these
13
attempts was actually a consumer-oriented policy. Most of regulatory changes as
well as fiscal stimuli are strictly business-oriented measures aimed at providing
relief not even to the economy as a whole but to particular industries. Most of
them do not improve an individual consumer’s welfare. The same can be argued
about the changes to the pension system, health insurance, social security,
environment protection regulations and so on. In reality, the policy measures had
rather adverse effects on consumer spending, while not providing enough relief to
businesses.
Increased spending on public works projects failed to stimulate individual
consumption because Japanese public is aware of negative fiscal effects of such
projects and about their rather limited rationality. Numerous scandals of bid
rigging, corruption among bureaucrats and politicians involved in public work
projects, poor quality of construction work, and involvement of organised crime
further undermine the confidence of Japanese public which seems to view those
spending as a waste of public funds rather than genuine steps leading to an
economic recovery. This, not less than deflation itself, discourages consumer
spending. Anecdotal evidence says that, for the same reason as consumers, the
companies participating in public works projects use leased machinery and
equipment and hire poor quality part-time labour force instead of making capital
investment and hiring full time employees as the primary objective of the public
spending ought to be.
In order to restore public confidence, it is necessary to make substantial
changes restoring rationality to the national pension system, health insurance,
education and other areas of public spending. At the same time, in order to make
a one-time boost to consumer spending, 0% consumption tax rate on certain
goods (e.g. housing, vehicles, electric and electronic appliances and so on) for a
limited period of time (for instance one year) could be introduced. Because the CPI
deflation rate is still lower than the consumption tax rate, it could result in a
substantial increase in consumer spending. On the contrary, the increase of
consumption tax rate, as advocated by some of the business leaders, may rather
result in 1997-like recession with similar results for the economy in general and
banking system in particular. Given rapidly dissipating public confidence and
much worse government’s fiscal position, the recovery may be even more difficult
than last time. However, reducing its transfers for thousands of government and
semi-government agencies of questionable usefulness, that receive hefty subsidies,
or at least do not pay taxes, the Japanese government has still enough room for a
substantial fiscal improvement. Therefore the adverse fiscal effects of such a
consumption tax reduction need not to be catastrophic.
No less important is resolving the problems arising from the presence of
the government sponsored financial institutions. This especially applies to the
postal saving services, which currently rule over more than 20% of total fund-
raising and is effectively crowding banks out from their primary deposit market.
But privatisation of postal services, promoted as a trademark for structural
reforms, is not likely to provide a sound solution to the problem. If simply
privatised, as the current administration views it, the postal savings would
become a near monopoly player on the country’s deposit market, several times
bigger than the biggest of banking groups. Additionally, transforming postal
services into ordinary banking business would require finding solutions to its
insufficient capital base (currently its own capital is less than one percent of its
total assets, the remaining part would have to be provided by injection of public
funds) and building its lending facilities from naught. As the experience teaches
this is not a simple task. Needless to say, this would require a substantial fiscal
effort while creating at the same time a lot of confusion on the financial markets.
14
Without addressing this issue, a fragile recovery in the banking industry may be
seriously damaged.
4. Conclusion
Except for its extraordinary length, the banking crisis in Japan is very
typical in all aspects. Therefore the lessons that can be drawn from its experience
are also very typical, although quite often overlooked.
1/ Though it is often viewed as a crisis that triggered a change, in fact the banking
crisis in Japan is the product of a change that resulted in a crisis. Poor
coordination and sequencing of liberalisation allowed for “regulatory arbitrage”
(unequal treatment of different institutions engaging in similar activities) while
failure to address the overcapacity in the banking industry led to “survival
myopia” (engaging in increasingly risky activities in order to survive on the
market). Similarly, a specific Japanese corporate governance system (including
the presence of a main bank, weak corporate management, low profitability etc.)
that worked almost flawlessly in highly regulated environment of a post war
growth era had to fail if confronted with new conditions of (partially) liberalised
economy. The origins of this failure can be traced to the simplistic view of a
market mechanism as a simple demand-supply game. On the contrary, viewing
market as an internally coherent set of institutions that have to be created could
result in much better prepared reforms.
2/ It is very difficult to perform economic policies in isolation. Although perhaps
not directly responsible for the banking crisis in Japan, the Plaza Accord led to
series of events that ultimately triggered the crisis and aggravated it depth. The
ability to foresee what a particular policy may actually result in is a very
important but often neglected element of policy making. Making short-term ad
hoc international policy commitments without undertaking proper adjustment
has continued also later as liberalisation of imports (often blamed for the
ongoing consumer price deflation) within the WTO framework proves.
3/ Many policy mistakes made by authorities may be attributed to the specific
character of policy making in Japan. The policy making process is under control
of bureaucrats and therefore the quality of bureaucratic cadres is essential to its
effectiveness. The central bureaucracy is largely comprised of graduates from
undergraduate schools of law (it should be noted that in Japanese education
system it means approximately two years of general education and two years of
rather general studies of law). Bureaucracy is generally closed to economists
(Heizo Takenaka’s appointment as Minister for Finance and Economy is a very
rare exception) and for that reason economics, finance and economic policy are
self-taught subjects at the highest level of Japanese decision-making process
(Cargill, 2000). Involving country’s economists in decision-making process, as it
happens in other countries, might have spared Japan some of its current
problems.
15
NOTES
1 Posen (2002).
2 For instance Bayoumi (1998), Sekine (1999), Woo (1999) and others.
3 Japanese banks’ obsession with their market share has been in a sense a
remnant of the overly regulated financial system. Under the interest rate
controls, the banks’ lending spreads were more or less fixed and the size of their
net income was determined by the sheer size of their outstanding loans (Kanaya
and Woo, 2000).
4 The agricultural cooperatives were under the supervision of the Ministry of
Agriculture and the limits on lending to real estate did not apply to them.
5 The largest creditor group of the Jusen companies, the agricultural cooperatives
were charged with only minor part of the bail out costs due to the political
pressure from the Ministry of Agriculture (Kanaya and Woo, 2000).
6 The FSA took over the supervision of banks, insurance and security companies,
and non-bank financial institutions from the MOF, Shinkin credit associations
from Regional Financial Bureaus and credit cooperatives from local
governments. As one of the first changes, the FSA started investigating the non-
performing loans on its own; until then banks calculated them themselves. In
July 2000, the Financial Supervision Agency was renamed as the Financial
Services Agency and from January 2001 took over the duties of the Financial
Reconstruction Commission.
7 The fifth one, the Resona Group was launched in March 2003 only to file for
protection two months later.
8 Much of the criticism directed at the BOJ policies during the bubble economy is
based on our current knowledge of the events that took place later and our
much-improved modern analytical tools. It is however very hard to estimate how
aware of the problem were the officials at that time. The discussion on the
responsibility for creating the asset bubble economy in Japan is far from being
closed, just to mention recent papers such as Bank of Japan (2000), Kosai, Ito
and Arioka (2000), Shiratsuka (2000), Bernanke and Gertler (2001), Okina and
Shiratsuka (2001) and so on.
9 The practice, where top bureaucrats assume management posts in private
businesses after retirement contributes greatly to the supervision failures (it
happens that government controllers have to control their own mentors) and
policy mistakes. Although the practice of amakudari seems to have subsided in
recent years, still between 5.7% (regional banks) and 9.6% (second tier regional
banks) of top management posts is occupied by former bureaucrats (Asahi
Shinbun, September 13, 2003).
10 In March 2003 (end of fiscal year 2002), deferred taxes accounted for 60.8% of
Mizuho Group own capital (in September the share decreased to 43.6%), for
58.7% (51.5% in September) in Mitsui Sumitomo, 41.6% (26.8% in September)
in Mitsubishi Tokyo, 59.4% (51.6% in September) in UFJ, and 99.3% (12.6% in
September) in Resona. The September low figures for Resona are due to public
funds injection (Asahi Shinbun, Nov. 26, 2003).
16
11 Although the half-year results released in November 2003 show substantial
improvement (except for Resona Group) there remain certain doubts about
sustainability of the recovery in the banking industry (Asahi Shinbun, Nov. 26,
2003).
12 Because of outdated fund transfer system, unable to accept Chinese characters
banks cannot compete with 24-hours bar code reading terminals at convenience
store chains.
13 It has grown from 7.1% in the first quarter of 1993 to 9.1% in the first quarter
of 2002 (Iwatsubo and Hino, 1998, Posen, 2002).
14 This is more or less exactly what happened in Korea after the 1997 crisis. See
Nakai (2003) for comparison of Japanese and Korean approaches to banking
crises.
REFERENCES
Bank of Japan (2000): “A Symposium on the Monetary Policy During the Bubble
Period”, Institute for Monetary and Economic Studies (IMES) Discussion Paper
(in Japanese).
Bayoumi, Tamim (1998): “The Morning After: Explaining the Slowdown in
Japanese Growth in the 1990s”, IMF Working Paper.
Bernanke, Ben (2003): “Some Thoughts on Monetary Policy in Japan”, a speech
delivered before the 60
th
Anniversary Meeting of the JSME, Tokyo, May 31,
2003.
Bernanke, Ben, and Mark Gertler (2001): “Should Central Banks Respond to
Movements in Asset Prices?” American Economic Review, 91 (2), pp. 253-257.
Cargill, Thomas F. (2000): “What Caused Japan’s Banking Crisis?” in Hoshi and
Patrick (2000).
Fujiwara, Kenya (1998): “The Fragility of Banking System – Issues of
Rehabilitation”, in Ishigaki and Hino (1998).
Fukao, Mitsuhiro (2002): “Barriers to Financial Restructuring: Japanese Banking
and Life-Insurance Industries”, paper presented at the Conference on “East
Asian Monetary and Financial Co-operation; Concepts, Policy Prospects and the
Role of the Yen”, Hamburg Institute for International Economics, May 29, 2002.
Horie, Yasuteru (2003): “Loan Classification and the Scale of Non-Performing
Loans”, paper presented at the annual conference of the JSME, Tokyo May 31,
2003 (in Japanese).
Hoshi, Takeo and Hugh Patrick (eds.) (2000): Crisis and Change in the Japanese
Financial System, Kluwer Academic Publishers.
Ishigaki, Ken-ichi and Hiroyuki Hino (eds.) (1998): “Restoring Sound Banking
System in Japan”, Kobe University Research Institute for Economics and
Business Administration (in Japanese).
Iwatsubo, Kamon and Hiroyuki Hino (1998): “Banking Crisis and Macroeconomy”,
in Ishigaki and Hino (1998).
17
Jinushi, Toshiki, Yoshihiro Kuroki, and Ryuzo Miyao (2000): “Monetary Policy in
Japan Since the Late 1980s: Delayed Policy Actions and Some Explanations”,
in: Mikitani and Posen (2000).
Kanaya, Akihiro and David Woo (2000): “The Japanese Banking Crisis of the
1990s: Sources and Lessons”, IMF Working Paper.
Kosai, Yutaka, Osamu Ito, and Ritsuko Arioka (2000): “Reconsidering the
Monetary Policy During the Bubble Period”, IMES Discussion Paper (in
Japanese).
Lee, Jongsoo (1999): “The ‘crisis’ of non-performing loans”, in Tsutsui (1999).
Mayes, David, Liisa Halme and Aarno Liuksila (2001): Improving Banking
Supervision, Palgrave.
Mikitani, Ryoichi and Adam S. Posen (2000): Japan’s Financial Crisis and its
Parallels to US Experience”, Institute for International Economics, Washington,
DC.
Miyajima, Hideaki and Yishay Yafeh (2003): “Japan’s Banking Crisis: Who has the
Most to Lose?” Waseda University Financial Paper no. 03/02.
Nakai, Hiroyuki (2003): “Dealing with Bad Loans in Japan: A Comparative
Assessment”, paper presented at the annual conference of the JSME, Tokyo May
31, 2003 (in Japanese).
Okina, Kunio and Shigenori Shiratsuka (2001): “Asset Price Bubbles, Price
Stability, and Monetary Policy; Japan’s Experience”, IMES Discussion Paper.
Posen, Adam (2002): “The Looming Japanese Crisis”, International Economics
Policy Brief 02-5, Institute for International Economics.
Sekine, Toshitaka (1999): “Firm Investment and Balance Sheet Problems in
Japan”, IMF Working Paper.
Shimizu, Yoshinori (1999): “Problems in the Japanese Financial System in the
Early 1990s”, in William M. Tsutsui (1999).
Shimizu, Yoshinori (2000): “Convoy Regulation, Bank Management, and the
Financial Crisis in Japan”, in: Mikitani and Posen (2000).
Shiratsuka, Shigenori (2000): “Asset Prices, Financial Stability, and Monetary
Policy: Based on Japan’s Experience of the Asset Price Bubble”, IMES
Discussion Paper.
Tsutsui, William M. (ed.) (1999): Banking in Japan, Routledge, London.
Taniguchi, Tomohiko (1999): “Japan’s Banks and the “Bubble Economy” of the
late 1980s”, in William M. Tsutsui (1999).
Woo, David (1999): In Search of Credit Crunch: Supply Factors Behind the
Slowdown in Japan”, IMF Working Paper.
doc_409007546.pdf