Skip to content

Japan’s Banking Structure


The challenge to shoulder the financial burdens of Fukushima is posed to a financial system which is still in the process of  recovering from a long-lasting economic and financial crisis that resulted in a fundamental restructuring of the financial industry.  Whether and to which extent Japan’s financial institutions will succeed in smoothly fulfilling the extraordinary task to provide the huge amounts needed for compensation, reconstruction and economic recovery, and channel them into the right uses, will also depend on the question to which extent the system has digested the preceding crisis. A look at the transformation of the Japanese banking structure will provide some clues.

From the bubble economy to recovery

In the 1980s Japanese financial markets experienced a dramatic rise in land and stock prices. This triggered a speculative bubble which gradually seized the whole economy. Financial institutions of all sorts were involved and suffered badly when in 1990 the bubble burst. In the restructuring process that followed many did not survive. Others sought shelter under the roof of the giant holding companies that emerged in the orchestrated effort of government, bureaucrats and industry to restore financial soundness and stability.

Going backwards in tracing the transformation process from the current state to the “pre-holdings” era at the end of the 1990s (compare Japanese Bankers Association) demonstrates the various steps of this process.

These days, the Japanese banking system consists mainly of few big holdings or “Mega Banking Groups” as the Japanese Banking Association calls them:

These holding structures are a rather new phenomenon. As you can see from the slide below, for 2002, we find most existing in one form or another, some of them under a different name.  Among the big players, UFJ Holdings and Mitsubishi Tokyo Financial Group were still two separate entities:

One year earlier, in 2001, restructuring was very much in flux. There were already some holdings such as Mizuho Holdings and Daiwa Bank Holdings, but names and composition differed from later years and there were still some independent banks as remainders of the old structures:

In 1999, the restructuring process had just started and many of the old structures were still intact:

(A ppt-file of all four graphs for a more lucid impression is here.)

In the former system, there were no holding companies, only different categories of banks:

  • big commercial or city banks,
  • long-term credit banks,
  • trust banks,
  • regional banks and regional banks II (the former mutual savings banks or sôgô ginkô),
  • credit associations,
  • credit cooperatives,
  • central cooperative banks (of which the one for agriculture and forestry, Norinchukin, is still among the biggest international players)
  • and one specialized foreign exchange bank, Bank of Tokyo, the successor of the Yokohama Specie Bank.

For the details I would like to refer you to my book on the Japanese Foreign Exchange Market, chapter 2.

In addition, there were countless nonbank financial institutions such as money lenders, leasing firms and real estate agents which during the bubble became the vehicles for speculation allowing banks to circumvent official regulations.

In the transformation process, numbers varied considerably in the different categories as the following table demonstrates. Since 1990, the number of city banks more than halved, there are lesser regional banks II than before and long-term credit banks ceased to exist. On the other hand, the number of trust banks rose before somewhat declining again and a wholly new category, simply named other banks, emerged that did not exist before:

Numbers of Japanese Banks

End of March





City Banks





Regional Banks





Regional Banks II





Trust Banks





Long-term Credit Banks





Other banks



This category includes two institutions which are directly linked to the transformation process, the Second Bridge Bank and the Resolution and Collection Corporation (RCC). Both were  established as subsidiaries  of the Deposit Insurance Corporation of Japan (DICJ) and are in charge of disposing of undesirable remainders of the financial crisis. Details … Their very existence is a first indication of still prevailing unsolved problems.

Changes in banking structures also altered industrial structures in Japan. Traditionally, different kinds of banks are strongly linked to the big Japanese industrial groups or keiretsu. Those are seemingly loose groupings of manufacturers, distributors and finance companies characterised by cross shareholdings, regular meetings of executives from member firms, interlocking directorships and countless unwritten obligations (compare Ito 1992).  In 1990, the eight biggest groups contained about 900 companies – plus banks and insurance companies. Each keiretsu can be imagined as representing a cross-section of the economy, and the banks fulfill special tasks in this network. The most important ones are city banks that serve as so-called main bank for their respective groupings. Beside providing funding they exert a strong informal influence on members and assume far-reaching responsibilities.

There are two different kinds of keiretsu, old and new ones. The old ones have their roots in the pre-war zaibatsu, large conglomerates which were said to exert virtually perfect control of the economy up to their break-up after World War II. This group consists, above all, of the “Big Three”, Sumitomo, Mitsubishi and Mitsui.

A look at the prevailing structure of financial holdings shows that despite many changes, much of these old relations apparently survived. It is also an indication that the old ways of policy making by “administrative guidance” exploiting these relations and the accompanying close ties between government institutions and private firms may be still more or less intact. This raises the question how successful the transformation has been so far and whether, at least to some extent, recent institutional changes served to disguise remaining weaknesses rather than contributing to their elimination.

The later worries are indirectly confirmed by the results of a paper by Maximilian J.B. Hall who studied the challenges of the Japanese banking sector several years ago. On the one hand, he found signs of recovery. Banks’ overall performance in terms of profitability, asset quality and capital adequacy had improved and NPL (non-performing loans) ratios were declining.  On the other hand, core profitability remained weak, lending margins were “wafer thin”, loan demand from corporations was still sluggish, diversification from traditional lending activities appeared insufficient and the same held for the share of fee and commission incomes. Land prices had not generally recovered and banks had experienced new losses from exposures to the US sub-prime market. In addition, Hall noticed a remarkable still prevailing regulatory tolerance. In his view, a stricter interpretation of BIS rules would make it necessary to

  • fully deduct some of banks’ loan loss provisions from capital;
  • deduct a greater proportion of cross-shareholdings from capital to overcome “double-gearing”;
  • further restrict the inclusion of deferred tax assets (DTAs) and
  • make banks get a more realistic view on the value of loan collateral and borrowers’ capacity of repay with respective consequences for reported capital ratios.

If this impression holds true, Japan’s financial system appears badly prepared for the latest challenges and authorities might well regret not to have shown more courage when there was only one crisis to cope with.

From → Policies

Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: