Systemic Banking Problem: A Japanese Experience

I have been reading excellent analyses by Chris Whalen on potential systemic banking crisis. I missed the RGE Monitor Panel Discussion, but I enjoyed reading the report by Naked Capitalism. I find Whalen’s and others’ argument that a systemic banking crisis involving many U.S. banks is likely, starting with the failure of the Bank of America Corp. (BAS)’s attempt to absorb Countrywide Financial Corp (CFC) convincing.

The Wall Street Journal (May 29, 2008) reports that BAC is pushing aside David Sambol, chief operating officer of CFC, and having BAC’s chief technology officer Barbara Desoer to head the combined mortgage operation after BAC acquires CFC. The article argues “Speculation has swirled for months that the deal could fall apart, but putting Ms. Desoer in charge is a sign that Bank of America is turning its attention toward how to squeeze value out of Countrywide.” And adds “Bank of America officials increasingly believe that the Calabasas, Calif., lender is more valuable because of its broker and branch network than its corporate culture. As a result, the bank indicated the Countrywide brand name is likely to disappear at some point.” It seems still possible that with further scrutiny BAC concludes that it is more efficient for them to force CFC to restructure through a formal bankruptcy procedure before absorbed into BAC, as Whalen has been suggesting.

Assuming Whalen’s scenario is the right one, it will be very important for CFC’s bankruptcy to proceed smoothly. The case of CFC will be an important test case for a lot more insolvency cases that will follow. As Whalen concludes his blog on May 12, 2008,

And a Chapter 7 filing by a creditor of CFC will prove once and for all that a large bank holding company can go through a market-based resolution without a subsidy from Washington. For that reason alone, we’ll buy dinner at Sparks Steakhouse for the holder of CFC debt that pulls the trigger first.

I wrote in my blog on May 13, 2008, there are many examples of “not-to-do’s” in the Japanese experience with the banking problems. There are, however, some positive lessons as well. One of them concerns a systemic approach to the problem of systemic insolvency in the banking sector. The example comes from the Japanese experience in 1948.

No, it is not a typo. It is nineteen FORTY eight, sixty years ago. After the World War II, many Japanese companies found themselves insolvent, because the government decided to repudiate wartime liabilities to those corporations as well as wartime guarantees on their bank loans. The total losses are estimated to have been 92 billion yen, almost 20% of GNP in 1946 (and we should note the level of GNP used here is much smaller than its prewar peak). This meant that many major industrial firms in Japan were near insolvent and almost all the banks which lent to them were insolvent.

Japanese government and the Occupational Force (led by the U.S.) implemented a system-wide scheme to solve the problem. Here I give a very brief description of the restructuring of the banks under the scheme. I once wrote a longer and much more detailed description of the scheme (“Cleaning up the Balance Sheets: Japanese Experience in the Postwar Reconstruction Period,” in Masahiko Aoki and Hyung-Ki Kim (Eds) Corporate Governance in Transitional Economies, The World Bank, 1995).

First, the balance sheet of each bank was separated into two parts as of August 11, 1946. The “old account” included the assets that may be uncollectible as a result of suspension of wartime compensation and (on the other side) unprotected portion of the deposits and the capital. The “new account” included the assets that are safe (such as cash and government bonds), interbank loans, and protected deposits. The idea was to allow banks to continue operation using the new accounts while the old accounts are restructured, so that the system-wide restructuring does not interfere with the working of the financial system very much.

Then, the old account was reorganized. The assets were re-evaluated, and the liabilities were reduced to match the re-evaluated level of assets. When the liabilities and capital in the old account was insufficient to cancel the losses in the assets, the government paid the difference. Many banks ended up losing almost all capital and substantial majority of unprotected deposits. Surprisingly, only one bank ended up needing the government compensation, so the direct cost of bank cleanup for the government was just 0.4 billion yen (less than 0.1% of GNP). (The situation was different for agricultural banks that had much higher proportion of small and hence protected deposits.)

Finally, the banks were forced to be recapitalized by issuing new shares to the public. Many banks finished restructuring the old accounts, merged the cleaned-up old accounts with their new accounts, and issued new shares in late 1948. Contrary to the initial concerns, most issues were well subscribed, and many banks succeeded in achieving the goal of having the capital ratio above 5%.

The Japanese episode of handing the system-wide insolvency problem gives several useful lessons. An obvious one is that it is possible to deal with systemic insolvency even under the environment as chaotic as the immediate post-war Japan. The insolvency problem was solved without costing taxpayers very much. It is also remarkable that the recapitalization of the banking system happened without serious problems. This suggests that there was sufficient amount of capital that was willing to come in, once the balance sheets were cleaned up, even right after substantial amount of capital had been written off.

The large deposits, which also suffered from the restructuring, also came back relatively quickly. A paper by Herbert Baer and Daniela Klingebiel (“Systemic Risk When Depositors Bear Losses: Five Case Studies,” in George G. Kaufman (ed) Banking, Financial Markets, and Systemic Risk, JAI Press, 1995) examined what happened the level of deposits after the restructuring in Japan. They found that the amount of deposits declined immediately after the restructuring of the old account, but started to recover within a couple of years. This is what they found for the four other cases from other countries (including the U.S. after the Great Depression) where depositors incurred substantial losses in banking crises. Thus, depositors are also willing to come back to the banking sector once they believe that the balance sheets have been cleaned up.

A key question for the U.S. is whether the current institution (bankruptcy code and other restructuring mechanism) is enough to handle the potential large scale insolvency problem in the banking sector. The case of CFC seems to become an important test case.

2 Responses to "Systemic Banking Problem: A Japanese Experience"

  1. Steve Waldman   May 30, 2008 at 7:31 pm

    Very interesting, nice piece.Two comments — 1) If creditors do pull the trigger on CFC triggering a formal bankruptcy, it almost certainly won’t be accurate to characterize the unwinding as "without a subsidy from Washington". One of CFC’s creditor’s, and I’ll bet its largest, is the Federal Home Loan Banking system. If I recall, FHLB’s exposure was ~50B, but BAC agreed to redeem this debt quickly when the deal closed. (I bet this agreement gets revised, of the deal is going to close at all.) Anyway, if CFC goes under in a formal bankruptcy, US taxpayers are likely to get stuck with a big chunk of the loss.2) One aspect of the Japanese experience you describe leads me to question its comparability to the US situation. Japan’s participation in WWII can be viewed as a political choice, exogenous to the financial system, as a devastating random shock. Correspondingly, there were few "moral hazard" issues with a rearrangement scheme that worked to hold back collapses and retain existing human and institutional arrangements as well as possible as the banking system struggled to revival. The US financial collapse cannot be viewed as exogenous, it was quite plainly due in large part to choices banks made, and made poorly. There might be costs, in terms of future systemic risk, to minimizing disruption in the present US crisis that would not have been relevant to the postwar Japanese case.

  2. Takeo Hoshi   May 30, 2008 at 10:22 pm

    Steve,Thank you for the comments. Both are valid points. Taxpayers may be actually better off if BAC absorbs everything. On the "moral hazard" issue, the initial plan of the Japanese government was to honor the wartime contracts to the corporations by taxing the assets of taxpayers. (To the extent that wealthy taxpayers were also shareholders, this would have similar effects, but …) The U.S. occupational force did not allow this, because they did not want any Japanese business to have profited from the war. So this solution was partly motivated by a "moral hazard" concern in a sense.