Summary of the Workshop on Non-Performing Loans

Anil Kashyap

Professor of Economics
The University of Chicago
Graduate School of Business


  There was a strong consensus among the participants about almost all of the key aspects of the problems.  This summary reflects that consensus, although given the size of the group it should be understood that individual's views may differ on specific points.

  The participants argued that the non-performing loan problem is a reflection of the deterioration in asset quality of many borrowers.  Much of this deterioration is due to weak macroeconomic conditions present in Japan.  The group asserts that until the deflation is arrested, an improvement in creditworthiness is very unlikely.

  More importantly, the current sharp economic slowdown has meant that new problem loans are appearing faster than the financial institutions can provision for them.  This situation is very dangerous and calls for a strong coordinated policy response.  The Bank of Japan must play a role in this effort.  The group rejects the notion that the BoJ cannot stop deflation and encourages the Bank to take all necessary steps to do so immediately.
Several specific proposals were put forward as to how the Bank might proceed.

  One strategy, first proposed by Lars Svensson, focuses on establishing a temporary peg for the Yen at a rate far below its current level.  This peg would be maintained until prices began to rise.  When prices had returned to an appropriate level the peg would be broken.

  Another option was to have the Bank of Japan buy the debt issued by the Resolution and Collection Corporation (RCC).  If the RCC used this money to buy bad loans from financial institutions this would amount to monetizing the bad loans.  If these bad loans were purchased at a price above the current secondary market prices this could also provide some new capital for the banks.  (This procedure would also allow the recapitalization to be kept off-budget, something that has precedent in a number of other cases including the U.S. Savings and Loan bailout.)

  The group did not settle the issue of which particular plan should be pursued, but there was unanimity that prompt action was needed.  The group strongly rejected the idea that the Bank should wait to begin actions until other aspects of the full policy response can be finalized.

  Nonetheless, the participants also indicated that actions by the Bank of Japan in isolation will not be sufficient to solve the problem.  A successful plan will require changing the incentives for the regulators, banks, and corporations to identify and provision for non-performing loans.  Similarly, the final resolution of the problems should also recognize that there are significant losses in the insurance sector and in many government- sponsored agencies.  The sources of these problems and likely solutions to the problems may differ from the sources of and solutions for the commercial banks troubles.  But it is imperative that these problems are not ignored.  In short, resolving the NPL problem will require a comprehensive attack that involves coordination among many parts of the government.

  The group did not fully explore all of the options that merit consideration for addressing these issues.  However, there were several points of agreement, including:

  1.   A very rapid response is needed.  The situation is continuing to deteriorate. Delay raises the cost to the taxpayer and makes most choices more difficult.  The experience in the US in the early 1990s (and to some extent already in Japan) suggests that releasing bad news and attacking problems often is greeted favorably by markets.


  2.   The banking sector is going to require a further capital injection and this injection should be done recognizing that Japan is over-banked.


  3.   It is essential that incentives are put in place to prevent further capital from being squandered.  This might be done by using a good bank/bad bank approach.  Alternatively, a carefully designed temporary nationalization scheme might be even appropriate.


  4.   The regulatory agencies do not have enough personnel to undertake the intensive monitoring that will be needed.  The regulators need to be held accountable for not detecting problems sooner and should be discouraged from protecting bad borrowers or poor managers.


  5.   The total cost (including personnel, depositor payoffs and capital injections) for the government is going to be huge.  A full resolution of the problems will require the government to spend far more than can be done within the 30 trillion Yen debt issuance target for the next two years.  (Although one could argue that the bad loan clean up costs should properly be separately funded and should not count against this debt target.)