Policy Update 035

The 1990s in Japan and Lessons for the Current Financial Crisis

KOBAYASHI Keiichiro
Senior Fellow, RIETI
Visiting Professor, Chuo University
Research Director, Canon Institute for Global Studies

1. Crisis

What is the similarity between Japan's crisis and the current financial crisis?

Both crises were severe recessions that followed the collapse of asset-price bubbles and both bubbles emerged in real estate markets. The basic features of Japan's crisis are as follows: economic transactions were disrupted due to a malfunction of the payment system; the government responded with massive fiscal stimuli, which worked only temporarily and failed to get the economy back on track; the central bank conducted extraordinary monetary easing, the effect of which was not clear; and the most notable feature was that the nonperforming loans problem in the banking sector in Japan was prolonged, continuing for more than a decade from 1991 to 2005.

2. Bubble

What causes bubbles? Can we prevent the recurrence of bubbles?

It is often argued that financial deregulation and/or securitization of bank loans are the causes of a bubble. However, Japan experienced a bubble without these factors. In the 1980s when a huge bubble emerged in Japan's real-estate market, financial institutions were heavily regulated in Japan. There was no securitization technology in the 1980s and 1990s in Japan and bad assets were bad loans on the bank balance sheet.

These facts imply that imposing new regulations or prohibition of securitization cannot prevent the recurrence of bubbles. Over leverage was the key driver for the asset price bubbles. There are several ways to prevent over leverage: introduction of new monetary policy that responds to the asset prices; capital requirements to secure the bank balance sheets; and Glass-Steagall type prohibition, which means that banks who form the payment system cannot invest in a high-risk assets. To prevent excessive risk taking, eradicating the side effects of limited liability may be essential. Limited liability of bank CEOs made them risk lovers. Therefore, unlimited liability may be a solution. If bank CEOs were personally liable for their banks' debts, they would become very cautious.

3. Fiscal Policy

Why fiscal policy did not work in Japan in the 1990s?

Fiscal policy was a well established "pain reliever" in the Post-War Keynesian regime of economic management. Policymakers in Japan did not recognize that huge bad loans exerted perverse and persistent externality on the Japanese economy. They just conduct the fiscal policy reflectively, on the premise that: (1) the economy will recover spontaneously as per usual recessions, and (2) only one or two rounds of fiscal stimulus are sufficient to buy enough time for economic recovery. Moreover, politically, the easiest response was repetition of pork-barrel fiscal stimuli. But the effect of fiscal stimulus is by nature temporary. The economy remained stagnant throughout the 1990s.

4. Deflation

Why did deflation continue for a decade in Japan? Why didn't monetary policy work?

Popular argument among macroeconomists and monetary economists concerning the deflation in Japan is that deflation prolonged Japan's recession. Prominent economists like Krugman, Bernanke, Auerbach, among others, argued that the Bank of Japan should have conducted a much more aggressive monetary policy. These arguments are based on a variant of the new Keynesian view that deflation causes recession. In this view monetary policy is still effective only if it can change expectations on the future inflation rate. The BOJ was skeptical about the effectiveness of monetary policy because they believe that the systemic problem in the banking sector hinders the effect of monetary policy. But the general public formed the opinion that, "the Recession will not end unless the BOJ stops deflation." Pushed by this public pressure, the BOJ was forced to implement the zero nominal interest rate policy in 1999 and the quantitative easing policy in 2001. What the economic performance during the early 2000s showed, however, is contrary to the widespread theory that the recession will not end as long as deflation continues. Japan experienced a long economic recovery in 2003-2007, while deflation still continued during this period.

5. Bad Assets

Why did the nonperforming loans problem persist for a decade? Why is government intervention necessary to resolve the bad-debt mess?

My conjecture is that, "bad assets hidden in the financial sector impede economic growth." Bad assets destroy inside money, in other words, bad assets disrupt the payment system. Disappearance of money disrupts economic transactions. Output declines and unemployment increases. Deflation is also a consequence of a protracted bad asset problem, because deflation is caused by the disappearance of money.

We may need the following "Bad-Asset Theory of Financial Crisis": hidden bad assets destroy inside money through information asymmetry (Akerlof 1970)(*). Since money is a medium of exchange in economic transactions, it is a public good. Inside money is short-term debts of financial institutions (e.g. deposit money, repo), the value of which is guaranteed by the assets of the financial institutions. In this situation, the emergence of bad assets causes severe information asymmetry about asset quality that leaves market participants unable to distinguish good assets from bad assets. As a result, confidence in the value of assets breaks down completely (the lemon problem). A run on short-term debt occurs just as the bank runs in the 1930s. As a result, money disappears and financial intermediation malfunctions, leading to disruptions of economic transactions. Eventually, output declines and unemployment increases. On the other hand, the shortage of inside money causes deflation. Recessions caused by the bad assets problem tend to persist for many years. This is because the bad assets problem is essentially an external diseconomy, that is, the disappearance of a public good due to information asymmetry. Since the problem is an external effect, banks cannot internalize the social cost of the bad assets and they do not have proper incentive to reveal their own bad assets and will not dispose of them. As a result, hidden bad assets continue to exist for years as does the economic disruption. This argument also justifies the policy intervention. Since the bad assets problem exerts a perverse externality on the economy, private agents (i.e. banks and borrowers) do not have proper incentive to resolve the bad debts. The practical implications are as follows: bad-asset disposals are effective to resolve both recession and deflation; and government intervention is necessary to accelerate the bad-asset dispositions.

Unfortunately, we did not understand all these things in the 1990s in Japan. (We viewed the bad loans problem as purely a private problem between banks and borrowers.) This time around, we should manage the crisis more effectively, based on the lessons from Japan's experience.

*Keiichiro Kobayashi, Senior Fellow, RIETI wrote this column for his visit to The Caisse des Dépôts et Consignations (CDC)/L'École des hautes études en sciences sociales (EHESS) in October 2009.

Reference(s)

*  Akerlof, George A, 1970. "The Market for "Lemons": Quality Uncertainty and the Market Mechanism," The Quarterly Journal of Economics, MIT Press, vol. 84(3), pages 488-500

October 26, 2009

Article(s) by this author