بحران بانکی ژاپن و رشد اقتصادی : مفاهیم نظری و تجربی تضمین سپرده و مقررات مالی ضعیف

An endogenous growth model with financial intermediation is used to show how government policies towards the financial sector can lead to banking crises and persistent growth slumps. The model shows how government deposit guarantees and regulatory forbearance can lead to permanent declines in the growth rate of the economy. The effects of inadequate prudential supervision on asset price dynamics under perfect foresight are also derived in the model. The policies that are used in the analysis are based on essential features of Japanese financial regulation. The implications of the model are compared to the experience of the Japanese economy and financial system during the 1990s. We find that the dynamics predicted by our model are generally consistent with the recent behavior of economic aggregates, asset prices and the banking system for Japan. A policy implication of the model is that the impact on future economic growth depends upon the length of time the government fails to enforce loan-loss reserving by banks.

The persistent weakness of the Japanese financial sector and its contribution to the poor growth performance of the Japanese economy are among the most important current challenges to macroeconomics. The high levels of non-performing assets held by Japanese banks have led to a credit contraction and contributed to declines in economic growth. The current difficulties faced by the government in resolving the banking crisis echo a recent history of weak prudential regulation and supervision that are a primary source of the banking crisis in the first place. This paper considers the possibility that public policy towards the financial system is a root source of the Japanese growth crisis. The model presented here shows how a banking crisis can evolve endogenously and lead to a long-run drop in the growth rate. It implies that the longer the government allows banks to carry non-performing assets on their balance sheets, the further the growth rate falls. We develop a simple endogenous growth model with financial intermediation and solve for its equilibrium dynamics. The model incorporates three simple stylized facts about the Japanese financial system. Investment is financed primarily by bank loans and equity issues. The government guarantees the bank deposits of domestic savers. The government allows banks to raise its deposit insurance liability over time through weak prudential supervision and regulatory forbearance. This is represented in the model by the accumulation of non-performing loans against deposit liabilities when the government fails to monitor additions to bank loan-loss reserves. Production in the theoretical model displays constant returns to a single accumulable factor, capital, and is subject to idiosyncratic productivity shocks. Under standard informational assumptions, bank finance dominates direct lending by savers to firms, and bank loans are constrained optimal contracts. Productivity shocks are independent across a continuum of firms at each date and over time. There is no aggregate uncertainty, and both banks and households (as holders of corporate equity) hold perfectly diversified portfolios. This framework focuses the analysis on the role of government deposit guarantees (more generally, government bailouts) and inadequate enforcement of loan-loss reserves in a developing banking crisis and persistent growth decline. If the government continuously requires loan-loss reserving by banks, then the growth rate is constant over time in the model. With regulatory forbearance, the output growth rate declines gradually as non-performing assets held by the banks rise. The long-run growth rate falls towards zero with the length of time that the government forbears. If the government fails to intervene before a critical date, the banks will become unable to meet the deposit withdrawal demands of households for consumption and a banking crisis occurs. At moment of a crisis, the growth rate of consumption suddenly falls. The model allows us to solve for the equity value of banks and of output-producing corporations and how these change before and after the banking crisis. It also predicts a collapse in the market for collateral at the moment of the crisis. The dynamics for output, consumption and investment predicted by the model under regulatory forbearance and deposit guarantees are compared to the recent experience of Japan. We also compare the implications of the model for the stock market value of corporations and for banks, non-performing loan accumulations by the banking system and the value of collateral assets. The strict interpretation of the model in terms of the data requires us to date the onset of the banking crisis. However, a maintained hypothesis of the model is that savers and investors have perfect foresight. The possibility that households receive information about the state of the banking system suddenly with government announcements in the early 1990s is discussed in our interpretation of the data. We also test the relationships in our model using structural vector autoregressions following, for example, Tornell and Westermann (2002). We argue that the data are consistent with the simple dynamic model but recognize that some of the relationships are consistent with other explanations. A special feature of our model is that it implies that the ratio of the value of bank equities to the overall stock market declines before the crisis and ahead of other macroeconomic aggregates. We incorporate this equilibrium relationship in a structural vector autoregression as an orthogonality condition. From the calculated impulse responses, we find that the data supports the predictions of our theoretical model quite well. Our model of a banking crisis complements the traditional view of Japan's banking crisis in the 1990s. The traditional view puts the blame for Japan's banking crisis on the pattern of deregulation that started in the mid-1970s.1 Briefly, as alternative financial instruments were introduced, higher quality customers (that is, large, internationally-oriented firms) moved to raise funds directly in the domestic and foreign bond markets. When large banks started to lose their large clients in the late 1980s, they began lending to small- and medium-sized firms. As the asset-price boom in Japan in the late 1980s collapsed, the Japanese economy slumped into prolonged stagnation. The small- and medium-sized firms had difficulty paying back their loans and the non-performing assets of the banking system rose rapidly. Japan, however, continues to be a model of a bank-centered financial system despite a modest rise in direct corporate borrowing. Cargill et al. (1997) and Cargill (2000) emphasize, as we do, the importance of explicit and implicit deposit guarantees and regulatory forbearance in the crisis, although they do not propose a formal model of credit market dynamics.2 Our model is one of the few general equilibrium models that analyze the Japanese financial and growth crisis of the 1990s. The Japanese banking crisis has affected investment and growth, while the weak growth of the Japanese economy has aggravated the banking crisis. Because of the endogenous dynamics between banking crises and growth, partial equilibrium models are severely limited. Two other papers analyze the Japanese economy in the 1990s using a general equilibrium framework. Hayashi and Prescott (2002) build and calibrate a real business cycle model for Japan and show that the fall in hours and total factor productivity growth can account for the decline in growth in the 1990s. Hayashi and Prescott argue that credit constraints were not important in lowering investment but suggest that weakness of the banking sector may have played a role in lowering investment productivity, although they offer no empirical evidence. As in our paper, Barseghyan (2002) emphasizes delays in government bailouts of the banking system. In his overlapping generations model, the government postpones the actual bailout but insists that banks honor their obligations to depositors. Banks use new deposits to pay old depositors, as in a Ponzi scheme, leading to an immediate contraction in investment. Barseghyan calibrates his model to Japanese data and finds a favorable comparison to Japanese growth in the 1990s. The next section discusses the stylized facts about the Japanese financial system that motivate the simple assumptions of our analysis. Section 3 presents the endogenous growth model and its equilibrium. The following two sections derives the pattern of equilibrium growth and evolution of banking crises under regulatory forbearance. The comparisons with data for Japan and econometric analysis are discussed in 6 and 7, respectively. The last section concludes.

The model is stripped down to reveal the relationship between deposit guarantees, prudential supervision and economic growth that comes about when transfers to bank shareholders or depositors are financed by the public sector through future taxes. In the model, banks can effectively transfer future resources from taxpayers to shareholders through inadequately monitored government-backed deposit guarantees. A banking crisis can be interpreted as either the sudden realization of deposit insurance liabilities by authorities or a simple liquidity crisis in which banks cannot meet the withdrawal demands of depositors due to irreversible investment.19 The qualitative dynamics for the growth rates of gross domestic product, investment, consumption and stock market values for the model compare very well to the Japanese experience of the 1990s. Our analysis of how deposit insurance liabilities affect economic growth emphasizes a simple policy failure. If the government required banks to hold loan-loss reserves against accumulations of non-performing assets, the banks would not have the ability (hence, incentive) to accumulate deposit insurance claims against the public sector. The model also shows that the longer the government waits to intervene in the banking sector and stop the accumulation of unrealized deposit insurance liabilities, the lower the long-run growth rate that results. It can also be shown easily that letting the deposit insurance cost of the crisis accumulate after the crisis lowers the long-run growth rate in our representative agent model. The model supports the serious implementation of Prompt Corrective Action by the Japanese government at any time. In the representative agent model, shareholders and depositors are identical. Deposit insurance under regulatory failure affects economic growth because deposit insurance liabilities are paid by taxes on the interest earned by depositors. An alternative approach would be to use an overlapping generations model so that the transfer scheme redistributes from future generations to current households, even with lump-sum taxation. This approach is taken by Barseghyan (2002), but with two-period lifetimes, many of the dynamics of our model are lost. In an earlier paper, Dekle and Kletzer (2002a), we explore the consequences of expected government bailouts of depositors, implicit or explicit, with weak prudential supervision in an agency-based model of bank-centered financial intermediation. The first version of this paper (Dekle and Kletzer, 2002b) elaborated that model and applied it to the Japanese context. In that agency model, self-financing of a portion of corporate investment by a firm's shareholders reduces the risk of low productivity performance. With limited liability and deposit insurance, banks have incentives to renegotiate loans when current interest cannot be paid. In the process of doing so, the riskiness of a firm's investment rises as self-financing declines in proportion to the firm's capital. Over time, the growth rate of non-performing assets rises over time. The agency model can be added to the current model to include the extra dynamics of non-performing asset accumulation by banks discussed in our previous papers.