Lessons from the East Asian Crisis
Research by Raghuram Rajan and Luigi Zingales
Just a few years ago, it was fashionable to decry the shortsightedness of the American financial system and the tendency of U.S. financial markets to ignore long-term corporate prospects while focusing on quarterly earnings reports. There were repeated calls for the U.S. to adopt new laws that would permit financiers to take a longer view of their investments, and to move toward the more relationship-based investing model that prevails in Japan.
But it is amazing what a banking crisis or two will do to popular fashion. Now the trend is to talk about the virtues of "the market," the importance of competition and disclosure, and the horrors of crony capitalism.
In recent research, finance professors Raghuram G. Rajan and Luigi Zingales of the University of Chicago Graduate School of Business analyze the questions surrounding the recent Asian economic collapse. Specifically, they seek to answer the following questions: Why did the relationship-based financial systems, which had been credited with fueling the miraculous growth of East Asia, suddenly implode? Was the recent crisis a temporary setback to an otherwise successful system or does it herald its demise? Does the slow but steady ascendance of the public markets, even in Germany, suggest the eventual supremacy of the arm's-length, market-based, Anglo-American system?
Relationship-Based Versus Arm's-Length Systems
To answer these questions, the authors provide a general sketch of the salient features of these two systems.
A financial system, they say, has two primary goals: 1) to channel resources to their most productive uses, and 2) to ensure that an adequate portion of the return flows to the financier. The second goal is crucial to the first goal, because without the prospect of an adequate return, funds will not be made available for investment.
Relationship-based systems ensure a return to financiers by granting them some form of power over the firm being financed. The simplest form of power is when the financier has ownership of the firm, either directly or indirectly. The financier can also serve as the sole or main lender, supplier, or customer. In all of these forms, the financier attempts to secure his return on investment by retaining some kind of monopoly over the firm. As with every monopoly, this requires some barriers to entry. These barriers may be due to regulation, or to lack of transparency -- the collusive, protective nature -- of the system, which substantially raises the costs of entry to potential competitors.
Contrast this with the arm's-length, Anglo-American system, where the financier is protected by explicit contracts. In such systems, contracts and associated prices determine which transactions occur. As a result, institutional relationships matter less and the market becomes a more important medium for directing and governing the terms of transactions.
An important distinction between these two systems is their different degree of reliance on legal enforcement. Relationship-based systems can survive in environments where laws are poorly drafted and contracts not enforced. The relationship is largely one of self-governing parties intent on maintaining their "reputations" in honoring the spirit of agreements (often in the absence of any written contract) in order to ensure a steady flow of future business within the same network of firms.
By contrast, the prompt and unbiased enforcement of contracts by courts is essential to the viability of a market-based system. Moreover, since contracts are typically hard to write with the wealth of detail necessary to fully govern transactions, it is important that the law offer a helping hand. Under common law, the court tries to follow the spirit rather than the letter of a contract, thus enabling contracts of offer greater protection. For this reason, it is perhaps no surprise that market-based systems are found largely in countries with a common-law tradition, hence the "Anglo-American" model.
Another distinction between the two systems is the relative importance of transparency, or the ability to see clearly into corporate dealings. Market-based systems require transparency as a guarantee of protection. The authors cite the words of Franklin Roosevelt to convey this phenomenon: "Sunlight is said to be the best of disinfectants; electric light the most efficient policemen." By contrast, relationship-based systems are designed to discourage transparency, which has the effect of protecting the relationships from the threat of competition.
An Example: Credit
The authors provide the example of a transaction -- the extension of credit -- in each of the two systems. In a relationship-based system, a bank will have close ties with a potential borrowing firm, perhaps because of frequent past contacts or because of ownership links. In assessing the borrowing needs of the firm and its ability to pay interest and principal, the bank will consider not only the firm's current debt-servicing capability, but also its long-term ability to repay, and the various non-contractual levels the bank can push to extract repayment. The interest rate charged will be negotiated repeatedly over time, and may not have a direct relationship to the intrinsic risk of the project.
In an arm's-length system, by contrast, the firm will be able to tap a wider circle of potential lenders because more widespread financial information is available. The loan will be contracted for a specific period, and the interest rate will be a competitive one that will compensate the lender for time and risk.
Limitations on competition in a relationship system do not just give the financier power, they also strengthen his incentive to co-operate with the borrower. Studies of Japanese "keiretusus" show that the main banks went out of their way to help financially distressed borrowers. For example, Sumitomo Bank not only effectively guaranteed Mazda's debts when it got into trouble after the first oil shock, but also orchestrated a rescue, in part by exhorting employees within its keiretsu to buy Mazda cars. Sumitomo's incentive to help would have been considerably weaker if Mazda, once it emerged from distress, had had the option of giving the lion's share of its business to some other bank.
The absence of competition and disclosure in a relationship-based system imply that there are really no price signals to guide decisions. Unlike an arm's-length system, where a number of competitive lenders can give a borrowing firm independent assessments of the costs of undertaking a project, the cost a borrower faces in the relationship-based system is simply what the relationship lender and the borrower negotiate.
But is this necessarily a bad thing? Are lending and investment decisions always inefficient if the cost of the funds differs from the true costs? Are there no redeeming features of a relationship-based system? The answer is no to all three questions, say the authors. In the real world with all of its imperfections, an imperfect cost of funds can sometimes produce the right investment decisions.
For example, consider the example of the distressed Mazda corporation. In a relationship-based system, the lender -- confident in the strength of its relationship with Mazda -- can offer a below-market rate in the short run to bail out the company and then recoup its losses with an above-market rate over the long run when the firm is healthy and can afford high payouts. A firm in trouble is less likely to be bailed out in a competitive, arm's length system where there are no guarantees of long-term payouts for the lender.
It is this kind of ability to trade off short-run losses for longer-run gains that led many economists to defend the efficiency of relationship-based systems. But it is easy to see the problems that can arise in such systems, say the authors. Perhaps most important, the relationship-based system does not pay much attention to market or price signals. And this indifference to price signals becomes self-fulfilling. If investment decisions are not driven by prices, then prices become less effective in providing economic directions because they reflect less information. The message from existing research is that although relationships may increase or preserve value in some cases -- particularly when contracts are hard to write or enforce -- they also have a downside that they do not rely on price signals. The consequence has been a widespread and costly misallocation of resources.
Making Sense of the Asian Crisis
Until the end of the 1980s, the East Asian economies were overwhelmingly relationship-based systems. At the outset of liberalization, the volume of profitable investment opportunities greatly exceeded the available capital. This capital shortage in turn prompted a momentous change in the environment: namely, the opening up of these economies to capital flows--a development that coincided with the increased desire of Western banks and fund managers for international diversification.
But, there was a problem: A flood of foreign capital poured into these countries at a time when the institutional infrastructure was not adequately developed. Essentially, the arm's-length capital was lent to a relationship-based system that did not have adequate price signals to deploy the massive inflow of capital properly.
Not only did foreign lenders not always know whether their funds were being deployed appropriately, they also did not have the institutional safeguards to protect their investment. Therefore, they took the next best route; they kept their loans and investments short term so that they could pull out at any indication of trouble. So long as the countries could not offer adequate institutional safeguards, short-term financing was the cheapest way for the countries to obtain the large amounts of capital being offered. Both sides were happy provided the economies continued to hum along.
But then prospects changed. It is hard to say whether the trigger was the depreciating yen (driven down because of loose Japanese monetary policy), poor macro-economic policies, or the realization that capital was being poorly invested by the relationship system. At any rate, once some foreign arm's-length capital started to pull out, it did not make sense for any to stay in. Since the relationship-based system ensured that the pain would be spread through invisible cross-subsidies, and not contained within a few specific "bad" institutions, it made sense for every outsider to pull out. This was not necessarily a panic, but a rational move to the exit doors by arm's-length capital providers who knew they had inadequate protection for the long run, and were not sufficiently part of the relationship system to get any of the benefits of staying.
The authors do not rule out the possibility that moral hazard may have been behind some of the investment that flowed in, or that there was some panic in the search for exits. Yet the authors conclude that much of what happened can be explained as the consequence of two financial systems that are essentially incompatible coming into contact with each other. The mistake, if any, may have been on the part of East Asian countries in underestimating the risk involved in accepting such capital flows without a clear plan to reform, and in not reforming their institutions once the flows began.
One clear policy implication of this analysis is that a country faced with the prospect of substantial financial inflows has to either accept the risk of financial fragility or improve its financial infrastructure before it accepts the inflows. Institutions such as exchanges and custodial services have to be set up, monitors such as rating agencies, auditors and supervisory authorities have to be established and strengthened, accounting standards and disclosure laws improved, and bankruptcy and contract law made more effective.
Where Do We Go From Here?
Given the flight of foreign capital and the ensuing capital shortage now confronting them, East Asian economies might appear to be justified in returning to their traditional relationship-based systems. But is a return to the old system likely to restore these economies to their former strength? And is a relationship-based system really a viable, long-run solution for these economies?
The authors' analysis thus far would suggest that a return to relationships is the best way to go in the short run. Yet, there is a fundamental problem with a relationship system -- that is, its resistance to change. The hidden and collusive practices that sustain a relationship-based system entrench the current players at the expense of potential new entrants. Moreover, the very lack of transparency also makes it hard for democratic forces to detect all the abuses in the system, allowing current players to resist reform.
One of the effects of a crisis is to create such immense problems that even the relationship system cannot hide them. The evidence of gross abuse can be a powerful weapon for democratic and liberal forces in pressing for reform. An example is the financial legislation that was rushed through Congress soon after the onset of the Great Depression, and in 1933 and 1934. Much of that legislation -- including the Glass Steagall Act and the Securities Act of 1934 -- has been attacked by economists as politically motivated and a source of inefficiency in the U.S. economy. What is rarely pointed out, however, is that such legislation laid the framework for the modern U.S. financial sector. The crisis of the Great Depression provided an opportunity for democratic forces to combat the concentration of power on Wall Street, and the legislation that resulted from the crisis essentially ended relationship-based finance in the United States. In so doing, the Glass Steagall and the Securities acts can be seen as providing the impetus for the present variety and competitiveness of U.S. financial institutions.
In the near term, then, East Asia's crisis and capital shortage provides a rare opportunity for institutional reform. Over the longer run, however, such economies can be expected to move from the current condition of capital shortage to situations that once again test the system's ability to allocate capital. When that day comes, the arm's-length system is likely to be more efficient. If the recent crisis can be weathered, the long run need not be that long for the East Asian economies. Rather than reconstituting the old monopolies and inside deals, these economies would be well advised to follow the U.S. example in the 1930s and take advantage of the financial crisis to improve transparency and accountability in their financial system.
Raghuram Rajan is the Joseph L. Gidwitz Professor of Finance at the University of Chicago Graduate School of Business. Luigi Zingales is a professor of finance at the University of Chicago Graduate School of Business.