SME Financing in Japan: What We Have Found

Throughout the periods of the bubble economy and the subsequent "lost decade," the main players of the Japanese economy have been major banks and large-sized companies. These institutions were the driving force behind the creation of the bubble. They were also the biggest sufferers of the bubble burst; forced to take massive losses, their anguish continuing for many years. Needless to say, however, the Japanese economy does not consist only of major banks and companies. On the contrary, some 70% of Japanese workers are employed by small and medium-size enterprises (SMEs) and half the total value added in Japan is generated by SMEs. In this sense, the coexistence of major companies and SMEs is how the Japanese economy is supposed to be. Based on this perspective, the past 20 years can be redefined as an unusual period in which the Japanese economy deviated from the norm in that the presence of major companies and banks were overstated. A return to monozukuri, the craftsmanship in developing, creating and manufacturing products, is called for in pursuing an ideal form of the Japanese economy as the nation re-emerges from the lost decade, while banks are advocating the importance of SME lending as a future revenue source. Both of these phenomena stand as evidence of the fact that the Japanese economy is now beginning to return to its normal state where major companies and SMEs coexist.

How SMEs manage to obtain financing smoothly is of great interest both within and outside Japan. Success or failure in obtaining finance exerts a far greater and more direct impact on the fate -- i.e. survival or exit -- of SMEs, which are primarily dependant on loans from financial institutions, than on the fate of major companies, which have access to diverse forms of financing including the issuance of shares and/or bonds¹. Japanese financial institutions are trying to respond to the needs of SMEs by diversifying financing options. Lending based on a scoring model, securitization and asset-backed loans are some examples of new financing options for SMEs. However, these efforts are far from enough to alleviate many of the problems incurred by information asymmetry in SME financing. Facilitation of SME financing is both an old and new problem. Meanwhile, in developing countries where major companies are nearly non-existent, the financing problem of domestic companies is synonymous to the problem of SME financing. Thus, these countries are keen to learn SME financing mechanisms used in developed countries from the viewpoint of what sort of financial system should be established in order to achieve economic development².

Based on the recognition discussed above, RIETI's Corporate Finance Study Group (Leader: Tsutomu Watanabe, Professor at Hitotsubashi University / Manager: Iichiro Uesugi, Fellow, RIETI) has been trying to analyze the behavior of SMEs particularly in financing activities. More specifically, by using a recently developed set of microdata on SMEs, we are examining statistical regularities observed in various aspects of SME financing, thereby attempting to clarify the financing environment faced by SMEs. Members of the study group have hitherto published 10 discussion papers through RIETI. On February 16 and 17, 2006, our study group and other study groups of RIETI will jointly organize an international policy symposium, where we will hold extensive discussion on the results of our research, and be joined by external researchers and experts including an economist at the United States Federal Reserve Board. In the following, I will discuss the conditions under which the SME financing environment, which is said to be very severe, changes and the implications of such change, constructing arguments based on three of the discussion papers of which I am a coauthor.

SMEs, though referred to in bundle, are in fact extremely diverse in nature, ranging from high-tech startups to wagashi (traditional Japanese confectionery) shops with generations of history. Despite such diversity, however, certain regularity can be found in the pattern of financing for SMEs. Practitioners note that the longer the history of SMEs, the lower their cost of borrowing. Thus, we capture the evolution of SME financing, using borrowing rates as a variable and focusing on the age of SMEs³.

Earlier studies provided a theoretical model which assumes that SMEs accumulate reputation as they continue to stay in business, thus their borrowing cost falls as they age. On the other hand, some people point to the effect of natural selection, that is, with the passage of time poorly performing borrowers would be selected by banks, subjected to higher interest rates, and eventually forced to exit. These two views differ in focus; the first looks only at surviving companies whereas the second pays special attention to the difference between surviving companies and defaulters. We have integrated these two different viewpoints. Specifically, we have defined the relation between corporate age and borrowing cost through two channels, i.e. "selection," a process in which poor performers are separated from good ones and then forced out of the market, and "adaptation," a process in which surviving companies change their behaviors as they age. This allows us to examine how each of these two channels functions and which has a greater impact on the financing environment of a company⁴.

The results of our analysis show that older firms are benefited from lower interest rates. But this is not the only finding. First, observation of the selection channel has found that companies with low return on assets (ROA) and equity ratios are paying high interest rates and eventually defaulting. This indicates that the mechanism of natural selection works properly for SMEs. Previous studies on banks' lending to major companies have found that so-called forbearance lending to low-quality companies became conspicuous from the latter half of the 1990s. That is, natural selection, a mechanism failing to function for major companies, is properly functioning for SMEs. Next, our study finds that the effect of adaptation is a greater contributing factor than the effect of selection in improving the financing environment. Of the decrease in interest rates resulting from the aging of borrowing companies, one third is attributable to the selection effect and two thirds to the adaptation effect. There is a variety of ways in which adaptation mechanism works, which are exemplified by the close relationships between firms and banks or original innovation systems. Through adaptation these function better in reducing the interest rates than does the selection mechanism.

Since around the bubble period, the presence of excessively indebted and poorly performing companies -- particularly, in the so-called structurally depressed industries such as construction, real estate, and wholesale and retail sectors -- has been pointed out. Of course, inefficient companies should exit from the market and this sort of Schumpeterian creative destruction is surely effective in improving the financing environment for SMEs. However, what is suggested by the results of our latest study is that a gradualist approach that seeks reform in the form of adaptation by making use of existing companies is more important in lowering the cost of borrowing.

Firm age is not the only factor affecting the financing environment. In reality, the provision of collateral and guarantors would make it easier to obtain loans. Collateral and personal guarantees are important particularly for SMEs, whose credit risks are difficult to assess, and extensively used in the U.S. and many other countries.

Despite this, the role of collateral and personal guarantees tends to be viewed in a negative light. In 2003, the Financial Services Agency drew up an "Action Program Concerning Enhancement of Relationship Banking Functions." A report underpinning the program cites the concern that Japanese financial institutions might have been excessively relying on the availability of collateral and personal guarantees in making loan decisions while failing to properly utilize invisible information obtained through relationships with respective borrowing companies. There, it is assumed that lending based on collateral and personal guarantees is inefficient and thus should be replaced by relational banking once an appropriate lender-borrower relationship is established. But is the presence of collateral and personal guarantees truly of no use in obtaining loans in Japan? To find an answer to this question, we seek to verify a series of theoretical hypotheses concerning the role of collateral and personal guarantees, using microdata from surveys conducted from 2001 on SME financing⁵.

The implication of the results is that collateral and personal guarantees play positive roles in obtaining loans. Specifically, our findings include: 1) high-risk companies that are prone to moral hazard provide more collateral and personal guarantees; and 2) companies providing collateral and personal guarantees are subject to more frequent monitoring by lending banks and more likely to build long-term and multi-faceted relationships with lending banks as compared to companies of equivalent credit risks but without collateral and personal guarantees. In today's Japan, the provision of collateral and personal guarantees is complementary, not substitute, to monitoring by banks and relationship lending.

Because data from earlier years are not available, we cannot quantitatively compare these findings with the general perception that "Japanese banks opted to forgo monitoring when borrowers offered real estate as collateral" during the bubble period. But as far as we can see based on our discussions with practitioners, the treatment of collateral in banks' loan decisions today is quite different from during the bubble period, due partly to sharp falls in asset prices, and it is unlikely banks become lax in monitoring efforts simply because they have taken collateral. Judging from all the above, it seems that companies with pledgeable assets are receiving various benefits, more than just easier financing, by providing collateral and personal guarantees.

Finally, let me discuss the influence of government interventions, particularly, the public credit guarantee system, in the credit market for SMEs⁶. Japan's credit guarantee systems have a very long history with the total outstanding amount guaranteed as of today at about 30 trillion yen. The system grabbed public attention with the introduction of the government-sponsored "special credit guarantee program," a scheme designed to alleviate a credit crunch faced by SMEs, from October 1998-March 2001. At the time, the Japanese economy was on the verge of slipping into a deflationary spiral and SMEs were seriously affected by the then pervasive credit crunch. Under the scheme, the government provided repayment guarantees worth 30 trillion yen, which would cover roughly 10% of SME loans then outstanding nationwide. The safety measure, however, provoked sizable criticism with its legitimacy questioned by those pointing to the possibility that moral hazard and adverse selection might occur in the credit market.

In our study, we track the records of 3,000 companies, including companies that turned to the special program and those that did not, using data in their financial statements for eight years from 1996-2003. Then, we measure the effect of the special credit guarantee program by comparing the financial data of these companies during the period before and after the introduction of the program. We find that the program has helped improve not only the financing environment but also business performance of companies. More specifically: 1) leverage increases more sharply at companies using the program than those not; 2) the ratio of tangible assets to total assets rises, albeit slightly, at companies using the program; 3) profit margins at companies using the program, when look at in average, increase more than those not; and 4) the degree of effect of using the program on profitability, however, greatly differs, i.e. the positive effect of using the program is not observable at companies with a low equity ratio, an indicator of high credit risk, whereas comparison within the low-risk samples -- companies with a high equity ratio -- show a conspicuous rise in profits margins at companies using the program.

These results indicate that the government's special credit guarantee program did alleviate the financial constraints of SMEs then seriously affected by the credit crunch, thereby inducing their investments. The implementation of investment projects that would have been impossible without the easing of financial constraints resulted in the improvement of profitability in terms of the average for all samples. However, a closer look shows that the easing of financial constraints did not necessarily lead to efficient investments at high-risk companies as compared to low-risk companies.

In this article, the current state of SME financing in Japan is analyzed from various angles based on three papers in which I was involved. Each of the examined factors -- reputation as embodied by corporate age, the availability of collateral and/or personal guarantee, and the presence of the public credit guarantee system -- is generally conducive to improving the efficiency of credit allocation.

There are several policy implications. First, the finding that the financing environment of SMEs changes more greatly through innovations by SMEs than through the exit of failing companies points to the importance of smooth business succession in ensuring the survival of existing SMEs. This is quite important given the ongoing reality that the number of individual proprietorships is rapidly decreasing due to the aging of business owners.

Next, it has been found that the provision of collateral and/or personal guarantee, which is complementary to monitoring and relationship banking, is not properly reflected in the interest rates and other terms and conditions of lending. Thus, it is hoped that the lowering of credit risks by the presence of collateral and/or personal guarantee will be properly processed in banks' credit screening, which is also important in creating an environment that facilitates the provision of collateral and/or personal guarantees where possible. At the same time, it is also urgently needed to reinforce the lending mechanism for companies having no collateralizable assets.

Finally, the contrasting results between low-risk and high-risk companies concerning the use of the public credit guarantee system -- i.e. performance improved at low-risk companies whereas profitability remained low at high-risk companies -- provide important suggestions in rethinking the scope of the target for this particular policy measure. Assistance to high-risk companies may be justified from the viewpoint of preventing social anxiety that can be caused by a rapid increase in corporate bankruptcies. However, for the purpose of improving the economic efficiency of SMEs, the target of the public credit guarantee system should not be limited to high-risk low-profitability companies. Another point that should be noted is the need to examine whether this scheme is fiscally sustainable, that is, whether the benefit of the scheme that is embodied in the form of SMEs' improved efficiency as observed in our research pays off the cost which would be incurred in the form of the government's payment of defaulted loans as a guarantor. By doing so, policy evaluation of the effect of the public credit guarantee system becomes possible.

It is hoped that the research results published by the Corporate Finance Study Group, including the aforementioned three papers, will be effectively used by those engaged in economic policy planning.

>> Original text in Japanese

1. With respect to major companies, it is rare that financial constraints per se become a problem. Rather, their financing activities, such as obtaining equity investments, are often raised as an issue within the context of governance.
2. Reflecting the growing interest both in developed and developing countries, an OECD Global Conference on Better Financing for Entrepreneurship and SME Growth will be held in March 2006.
3. See "Firm Age and the Evolution of Borrowing Costs: Evidence from Japanese Small Firms," RIETI Discussion Paper 05-E-026 (coauthored with Koji Sakai and Tsutomu Watanabe).
4. This analysis has been made possible by tracking the data of financial statements of more than 200,000 SMEs, for six years from 1997-2002, provided by the Credit Risk Database.
5. See "The Role of Collateral and Personal Guarantees in Relationship Lending: Evidence from Japan's Small Business Loan Market," RIETI Discussion Paper 05-E-027 (coauthored with Arito Ono).
6. See "Effectiveness of Credit Guarantees in the Japanese Loan Market," RIETI Discussion Paper Series 06-E-004 (coauthored with Koji Sakai and Guy M. Yamashiro).