Empirical Analysis of Small Business Financing

Small and medium enterprises (SMEs) support the Japanese economy in many ways, for instance, in the areas of employment, capital investment and technological innovation. When it comes to raising funds, however, SMEs often encounter various constraints not faced by major corporations. Financing available to SMEs is limited with only a small number having direct access to stock and bond markets. Also, SMEs more often face difficulty in obtaining bank loans than do large corporations, partly because banks do not have as much management information about SMEs. In recent years, Japanese banks have generally toughened their stance on lending to SMEs, creating a credit crunch characterized by kashishiburi ("reluctance to lend") and kashihagashi ("calling in of loans"). At the same time, however, they have eased conditions - reducing or waiving interest, deferring repayment of principal, and so forth - on some large borrowers that fail to make timely repayments because of deteriorating business performance. Against this backdrop, the problem of small business financing has drawn a great deal of public attention.

Naturally, the government is acutely concerned about small business financing and has implemented a series of measures designed to assist SMEs facing liquidity problems. In one such measure, the government introduced a special credit guarantee program in 1998 under which prefectural credit guarantee corporations provide loan guarantees to SMEs borrowing from financial institutions. In addition, the Bank of Japan, in addition to its primary responsibility to implement monetary policy by targeting on overnight call rates, is also purchasing SME-related asset-backed securities (ABS's) in its market operations as a means of encouraging capital flows to SMEs.

In this article, I discuss the state of small business financing in Japan from the standpoint of empirical economics and then examine the efforts of the Small and Medium Enterprise Agency (SMEA) and the Research Institute of Economy, Trade and Industry (RIETI) to improve the situation.

There have been few empirical studies of small business financing in Japan, based on the micro data from individual enterprises. An analysis by Suzuki and Yabushita (2002), which utilized corporate survey data gathered by National Life Finance Corp., and another work by Ogawa (2003), based on micro data of Financial Statement Statistics of Corporation collected by the Ministry of Finance are rare examples of such empirical studies.

Unlike large corporations, a small number of SMEs are listed on a stock exchange and their financial data are not readily available, which is the major reason why the literature on SMEs has been very slim. Furthermore, the absence of large-scale surveys that deal not only with corporate financial statements but also qualitative information on the relationships between companies and financial institutions - i.e., comparable to National Survey of Small Business Finances (NSSBF) in the United States - makes it difficult to grasp the actual state of small business financing in Japan. The NSSBF is conducted jointly by the U.S. Small Business Administration (SBA) and the Federal Reserve Board (FRB) roughly once every five years since 1987. Except for company names, the results of the NSSBF are fully disclosed on the Web and freely available for research use. To date, more than 80 research papers known to the SBA and the FRB have been written based on the data.

The SMEA has been trying to overcome the limitations in data availability regarding small business financing in Japan. One such attempt has been made through surveys done for the White Papers on Small and Medium Enterprises in Japan. The SMEA's Research Office conducted the "Survey of Corporate Procurement" (2001), the "Survey of Financial Environment" (2002), and the "Survey of Corporate Financial Environment" (2003) (hereinafter collectively referred to as "SFEs"). In each of these surveys, a questionnaire is sent to a total of 15,000 companies, mainly SMEs, of which 7,000-9,000 have responded each year. The questionnaires ask about the number of financial institutions the companies deal with, their relationship with their main bank, collateral and other forms of guarantee pledged or provided to secure loan repayment, and interest rates charged on short-term loans. In terms of detail, these surveys cannot compare to the NSSBF, a telephone survey for which a 200-page manual provides detailed instructions. However, the SFE may be more useful in analyzing how companies respond to changes in the external environment. More specifically, whereas the NSSBF is conducted every five years with subject companies reshuffled each time, a substantial portion of the companies covered by the SFE remains unchanged, with some 3,000 companies having responded to each round of the survey conducted over the past three years. Using such time series data collected from these companies, it is possible to determine how individual SMEs' financing activities have changed in response to changes in the external environment. (See note)

Another important effort, begun around the same time, was the Credit Risk Database (CRD), which was developed at the initiative of the SMEA's Finance Division, and which collects financial data on a large number of SMEs. The purpose of this database, which now covers the financial statements of nearly 1.4 million companies, is to provide numerical data so as to enable a shift from conventional small business financing, which relies heavily on collateral, to a system that focuses on the financial health of borrowers. The database is available for a fee to anyone researching the state of small business financing.

Already some results have been obtained through analysis of these newly available data. Among these, the 2002, 2003, and 2004 editions of the White Paper on Small and Medium Enterprises in Japan provide an extensive analysis based on the SFE. They have identified and clarified various fundamental facts concerning small business financing that previously had largely gone unquantified: for example, the extent to which interest rates applied to a corporate borrower differ depending on the size and capital adequacy ratio of the borrower, what kinds of companies are providing collateral or credit guarantees, and so forth. In addition, responding to the growing interest in the credit crunch problem and the changing relationships between banks and their corporate customers, the SME white papers analyze the lending attitudes of main banks vis-a-vis various corporate customers; for example, whether the bank rejected a request for an additional loan, whether the bank raised or tried to raise the interest rate, whether the bank demanded additional collateral, and so forth.

However, most of the analyses conducted for the white papers focus on the cross-section dimension of the data collected in each survey year and no attempt has been made to utilize them as panel data. At the same time, economists in Japan and many other countries are trying to determine the state of small business financing and bank behavior based on an enormous number of hypotheses. For them, data collected through the SFE are extremely useful to test these proposed hypotheses. Indeed, based on these data, several studies, apart from those for the white papers, have been conducted. I introduce two such works one by Hosono, Sawada, and Watanabe (2004) and another by Uesugi (2004) below.

Hosono, Sawada, and Watanabe (2004) examine the characteristics of companies that manage to obtain loans from other banks and remain viable even when their main bank rejects their request for a loan.

When a financial crisis occurs, it raises concerns over significant adverse impacts to companies. The grave concerns about the negative impact on local companies in Hokkaido caused by the collapse of Hokkaido Takushoku Bank in 1997 are still fresh in our memories. The data collected in the SFEs, however, show that some 20% of the companies whose loan requests had been rejected by their main bank between 1999 and 2001 were able to obtain loans from other lenders. The study shows what it takes for a company that has been jilted by its main bank to be picked up by another.

First, the authors emphasize the importance of companies doing business with more than one bank during normal times and keeping those banks informed of one's financial and management conditions. This, the authors say, serves as a kind of "insurance," which companies secure loans elsewhere if their main bank refuses to lend. It has been pointed out, somewhat critically, that in Japan even small businesses deal with many different banks whereas most of their American counterparts deal with only one or two. But given the above-described point made by the three researchers, it seems quite rational for a company to deal with many different banks.

Second, they point out that the worse the financial condition of the main bank that has refused to lend, the greater the chance for the abandoned SME to be picked up by another bank. Suppose there are two companies whose financial conditions are equal and that Company A's loan request has been turned down by an undercapitalized main bank and Company B's by a sound bank. According to Hosono et al. (2004), Company A has a greater chance to obtain a loan from another financial institution because there is a greater chance that refusal of the loan request stems from the poor financial condition of the lender rather than the borrower. This indicates the financial condition of the initial lending bank serves as a signal to other banks in making credit decisions on potential borrower companies.

Uesugi (2004) expands the scope of the investigation to shed light not only on bank loans but also on trade credit widely provided through corporate commodity transactions.. Uesugi investigates the relationships between trade credit and bank loans based on the SFE data. Previous studies assume that providers of trade credit have an advantage over banks, both in making credit decisions and taking appropriate actions in a timely manner. Through day-to-day commercial transactions, creditor non-financial companies are able to get a fairly accurate picture of the financial and business conditions of each borrower company. This enables creditor companies to judge the borrowers' credit risk more accurately. Also, they are able to make a relatively accurate judgment on the inventories held by borrower companies and, if necessary, dispose of them in a timely manner. The chart below shows how creditor companies' attitudes in trade credit change in response to changes in the credit risk of borrowers.

The horizontal axis represents the change in scores for each borrower company as determined by private-sector rating agencies. Samples are divided into four quartiles based on the degree of change in credit scores from 2001 and 2002. Toward the left end of the axis, scores are lower and credit risks higher. When the ratios of trade credit and bank loans respectively against total assets are plotted on the horizontal axis, the resulting chart shows that the ratio of trade credit goes down but the ratio of bank loans goes up as the credit risk increases. The reverse movements occur when the credit risk improves. Though differences in the nature of the two types of credit (i.e., trade credit and bank loans) need to be considered, these results basically support the above-described assumption that providers of trade credit, as compared to banks, are able to make more accurate judgments on the credit risk of borrowers, and are thus quicker to distance themselves from risky firms.

As discussed above, some analyses have been already conduced based on the results of the SFEs. Concerning the real state of small business financing, however, much has yet to be empirically proven. Thus, in fiscal 2004, the Research Institute of Economy, Trade and Industry formed a study group on corporate financing to conduct empirical analyses of small business financing. The study group with some 15 members from academia as well as bank and government institutions is chaired by Hitotsubashi University professor Tsutomu Watanabe. It plans to utilize not only the results of the SFEs, but also resources available on the CRD. In the first year of the project, the members will work on their respective research themes, which include the forbearance bank lending to virtually insolvent companies; the relationship between corporate financing and real business activities including capital investment and bankruptcies; the possibility of developing a medium-risk, medium-return market; and the roles of credit guarantees and public finance. I hope that these research efforts will help draw an accurate picture of the actual state of small business financing and thereby contribute to the development of effective government policies.

Meanwhile, the SMEA is planning to conduct further surveys to capture the current status of SMEs that also cover financing. The current SFE is targeted at relatively large SMEs and it remains difficult to asssess the situation of smaller firms such as sole proprietorships and very small enterprises. The Small and Medium Enterprise Basic Law, however, mandates the government to regularly conduct surveys so as to clarify the overall state of SMEs. The SME Research Office and Research and the Statistics Department of the Ministry of Economy, Trade and Industry have jointly designed a new survey called "Basic Research on the State of SMEs." The first round of the survey will be carried out within this fiscal year and cover some 100,000 corporations and proprietorships. It is expected that this new survey, which is designed to determine financing environment of SMEs, will complement the existing financing environment survey.

• Uesugi, Iichiro. 2004. Nihon ni okeru Kigyo-kan Shinyo: Kinyukikan Kariire tono Kankei [Trade credit in Japan: Relationship with bank loans]. RIETI Discussion Paper Series 04-J-001.
• Ogawa, Kazuo. 2004. Kashishiburi wa Sonzai Shitanoka: Kigyo no Setubitoshi Kodo to Ginko Shinyo [Was there a credit crunch?: Corporate capital investment and bank credit]. Chapter IV of Daifukyo no Keizai Bunseki [Economic analysis of Japan's Depression in 1990s], Nihon Keizai Shimbun, Inc.
• Suzuki, Kumi and Shiro Yabushita. 2002. Chusho Kigyo eno Kashituke Kinri ni kansuru Paneru Deta Bunseki [Panel data analysis on interest rates to SMEs]. Presented at the 2002 spring conference of the Japan Society of Monetary Economics.
• Hosono, Kaoru, Mitsuru Sawada, and Tsutomu Watanabe. 2004. Suteru Kami Areba Hirou Kami Ari: Kinyu Kiki-ka ni okeru Chusho Kigyo no Shikin Chotatsu. [When one god deserts you, another will pick you up: Small business financing under a financial crisis]. Unpublished manuscript.

>> Original text in Japanese