Japan's Financial System: Revisiting the Relationship between Corporations and Financial Institutions

Session Chair: WATANABE Tsutomu (Faculty Fellow, RIETI / Professor, Institute of Economic Research, Hitotsubashi University)

Presentation 1: ONO Arito (Senior Economist, Mizuho Research Institute Ltd.)
"The role of collateral and personal guarantees in relationship lending: Evidence from Japan's small business loan market"

Presentation 2: UESUGI Iichiro (Fellow, RIETI)
"Effectiveness of credit guarantees in the Japanese loan market"

Presentation 3: WATANABE Wako (Associate Professor, Graduate School of Economics and Management, Tohoku University)
"How are loans by their main bank priced? Bank effects, information and non-price terms of contract"

Comments: HIRAI Hirohide (Director, Corporate Finance and Tax Affairs Division, Business Environment Department, Small and Medium Enterprise Agency, METI)

Three presentations were made during the session from the critical perspectives set out below. The presentations examined whether positive significance can be found in various measures, such as collateral and guarantees and public intervention through the credit guarantee system and other programs, to ease restrictions that arise when firms wish to borrow, such as credit rationing and high interest rates.

• There is a tendency to be negative toward financial institutions' requirement for collateral and guarantees, but do they not also function effectively in some respects when SMEs raise funds? If so, what are the effects?
• How great are the negative effects of public intervention, such as moral hazard and adverse selection? Do they outweigh the positive effects?
• What impact on interest rates does the use of collateral and the credit guarantee system have?

Dr. Ono's presentation was on the subject of "The role of collateral and personal guarantees in relationship lending: Evidence from Japan's small business loan market," setting out the following points.

Collateral and personal guarantees play a useful role in mitigating debtor moral hazard.

There is no evidence that main banks neglect the monitoring of SMEs if their loans are secured by collateral and personal guarantees, or of a negative correlation between the use of collateral and personal guarantees on the one hand, and close relationships on the other.

Mr. HIRAI Hirohide put forward the following argument concerning Dr. Ono's presentation.

The government is preparing to make changes to the law relating to collateral and guarantees to permit collateral to be pledged not only in the form of land and other real estate but also movable property such as trade receivables. Numerous problems remain, however, such as the difficulty of use and lack of a market for disposing of these, so we would like to further study measures to enhance the system for pledging collateral, including a review of the legal system.

With respect to personal guarantees, there are moves on the public finance side to ease the conditions relating to third-party guarantees, but because of problems such as moral hazard, easing is not necessarily desirable for guarantees by principals.

The following response was made to the above argument.

I agree with making active use of movables for collateral, but we should avoid excessive expectations for this. In many cases the movables pledged as collateral are irrecoverable if a company goes bankrupt. Given this, inevitably financial institutions give it a very low weighting as collateral, and I believe that in practice it would be difficult to use.

With regard to third-party guarantees, in practice they are now rarely used by financial institutions, and there is no sense of there being anything untoward about the moves to reform the system. Nevertheless, as far as guarantees by principals are concerned, I am skeptical about the moves toward easing, and my feelings have been reinforced by the comment you just made.

The following question was posed from the floor.

What is the institutional reason for the fact that trade receivables and movable property have not been used as collateral in Japan?

The following response was made.

The institutional factors are rumor damage resulting from the use of trade receivables, the existence of special agreements prohibiting the cession of claims, and the lack of development of the legal system for the establishment of trade receivables and movable property as collateral.

Dr. Uesugi's presentation was on the subject of "Effectiveness of credit guarantees in the Japanese loan market," setting out the following points.

The Special Credit Guarantee Program, introduced for the period of 1998-2001, was criticized from the outset as inviting moral hazard and adverse selection on the part of both the borrowers and the financial institutions, and thus possibly having a negative policy impact.

By observing changes in the performances of firms that used the Special Credit Guarantee Program and those that did not, we examined which effect of special credit guarantees was stronger: the positive effect in the form of increased availability of funds and improved efficiency, or the negative effect in the form of the worsening of information asymmetry. As a result, we found that the positive effect was stronger.

We were able to conclude that the Special Credit Guarantee Program increased the efficiency of surviving firms, though a problem that remains is that of identifying the cost needed for achieving this.

It should be added that the Special Credit Guarantee Program changed relationship lending for firms immediately prior to bankruptcy, and it was possible that the granting of credit guarantees weakened the relationships.

Mr. Hirai put forward the following arguments concerning Dr. Uesugi's presentation.

Studies are being made in a number of areas with respect to credit guarantee schemes, including enhancing support for rehabilitation, enhancing the flexibility of new guarantees for holders of indemnity rights, making guarantee schemes more diversified and flexible, and introducing a partial guarantee scheme. Of particular note is that the system has not had the desired policy effect with the conventional 100% guarantees, and so it makes a great deal of sense to introduce a partial guarantee scheme.

With regard to the reform of government financial institutions, discussions are under way with a view to reducing direct lending and making it more efficient.

The following response was made to the above arguments.

With respect to the desirable ways of operating credit guarantee schemes, our analysis showed that in ex ante high-risk firms the credit guarantees did not have a policy effect. Conceivable ways of remedying this could include requiring collateral and raising the guarantee premium.

An important debate in the future will be about whether, when providing firms with financial aid, it is more desirable to give them credit guarantees or direct loans. For example, at the end of my report I showed that relationships were weakened by credit guarantees. If direct loans were provided it may be that relationships would be sustained.

The following questions were posed from the floor.

Isn't there a selection bias, because users and non-users of special credit guarantees have not been able to be separated randomly?

The post-crisis period in the analysis corresponds with the period of improvement in bank balance sheets. Does banks' oigashi (additional lending to keep ailing borrowers afloat) have an impact on changes in company leverage?

The following response was made to these questions.

Taking into consideration the endogeneity of choice of system, two-stage estimates were finally made, and conformable results were obtained.

I believe that oigashi does have an impact, but, that notwithstanding, the efficiency of firms that used credit guarantees showed an ex post facto improvement. That is to say, among the firms that actually used credit guarantees there appear to have been a considerable number of good firms that were unable to borrow from banks, despite having a positive net present value (NPV).

Prof. Watanabe's presentation was on the subject of "How are loans by their main bank priced? Bank effects, information and non-price terms of contract," setting out the following points.

Main banks do not lower interest rates for firms that offer collateral voluntarily, which may be interpreted as meaning that the pricing for claims guaranteed through collateral is not appropriate.

Firms that apply for public credit guarantees to credit guarantee corporations on their own initiative pay higher interest rates than firms that do not, which may be interpreted as meaning that firms that apply for public credit guarantees on their own initiative are risky.

Mr. Hirai put forward the following argument concerning Prof. Watanabe's presentation.

In the policy field, too, emphasis is being placed on how much to reduce information asymmetry. This is reflected in, for example, the building of the CRD (Credit Risk Database), the development of accounting systems for SMEs, and inter-regional CLO (Collaterized Loan Obligation).

The following response was made to the above arguments.

I believe that policies to alleviate information asymmetry, such as the development of the accounting system for SMEs and the enhancement of disclosure, are very important. Alleviating information asymmetry should also broaden the range of choices for firms in their financing.

(This report was compiled by SAKAI Koji)