«by MICHAEL GUDGER FAO AGRICULTURAL SERVICES BULLETIN 129 The designations employed and the presentation of material in this publication do not imply ...»
• Greater skill in managing risk, not merely transferring it, is accumulated through experience in making more risky loans, creating a permanent enhancement in the risk management skills of participating banks that would be reflected in improved credit policies and lending strategies.
• Better risk management techniques developed from experience with guaranteed lending are likely to be applied to lending in general and to be adopted and adapted across the banking community, creating social benefits as experience is gained.
• CEGs leverage financial capital: better risk management should increase banks' bid ratios enabling guarantee funds to underwrite even more lending.
• Moral hazard is greatly reduced or eliminated - banks would have undiminished incentives for good credit decisions and loan administration because their capital remains at stake.
• The process is entirely transparent, diminishing possibilities for corruption and perpetuation of bad practices and making it possible to identify additionality, which is the difference between what would happen with a guarantee and what would occur in the absence of a guarantee.
• Transparency and additionality are benefits that increase welfare for the economy and society as a whole.
• The auction process makes it impossible (except through collusion) for those responsible for operating auctions and setting conditions for guarantees to direct guarantees to specific borrowers. Administrative direction of guarantees would corrupt the system and usurp lenders’ role and responsibilities in credit allocation and risk management.
• Donor funds for action as capital enhancement guarantees can be disbursed very quickly.
• Administrative costs are trivial.
• No permanent guarantee organization is created, which means that future losses are contained and taxpayers are not burdened, as virtually all such organizations fail or require continued taxpayer support.
• When guarantee funds are exhausted through auctions, they remain at work in the banking system as capital or as experience with losses or both.
• There is no social loss, discontinuity or painful transition requirement created by the termination of the guarantee programme.
• Finally CEGs can be implemented very quickly compared to the years required to build a traditional guarantee mechanism.
POLICY REFORM APPROACH TO DEVELOP CREDIT GUARANTEES
If guarantees are viewed simply a means to an end and not an end in themselves,1 the targeting of credit to specific groups and the terms on which they can borrow can be facilitated through public policy. This can be done without the creation of any new special purpose organizations and without the injection of either public or donor funds, although not without some direct and indirect costs. The effects of guarantees can be replicated largely by modifying the policy environment in which banks operate so that the incentives to lend to specified groups are altered.
Changing the Banking Rules
Banks are generally held to two sets of criteria that arguably constrain lending to SME and micro borrowers as well as to other credit constrained groups. Banks are required to maintain an 8% capital adequacy ratio. That is, the bank must have $8 of capital for every $100 of loans. This ratio was established as part of the Belse Accord, in 1988, updated I 1997. With the backing of the bank for International Settlements, Balse,.it is widely recognized and used, though many argue that while 8% may be a reasonable number for developed countries, lending in developing countries is inherently riskier due to political and economic factors and should be supported by higher levels of capital. If it is clear public policy to channel credit to specific groups in the society, the ratio can be changed so those banks require less capital to make certain classes of loans. Undoubtedly, the result of such a policy of requiring less capital for more risky loans is that the “credit quality” will decline as a result. However, these loans are only one part of an overall loan portfolio. The credit deterioration on this small piece may be an acceptable economic price for the broader society to pay (more risky loans and weaker, more exposed banks) as the cost of bringing additional groups into the credit system.
Its “risk weighted assets” determine the capital of a bank. The value assigned to an asset is variable. Cash, government securities, and loans are given different weighting in the formula to determine the bank’s capital. As with the 8% rule, there is general agreement as to how the weighting is applied. However, if it is public policy to move credit to a specific sector, the weighting could be changed to in effect provide banks with additional capital by assigning Certainly many guarantee supporters and promoters with an intellectual commitment to mutual and selfhelp organizations would not agree that guarantees are simply a means to an end. They see these organizations as having social and political functions of building group solidarity and often political allegiances that extend beyond the economic realm of simply making more credit available to a specific group.
different weights to loans to SME or micro borrowers. Bankers would have a fairly strong incentive to seek out these loans if they were given favoured treatment for the simple reason that they could make more of them with the same capital. The societal costs would be the same as with a weakening of the capital adequacy ratio—riskier banks less able to withstand a financial shock. Again, that may be an acceptable social price as these loans are likely to be a relatively small part of the overall bank loan portfolio and the aggregate credit quality deterioration may be on balance slight.
Changing The Collateral Requirements
If we accept guarantee fund advocates’ argument that collateral is a severe constraint to lending to SME and micro enterprises, and if there is agreement that public policy should favour lending to specified groups, then the collateral constraint can be modified quite simply by requiring less collateral for specific types of loans. Bankers often report that they do not make loans to SMEs because bank regulators during loan reviews inspect the collateral. The collateral offered by SMEs may be of dubious quality in that banks cannot easily realize it and may not have a ready market in which it can be sold. The legal system may make realizing collateral a difficult, costly and lengthy process. And, often banks confront social organizations that help debtors resist bank collection efforts. The result of this situation is that the collateral is severely “discounted.” Banks and most probably bank regulators have an implicit discount factor in mind for certain types of loans. It is not uncommon to find that loans to SME and micro borrowers are supported by 200-300% collateral. However, when the implicit “discount” factor is applied, the collateral is likely closer to the actual size of the loan. If banks cannot obtain this level of collateral they are reluctant to lend because they fear that the regulatory authorities may “classify” the loans and force the bank to provision them. Setting up provisions (reserves against loan losses) is costly to banks, as they cannot lend the money that is put into provisions. Lending can be promoted at the same social cost noted above by modifying this constraint. The level of collateral for certain classes of loans can be set lower than the present hurdle. The social cost is the same: a somewhat weaker banking system with lower quality credits.
These measures all weaken the quality of a bank’s loan portfolio. Bankers would normally be unwilling to voluntarily participate in such a programme without some offsetting financial advantage to them. These incentives can be provided through some modifications to banking regulatory policies.
Extending The Term Of Lending
Banks often are unwilling to lend or to extend the term of lending because regulatory authorities require them to run a “matched book” where the maturity of their deposits parallel the maturity of their loans. This, for example, is the case in Poland where bankers report that central bank policy severely constrains term lending as the bulk of their deposits are for 30 and 60 days and they can only originate a one year loan when they have an offsetting one year deposit. It is typically the case in developing countries that most deposits are relatively short term; therefore most lending is short term. While the deposits may be short term, the total volume of deposits may be relatively stable over time. There is usually some quantity of deposits that can be classed as “core” deposits that will remain with the banks, although individual depositors may come and go. Regulatory policy that would permit banks to run a “long book” (borrow short and lend long) for certain classes of borrowers could change the terms of lending. Banks could for example be authorized make some (probably relatively small) volume of 2-3 year term loans to SMEs against 3 month and 6 “core” month deposits instead of having to “match” the deposit terms with those of the loans. Assuming a positive interest rate curve where longer-term loans pay higher rates of interest than shorter-term loans, bankers would have an incentive to seek out these more profitable loans. Banks would presumably want to make such loans on a floating basis in volatile interest rate environments so that an increase in rates would not leave them with below market rate term loans. They would likely want to index the loans to some inflation index or perhaps to some spread over the average cost of funds. Again, such a policy is not without costs, although the costs are difficult to quantify and are contingent in that they only become apparent during a financial crisis or an extended recession. Credit quality is lowered somewhat and banks are somewhat riskier as a consequence of running a long book.
Removing the “Hidden Tax”
Banks are often required to make non-interest-bearing deposits with the central bank. In one interesting case in Venezuela, the legal deposit (encaje legal in Spanish) was used to finance agricultural production. Banks have been traditionally very reticent to lend to agriculture both because of its inherent risk (due both to climatological factors and government price and import policy). Initially banks were required to lend 10% of their total volume of loans to agriculture but this policy was severely resisted and evasion levels were very high (as they were in Nigeria). Instead, the government gave banks the choice of deposits at the central bank at no interest or lending to agriculture at a slightly below market preferential rate. All banks chose to have some interest income versus having none. To protect themselves, they organized together with the universities and one large foundation with close ties to a large agroindustrial group, the technical assistance and arranged for insurance to cover most of the climatological risks. On balance, the programme worked reasonably well with low levels of loan losses due to careful selection and extensive extension work, although the programme has been criticized for its below market rates and its targeting to some politically influential groups of farmers who are not very efficient producers. The cost of such a policy is that the central bank is deprived of one (but only one relatively unimportant) tool of controlling the total volume of loans in the financial system.
Using Tax Policy To Promote Lending
Finally, the functional results of guarantees can also be promoted by tax policy. Banks could be allowed accelerated write-offs of losses on certain types of loans. Thus, banks when they suffered losses or even when they created provisions could write these losses off against profits in the year in which they occurred. Banks thus would be able to deduct provisions from taxes. If the loan were subsequently recovered either totally or partially, the write off would be reversed. There are clearly identifiable costs to this approach. Some tax revenues will be forgone as a result of accelerated write-offs.
Policy Reform: Costs And Benefits
This functional approach to guarantees may achieve the same results as establishing a fund. It is not suggested here that any one single policy will achieve the desired goals; in practice, it is likely that some mix of policies will be needed. Furthermore, it is likely that in a given country, once the paradigm is refocused on public policy instead of institutional creation, a broad range of other policies will suggest themselves as means to the same end of a more inclusive credit system. All of these approaches derived from a paradigm shift have several
• They require an agreement at the political level that some social and economic costs to the society are an acceptable price for a more inclusive credit system. This agreement is not intrinsically different than many of the other agreements and compromises incorporated into any banking system.
• In some cases, law change may be required; in almost all some rewriting of the regulatory norms is necessary. More difficult still will be to bring bankers and regulators alike to understand and apply the new policies.
• No special function organization is created. As noted above, once created these organizations are both expensive to maintain and become a political constituency that works to acquire and maintain a subsidy from either the state and/or donors.
• If the volume of loans to target groups is low for whatever reason, the policies supporting the lending are cost free and simply disappear into the background of unused law and regulations.
• The administrative costs are trivial. There is no staff to process guarantee applications and to pay claims. With a reduction to zero of bureaucracy the borrower deals only with the bank and does not have the duplicate documents and delays that characterize most guarantee operations.
• The enforcement mechanism to minimize fraud, abuse, and corruption already exists.