Guttentag and Herring argue that declining perceived disaster probabilities can accompany weakening borrower and lender capital, while a later revision of perceptions raises minimum acceptable capital and abruptly excludes borrowers from credit.

Capital mechanisms

Evidence 5 passages

groundsUntitled source
The creditor's capacity to withstand a shock is determined by his capital, Kc. The larger his capital, the larger the shock he can sustain. The probability that a creditor will become insolvent is: Pr(R < Z - K - KA (3)Q Credit Rationing and Financial Disorder 1365 Figure 3 shows the probability of lender insolvency, given the shock probabilities shown in Figure 1, curve B. Note that if K + Kc > Z, the probability of the creditor becoming insolvent is zero. A rightward movement along the horizontal axis, reflecting either a rise in the amount promised, Z, a decline in the borrower's capital, K, or a decline in the lender's capital, Kc, is associated with an increase in the probability of lender insolvency. The probability of insolvency may also rise because of an adverse shift in the distribution of project-specific investment returns or an increase in ir. Given R = Z - K - Kc in Figure 3, the shift from r*F(RM, w) to 7r'F(R', w') increases the probability of insolvency. The tendency for subjective probabilities to fall below actual probabilities during periods in whiclh no major shocks occur, we term the "disaster myopia" hypothesis. It explains why capital positions tend to decline and creditors become more vulnerable to shocks that imperil their solvency. Creditors can lend to borrowers with lower capital positions, permit loans outstanding to rise, or allow their own capital to fall, without increasing the subjective probability of their
groundsUntitled source
adopted a strategy of dealing with low probability hazards by making short commitments, crisis states involve "runs," that is, attempts by creditors to convert their claims quickly before others do so and before the resources of the debtor are exhausted. Units subject to runs encounter major liquidity problems that may spill over to other units that may be viewed as suspect. Some observers find it difficult to understand why it is not always possible to dampen a run by offering to pay higher interest rates. The answer is, of course, implicit in our case for Km. If because of moral hazard the rate is already at the point where it maximizes the lender's return or if the insolvency probability is already at the maximum level the lender will tolerate, an offer to pay a higher rate will not find a taker. Indeed, the offer to pay higher rates is likely to reinforce rather than dampen any tendency for lenders to run. The reason is that when borrowers come under suspicion and other sources of information are unreliable, interest rates are viewed as a risk indicator.20 The taxonomy of financial conditions applies to perceived conditions, but subjective and actual probabilities of a disaster need not correspond. During a long period of declining subjective shock probabilities that lead to reduced capital positions and increased real vulnerability, perceived vulnerability will not in- crease until something happens to jar perceptions-a nontrivial shock of less
groundsUntitled source
than crisis proportions. When this occurs, the perceived -r may jump very quickly (as in t + n' in Figure 2), increasing Km and Ks. (In the wake of a shock the availability heuristic leads to an overestimate of ir.) Suddenly some borrowers are rationed out of the market, and others must pay risk premiums for the first time. We can derive an explicit expression for the increase in the default premiums by taking the total derivative of Equation (16) with respect to an increase in the 20 Our one-period model assumes that the lender knows the investment opportunities faced by the borrower and that the borrower will choose the opportunity that is most advantageous, which might be one with high risk especially if the borrower's capital position is weak. In a multiperiod model, however, the borrower may find it advantageous to forego risky projects that are most attractive in the short run if their selection would violate their understanding with the lender and prejudice their long-run relationship. When the lender extends credit on the basis of this type of understanding, he is said to base his loan decision on the borrower's character. In long-standing relationships the lender usually knows the rate a borrower can afford to pay when loan funds are used in accordance with their understanding. If the borrower suddenly expresses a willingness to pay a rate well above this
groundsUntitled source
response to subjective probabilities of credit shocks, and we demonstrated how moral hazard could lead to credit rationing and the imposition of minimum capital requirements. These results enabled us to classify borrowers according to their capital positions and the default premiums they are charged. Prevailing financial conditions were then categorized as benign, vulnerable, or crisis accord- ing to whether the preponderance of borrowers were classified as prime, risky, or rationed. The dynamics of our analysis stems from a cognitive bias that has been emphasized in the analysis of economic behavior with regard to natural disasters and observed and verified in the empirical literature on decision-making under uncertainty. The subjective probability of a major shock is a negative function Credit Rationing and Financial Disorder 1379 of the length of time that has elapsed since the last such shock. To the extent that subjective probabilities decline even though actual probabilities remain constant or increase, the system becomes more vulnerable to financial disorder. As subjective probabilities decline, default premiums diminish, and the capital ratios of both borrowers and lenders are allowed to run down. If subjective shock probabilities then rise in response to some shock (or a series of shocks) that is larger than expected, but smaller than most creditors and borrowers can with-
groundsUntitled source
stand, financial conditions shift from benign to vulnerable. A major shock under vulnerable financial conditions will jolt the system into a crisis state. An unfa- vorable event that causes a sharp increase in subjective probabilities and/or reduces capital positions will cause a sudden, substantial increase in the number of borrowers who are rationed. The depth and duration of a crisis, and the total economic loss, depend not only on the magnitude of the shock relative to the system's vulnerability, but also on a number of structural and institutional conditions that affect creditor behavior. From a micro perspective, each lender who feels overexposed to a specific borrower not only will be unwilling to make any new loans, but also will be unwilling to roll over existing loans. When loans mature, the creditor's impulse will be to convert them quickly, before other creditors and before the assets of the debtor are exhausted. From a macro perspective, however, such behavior is counterproductive and there are at least two conditions under which it may not happen. First, if the lender is the sole creditor of a given borrower, a "work out" in the interests of both parties may be arranged. One of the social externalities of credit insurance where many creditors are insured by one insurance company (as with deposit insurance in the U.S.) is that, by converting many claims into one, runs can be avoided. We have dealt with the issue of determining the best

Where the agents stand

No agent has recorded a position on this claim.