Response of Distressed Firms to Incentives: Thrift Institution Performance under the FSLIC Management Consignment Program

• Published in: Financial management Vol. 22; no. 3; pp. 176 - 184
• Main Authors: Barrow, Janice M., Horvitz, Paul M.
• Format: Journal Article
• Language: English
• Published: Albany, N.Y Financial Management Association 01.10.1993; Financial Management Association International; Blackwell Publishing Ltd
• Subjects: Bank loans; Conservatorship; Consignment; Cost control; Credit risk; Deposit insurance; Failure analysis; Financial Distress Special Issue; Financial management; Generally accepted accounting principles; Impacts; Incentives; Insolvency; Investment strategies; Management styles; Net worth; Profit margins; Profitability; Receivership; Savings & loan associations; Solvency; Solvents; Studies; Thrift; United States > US
• ISSN: 0046-3892; 1755-053X
• DOI: 10.2307/3665937

The Management Consignment Program (MCP) was adopted in 1985 by the Federal Savings and Loan Insurance Corporation (FSLIC) in an attempt to minimize the acute adverse incentive problems present when insolvent thrift institutions are allowed to continue in operation. The management of problem thrifts were replaced by new management teams, selected by federal regulators and compensated on a contract basis. They were expected to maintain service to depositors and improve the condition of the thrift's books and records while more permanent solutions were explored. Without close monitoring by the FSLIC, problem institutions with low or negative net worth would have an incentive to take on risky strategies which could further erode net worth. Under the MCP, given the new incentive structure, agency theory predicts that there would be no incentive to exert effort in other than a risk-averse way. However, in an attempt to preserve asset values, the new managers may lock in negative or inadequate profit margins, thereby precluding the possibility of a return to solvency by a successful (lucky) gamble for large profits. Therefore, although the MCP may have reduced total costs to the FSLIC, as a result of the absence of a profit motive and the conservative strategies followed, the chances of the MCP institutions recovering to solvency may have been significantly lowered when compared to similarly insolvent institutions that were allowed to operate outside of direct government control. Previous research on insolvency and failure within the thrift industry has focused on techniques to develop early warning signals or to evaluate the costs to the federal insurance funds when institutions fail. The contribution of this paper is twofold. First, a limitation of previous failure studies is that their primary goal is to examine the determinants of closure when solvency may be the more meaningful event to explain. This paper uses generally accepted accounting principles (GAAP) to determine insolvency in establishing a prediction model. Second, using the bankruptcy prediction model, this paper examines the extent to which inclusion in the MCP and the resulting alternative management structures affected the behavior of financially distressed thrifts. The methodology is as follows: (i) A two-step logit bankruptcy prediction model is developed. The logit model provides an estimate of an observation's probability of falling into prespecified categories. In this case, the categories were GAAP solvent and GAAP insolvent. The model coefficients were generated by using proxies for the factors that are expected to explain solvency. The variables included were proxies for credit risk, liquidity risk, interest rate risk, operational risk, and variables necessary to control for external factors. The model was validated by using a holdout sample to test predictive ability. (ii) The chances of recovering to solvency for the MCP institutions were then compared over time by using the logit model developed in step one. The results were tested by comparing the change in the mean scores over time and by comparing the average change for each institution between two time periods. (iii) The MCP institutions were then compared with similarly insolvent institutions that were not placed in the MCP but allowed to continue to operate. The model's accuracy rate of 91.5% compares favorably with the results of similar studies. It also correctly classifies 90.1% of the observations in the holdout sample. In determining the impact of the MCP on the probability of solvency over time, the results show that the probability of insolvency increased at an increasing rate, with the rate of increase peaking in the third quarter after an institution was placed in the program. The changes are consistent with our expectations that placing an institution in the MCP will reduce its chances of recovering to solvency. Part of the change, however, may be attributed to better reporting by the new managers rather than reflecting a decrease in operational efficiency. A second test compared the rate of deterioration of the MCP institutions with similar problem institutions that were not placed in the program. The results show that although the chances of recovery had decreased for both groups, the chances were less for MCP institutions than for the control group. The implication is that the program does not reduce the probability of a need for FSLIC financial assistance in the final resolution. The reduced chances of recovery by the MCP institutions are presumably compensated for by a reduction in the size of the insurance losses. There are savings from resolving problem institutions as early as possible, but if resources are inadequate to do so in a timely manner, then an interim program is necessary. This paper contributes empirical evidence on the relative merits of alternative approaches for dealing with distressed depository institutions during the interval between the time a bank becomes critically undercapitalized and the time a permanent solution is in place.