A S 1 Udy of Debt Servicing Capaci ' Iy Applying Logit Analysis

This paper empirically investigates the importance of various economic factors in determining debt servicing capacity of borrowing countries. The paper builds on earlier work by Frank and Cline (1971) and Dhonte (1975) who tried to identify empirically the more important factors. Using logit analysis and significance tests, this paper suggests several factors which are important determinants of default probabilities. These findings are consistent with the descriptive literature on international borrowing. The estimates for predictive ability of the model are checked with alternative data and confirm a relatively low rate of error (around 4 percent). THE PROOFS SHOULD BE MAILED TO: Professor Richard E. Just Department of Agricultural and Resource Economics 207 Giannini Hall University of California Berkeley, CA 94720 A S1UDY OF DEBT SERVICING CAPACITY APPLYING LOGIT ANALYSIS Debt servicing capacity of borrowing countries is undoubtedly a subject of i nterest and concern for international lending organizations and institutions. But quantitative knowledge of the importance of determinants of debt servicing capacity is also useful for borrowing countries in formulating policies which affect their determinants and t hus their credit availability . and default probability. As in the case of private loans to individuals, lenders consider the possibility that the borrower will fail to r epay all agreed principal and interest depending on his ability to honor financial obligations; and interest rates are supposedly determined accordingly. At the international level, much discussion has focused on determining the economic variables which should be considered in making projections regarding the development of the borrower's performance [e.g., Alter (1961), Finch (1951), Mikesell (1962), Gulhati (1967), and Bittermann (1973)] . But most of this work has provided only verbal arguments for considering one variable or another. Only the studies by Frank and Cline (1971) and, to a lesser degree, Dhonte (1975) represent an attempt to quantify the importance of various potential factors suggested in other studies. Dhonte has used principal component analysis to investigate the relations among several economic variables which are considered to be relevant for debt servicing capacity. No formal testing procedure, however, is employed for the purpose of selecting these variables. Frank and Cline, on the other hand, investigated the quantitative importance of indicators in determining default probability using discriminant analysis to identify each observation as belonging to one of two possible populations: default or nondefault. Their results indicate that the debt-service ratio, the debt-amortization ratio, and the ratio of imports to reserves are important determinants of debt servic1 ing capacity. 2. The present study is an attempt to improve upon the methods offered by previous studies for analyzing debt servicing capacity. Logit analysis is used i nstead of discriminant analysis because it is a method specifically developed to deal with the binaryvalued, dependent-variable case. While discriminant analysis assumes two completely different populations, the logit approach assumes a discrete "event" takes place after the combined effect of certain economic variables reaches some threshold level • . The latter approach is especially suitable when several observations (of · both default and nondefault years) for a given country are included. That is, it makes more sense to claim that , in a specific period, the country was pushed beyond a critical level leading it to a rescheduling than to claim that the country suddenly became a member of another species. In addition, more appropriate statistical tests can be performed to determine the relevance of various economic indicators of debt servicing capacity. The predictive performance of the estimates is quite good within the sample period as there are only 6 to 11 errors (in predicting 2 default or nondefault) with 238 observations. The results also allow some interesting estimates of the probability of default (rescheduling) for debtor countries given their current economic status. 1. Indicators of debt servicing capacity Nine economic indicators of debt servicing capacity are used in the analysis but, to facilitate a brief discussion of them, seven are defined the same as in the Frank and Cline study. The two additional indicators used in this study are capital inflows and growth of per capita domestic product. The measure of noncompressible imports used 3. i n the previous study, however, is not included because the data for calculating it were not comparable among countries for all of the years used and because theoretical arguments have been developed which qualify this indicator. The usual argument is that imports of various consumption goods, which are not vital necessities, can be curtailed temporarily so as to increase availability of foreign exchange for debt servicing purposes. The assessment of this factor thus requires detailed data on import composition patterns. Moreover, there may be raw materials and intermediate goods that are imported for production of domestic nones~ sential goods which can be reduced; but separation of these from other intermediate goods is usually impossible. Furthermore, the possibilities for reducing imports may depend heavily on a government's internal political status rather than on the economic importance of import items. Thus~ it seems that the notion of compressible imports may be of little empirical use until a reasonable approach for including political status 3 i s developed. Following the classification suggested by Avramovic et al. (1964, p. 13), a number of variables are apparent that affect debt servicing capacity in the short run. These variables focus mainly on components of the balance of payments which reflect a country's short-run transfer problems. The first variable considered is the debt-service ratio, i.e., the ratio of debt service to exports. As indicated by Bittermann (1973), the debt-service ratio is one of the most COIIllllon rules of thumb for credit-worthiness evaluation. Supposedly, a high ratio (indicating a