Imperfect information and credit rationing equilibrium: evidence from Egypt

Abstract

This paper investigates the empirical significance of the concept of credit rationing equilibrium. The evidence from a disequilibrium model of the credit market with application to Egypt provides support for this concept. In particular, loan supply is not found an increasing function of the return on intermediated credit. Also, expectations on the state of the economy appear to explain the changing pattern of credit rationing over time.

Keywords:

• Imperfect information
• Loan contracts
• Credit rationing equilibrium
• JEL CLASSIFICATIONS: E44
• G21
• G28

Notes

 Stiglitz and Weiss (1986) have shown that loan collateral amounts to the provision of equity finance by borrowers for risk-sharing purposes. Accordingly, there is no loss of generality with the assumption that entrepreneurs exclusively finance their investments by borrowing from banks while putting up collateral.

 See, for example, Laffont and Martimort (2002, Section 2.15.4) for a setting in which a lender may screen borrowers according to the size of their loans.

 The existence of credit rationing equilibrium when banks use collateral as a screening mechanism is sensitive to the assumptions on market structure and attitudes towards risk. See, for example, Bester (1985) and Besanko and Thakor (1987).

 The foreign-exchange component of investment is important in the Egyptian economy, as reflected in a significant import bill of producer goods. Taking account of business loans in both domestic- and foreign-currency is therefore relevant to the present analysis.

 The interest on foreign-currency deposits in the domestic market is reported in central bank bulletins, and is determined by banks in line with the prevailing rates in the international financial markets (e.g. LIBOR). The assumed interest differential of 2% is consistent with the average practice in the Egyptian market.

 See Alastair (1994) and Ng and Perron (1995) on merits of the general-to-specific method for lag selection in the ADF unit root test.

 The correlogram of each series does not display a seasonal pattern, and so the unit root tests are carried out for the non-seasonally adjusted series.

 A brief summary of bank regulatory policy in Egypt may be found in IMF (1998, Appendix II).