The Effects of Screening and Monitoring on Credit Rationing of SMEs

Authors

  • Mariarosaria Agostino,

    1. Dipartimento di Economia e Statistica, Università della Calabria, Italy
    Search for more papers by this author
  • Damiano B. Silipo,

    Corresponding author
    1. Dipartimento di Economia e Statistica, Università della Calabria, Italy
      †Corresponding author: Damiano B. Silipo, Dipartimento di Economia e Statistica, Università della Calabria, Ponte P. Bucci, Cubo 0-C, 87036 Arcavacata di Rende (CS), Italy. Tel: +39 0984 492451; Fax: +39 0984 492421. E-mail: silipo@unical.it
    Search for more papers by this author
  • Francesco Trivieri

    1. Dipartimento di Economia e Statistica, Università della Calabria, Italy
    Search for more papers by this author

  • We thank doctor Giovanni Butera of Moody's KMV for kindly providing data on firms' risk. We are also grateful to two anonymous referees for their helpful comments and suggestions on previous versions of the paper.

†Corresponding author: Damiano B. Silipo, Dipartimento di Economia e Statistica, Università della Calabria, Ponte P. Bucci, Cubo 0-C, 87036 Arcavacata di Rende (CS), Italy. Tel: +39 0984 492451; Fax: +39 0984 492421. E-mail: silipo@unical.it

Abstract

In this paper, we seek to empirically assess which determinants of the capability and incentives of banks to screen and monitor firms are significant in explaining credit rationing to Italian SMEs. After testing for the presence of non-random selection bias and the potential endogeneity of some determinants of interest, the probit model results we obtain suggest that the average banking size and the multiple banking relationship phenomenon are statistically significant factors affecting credit rationing, presumably through their impact on the aforementioned banks' capability and incentives. Other potential determinants of banks' incentives to monitor and screen, such as local banking competition and firm' capacity to collateralize, are never significant. However, when we split the sample according to the level of competition in credit markets, we find that the estimated marginal effects of all significant determinants of interest are larger in absolute value than those obtained when using the whole sample.

Ancillary