Can Governments Mandate Hard Budget Constraints? Bank Lending and Financial Isolation in Romania

Abstract

This paper examines the empirical evidence on the impact of reforms in the financial sector in Romania in the period 1993-1995. The methodological framework of the paper is based on a theoretical model of intertemporal bank lending in a transition country with uncertain prospects for stabilization (Carare and Perotti, 1997). The model identifies an empirical test to measure whether the banking sector has started to act as a source of financial discipline, or just as a temporary buffer for enterprise losses. The idea is to test whether the correlation between bank lending and arrears is decreasing over time. We also seek to assess the structure and impact of the program of financial isolation initiated by the government in 1994 to impose tighter budget constraints on a set of insolvent state-owned firms. We use both a logit specification and a structural test to analyze the determinants of the selection of firm included in the program. Our results indicate that after the early stage of collective bailouts ended in early 1993, banks at first acted as channel to support insolvent firms. The evidence suggests also that by 1995, when the financial isolation program started, financial policies became more discriminative. While the criteria for the overall credit allocation appear to have improved, there are signs that support was shifted to a selected group of enterprises, perhaps in part through the program itself. However, only once more recent data becomes available it would be possible to investigate whether there has been a change in the initial discipline imposed on the financially isolated firms as a result of the program.