A dynamic model including a single lag of current loss ratio was estimated usingthe Arellano and Bond (1991) estimator. The lagged dependent variable has apositive estimated coefficient and was significant at the 10 per cent level, thoughunlike for stocks of non-performing assets, there is no strong reason to expecttrue state dependence over short horizons in the flow of current losses.Estimated coefficients on key explanatory variables in this model werequantitatively similar to Model C. Banks learning from their mistakes maycreate a negative relationship between credit losses in an earlier downturn andthose in a later downturn – something akin to the ‘institutional memoryhypothesis’ of Berger and Udell (2004). But only a small number of the banks inthe sampleduringthe global financial crisis episode were not present during theearly 1990s downturn, so testing this hypothesis is difficult.
No comments: