Sudden Stops, Output Drops and Credit Collapses

Abstract  This paper develops a model of a small open economy where Sudden Stops can be generated through abrupt downward shifts in expectations about future growth rates. Sudden reversals in capital flows can lead, in this model, to major drops in out-put due to the presence of financial frictions. The analysis is carried at the firm level allowing for firm heterogeneity. Within this framework one can study the main determinants of firms’ performance during a Sudden Stop episode. The paper documents the patterns displayed in firm level data from Southeast Asia during and following the Asian crisis and shows how these features can be explained by the model. In addition to the sudden reversal and the subsequent drop in output the model can explain the credit-less recovery pattern which has been recently documented to be a main characteristic of these episodes. This feature is also strongly displayed in the firm level data. Earlier models that rely on TFP shocks, produce procyclical debt levels and asset prices and therefore cannot match the patterns observed during therecovery. The model also provides an explanation for the strong positive size effect that this paper documents in the data. The argument relies on the hypothesis that small firms are “growth firms” which is supported by evidence from the data. The model produces a prediction that a higher pre-crisis Tobin’s Q should be associated with a lower performance during the crisis which is also consistent with findings from the data. Finally it is shown that growth shocks produce a strong leverage effect. In this model, it is the costly adjustment in debt levels that is responsible for the drop in output. Therefore, as observed in the data, firms that carry little debt are not significantly affected by the crisis.

Source: Worldscope, Thomson Financial.

                         Notes: This figure plots an index of average sales, debt, investment and credit where the 1997 value is normalized

                                    to 100. Firms are ranked by their size in 1996 based on the dollar value their fixed assets.  This figure com-

                                    pares the upper 35% (which we call large firms) to the lower 35% (the smaller firms).