| Popis: |
In this paper we use geographical and time variations to estimate the impact of war and occupation on firms in Croatia. We propose that during occupation firm exit was significantly subdued in the whole country, irrespective of occupation. After war, firm exit intensifies, dividing into three clusters. We find a clear and very strong positive effect of a firm being located in a municipality that was occupied at some point in time, on firm exit. The movement of war from 0 to 1 produced a 3.8 percentage point change in the probability of firm exit. As the sample mean amounts up to 12.2 percent, war increases the mean up to 16.0 percent. When we further separate occupied territory into "middle" and "east", the effect is much stronger in the "middle" territory, and it suggests that the movement of war from 0 to 1 produced a 4.3 percentage point change in the probability that firm exit occurs, or that war increases the mean up to 16.5 percent. We believe that this result comes from differences in the termination of occupation that was violent in the "middle" case, and peaceful afterwards. There is also a negative effect of the number of employees, capital and revenues, implying that larger capital and revenues, and higher number of employees decrease the probability of firm exit. For example, a 100 percent change in revenues generates a 0.9 percentage point decrease in the probability of exit. The probability of firm exit for a firm in the least capital-intensive sector goes up by 5.4 percentage points when a firm is in a municipality that was occupied in some period. For the smallest minimum efficient scale, the probability of firm exit increases by 4.3 percentage points when a firm is in a municipality that was occupied in some period. |
