The distribution of firm growth and business cycles

In any given economy, the majority of firms are small while only a handful of firms are relatively large. However, overall aggregate cyclicality is heavily dependent on the behavior of these few, large firms. Specifically, the behavior of large firms plays a fundamental role in the amplification of shocks.

This article revisits the role of large firms in business cycle fluctuations in the Portuguese context. The findings provide evidence that the effect of large aggregate shocks, while affecting all firms, both small and large, are particularly pronounced for the largest firms. The interpretation of the evidence is that while large firms may be relatively more diversified, they appear particularly exposed to the more extreme aggregate shocks. Consequently, it is not uncommon for a large firm to experience a 50% increase or reduction in sales in response to larger aggregate shocks.

The aggregate shocks may arise from either the demand or supply side. The paper attributes the empirical findings to demand shocks, following recent empirical evidence of heavy tails to output shocks. These shocks affect only the firms connected to the affected demand and are equally distributed around zero. Large shocks to demand, coupled with a concentration of firm sales in a few large customers, limit the potential for diversification, even for large firms. This, in turn, generates large firm-level demand shocks, which can be positive or negative.

The paper shows how these shocks have aggregate implications. The kurtosis of firm-level growth rates –a measure that assesses the likelihood of extreme positive or negative events—is strongly correlated with GDP growth, with a correlation coefficient of 0.78 (see figure). Taken together, the results challenge the notion of risk diversification.

Click here to go to the paper by Carlos Daniel Santos.