Aggregate and Idiosyncratic Political Risk: Measurement and Effects
|Speaker:||Laurence van Lent, Tilburg University|
|Date:||Wednesday 22 February 2017|
|Time:||14:30 - 16:00|
|Location:||Kolade Teaching Room, Building One|
We propose a new measure of political risk faced by individual US-firms based on textual analysis of earnings conference call transcripts: the share of the conversation between management and analysts that is devoted to political topics. Our measure correlates significantly with firm-level stock return volatility, even after controlling for firm and time fixed effects. We find that increases in idiosyncratic political risk are associated with decreases in investment and hiring, and that the dispersion of idiosyncratic political risk tends to increase significantly in times of high aggregate political risk. About two thirds of the variation in political risk is idiosyncratic in the sense that it is neither captured by firm or time fixed effects, nor by heterogeneous exposure of individual firms to aggregate political risk. Further decomposing our measure by political topic, we find that discussion of risk associated with corporate regulation and health care is associated with the largest decreases in investment. We also find that firms actively manage political risk through lobbying: firms that devote more time to discussing the risk associated with a given political topic tend to increase lobbying expenses on that topic that quarter. These effects are most pronounced for large firms and firms headquartered in states that are associated with higher levels of political corruption.