[This is another of a series of editorials by Executive Editor Andrew Karolyi at the Review of Financial Studies featuring recently published papers at the journal. This editorial features “Policy Uncertainty and Corporate Investment” by Purdue University’s Huseyin Gulen and University of Arizona’s Mihai Ion, an article in Issue 29(3) for March 2016. It was selected as an Editor’s Choice article on the Oxford University Press web site for RFS.]
It is hard to travel in the academic circles of Finance without engaging in some impromptu discussion about the currently volatile political environment in the U.S. Whether the conversation revolves around the candidates, the polls, the calendar of primaries, or the “substance” of a recent debate, it seems to capture our collective fancy. Intriguingly, those among our colleagues who hail from other countries around the world seem to be as caught up in the coffee-break chat as any others. I cannot help but wonder about the opportunity cost of this attention and count in my head how many important research papers could be hatched and fostered in the absence of these distractions.
Slowdowns in research paper production are one thing, but delays in corporate investment activity are another. And this is precisely the question that a new article by Huseyin Gulen and Mihai Ion pursues. Their paper focuses on heightened policy uncertainty and how it is associated with significant delays in investments. The economic magnitudes are striking: on average two-year delays and as much as a one-third decline in corporate investments were observed during the recent financial crisis and aftermath. The authors emphasize that the effect is most acute among firms for which investments are likely to be more irreversible, sensibly anchored in real option theory, and which are more reliant on government spending.
One of the more intriguing aspects of their paper—equal in my mind with how they quantify the link between general policy uncertainty and investments—is how they operationalize various measures of economic policy uncertainty. This measurement business has been a fruitful area of research in macro-finance, applied econometrics, and even political science. The authors agree with me that this part of their paper, in fact, may be one of their most important contributions that readers risk overlooking. The bulk of their paper employs a measure of economic policy uncertainty developed by Scott Baker, Nick Bloom, and Steven Davis (BBD). BBD construct a policy-related economic uncertainty index from three components: one based on newspaper coverage of economic policies, a second based on federal tax code provisions set to expire in coming years, and a third based on disagreement among economic forecasters. Details are at www.policyuncertainty.com, where interested readers can download the data and read the latest 2015 working paper on which it is based. Sure, there are others to consider, such as the VXO index or an intriguing index built using factor-augmented Vector Autoregression techniques by Kyle Jurado, Sydney Ludvigson, and Serena Ng in the American Economic Review in 2015. But what Gulen and Ion show is how resilient BBD’s index is to the inclusion of many different controls for general economic uncertainty. It makes one wonder which component of the BBD index is the live one for investment activity.
A special part of the Gulen-Ion paper (again, that might be overlooked by some) is that the BBD index’s link to delayed corporate investment activity seems to belie political election cycle forces, another research stream in Finance and Economics that has received a healthy dose of attention. Sufficient variation seems to happen outside election years, be it when debt-ceiling debates arise or when other government budget crises ensue.
So, the next time you and your colleagues gather around the water cooler to deconstruct the content—or name-calling—of the primary debate of the prior evening, feel comfortable that the delays in corporate investments (and perhaps financial economic research activity) are likely no worse than usual.