Two Essays on Competition, Corporate Investments, and Corporate Earnings
The general focus of my dissertation, which consists of two essays, is on how changes in the financial and economic environment surrounding a firm affect managerial incentives and firm policies regarding investment in physical capital, innovation, equity offerings, and repurchases. The first essay in my dissertation examines how product market competition affects firms' investment decisions. While competition among firms benefits consumers via lower prices, greater product variety, higher product quality, and greater innovation, recent studies provide evidence that competition has been declining in the U.S. economy over the past decade. The evidence shows that American firms' profits are at near-record levels relative to GDP and are persistent. Industries have become more concentrated as a result of mergers and acquisitions, and barriers to entry have risen and the rate of new entry has been declining for decades. Taking these findings at face value, we examine empirically whether companies feel less compelled to invest in physical capital and in research and development because they face fewer threats from rival firms.
Using both traditional proxies and recently developed text-based measures of industry concentration, we show that firms operating in competitive industries invest significantly more in both physical capital and research and development relative to their peers in concentrated industries. We also report that the propensity to invest less by managers of monopolistic firms is partially mitigated by superior corporate governance that reduces the agency problem, and by certain product market characteristics such as low pricing power and low product differentiation/entry barriers. However, after accounting for all these mitigating factors, the negative association between industry concentration and investment persists. Our results are robust to including various control variables and exclusion of firms from industries that face significant competition from imports. The results are also robust to controlling for endogeneity caused by missing time-invariant and time-varying industry level factors that could potentially be related to both the level of concentration and investments.
Overall, our results are consistent with the notion that firms in competitive industries have a greater incentive to invest and innovate to survive and thrive in a competitive environment relative to the managers of the firms in more concentrated industries whose incentive to invest and innovate is to maintain their monopoly rents. Our findings have obvious policy implications in that investment and hence economic growth is being adversely affected in the current era of increasing industry concentration and declining competition.
The second essay in my dissertation investigates whether information contained in equity issues and buybacks is fully incorporated into prices such that the market reaction to subsequent earnings announcements is unrelated to those corporate actions. Korajczyk at al. (1991) argue that firms prefer to issue equity when the market is most informed about the quality of the firm to prevent adverse selection costs associated with new equity issues. This implies that equity issues tend to follow credible information releases contained in earnings announcements. However, analyzing a sample of 19,466 SEO pricing dates between 1970 and 2015 and 15,106 buyback announcements between 1994 and 2015 shows that a considerable number of equity offerings and repurchase announcements take place before the announcement of earnings. About 28% of buybacks and 32% of SEO pricings are made in the three weeks prior to an earnings announcement. Given these statistics, we examine whether these corporate actions provide information about upcoming earnings announcements (earnings predictability) to the extent that new information has not been fully incorporated into prices by market participants.
We find evidence of earnings predictability: the market reaction to earnings following buyback announcements is higher by 5.1% than the reaction to earnings following equity issues over the (-1,+30) window when four-factor abnormal returns are used; the difference is 2.2% when unadjusted returns are considered. The results are robust to several alternate sample construction methodologies. There are at least two puzzling effects of earnings predictability that are difficult to reconcile with the market efficiency hypothesis. First, there is an incomplete adjustment to SEO pricings and buyback announcements that results in residual market reaction to earnings announcements. Second, prices continue to drift after earnings announcements: upward for buybacks and downward for SEO pricings. Unlike post-earnings announcement drift, the drift documented here does not depend on the market reaction to earnings announcement. We test several reasons for this anomalous behavior including prior returns, price, size of buyback or SEO, analyst forecast errors, and bid-ask spread. We find that information asymmetry proxies partially explain the persistence of earnings predictability following SEO pricings and buyback announcements.