The authors examine the real consequences of share repurchases driven by earnings management considerations. Firms that would have just missed the earnings per share (EPS) forecast in the absence of a share repurchase are more likely to repurchase their shares than firms that just beat the EPS forecast. This behavior creates a discontinuity in the probability of a share repurchase around the zero earnings surprise level. The authors use this discontinuity to assess the causal effect of such repurchases on other corporate policies using a regression discontinuity framework.

They find that firms that repurchase shares subsequently reduce employment and investment in capital. They also find that the stock market reacts positively to EPS forecasts that are met by using share repurchases at the time of the earnings announcement. On average, this market reaction is not different for firms that just meet their earnings targets. But the share price reaction is lower for those firms that reduce real variables, such as R&D and employment, to help finance repurchases.

The research adds to the literature that examines share repurchases. The authors make the argument that EPS-driven repurchases cause firms to reduce investment, employment, and R&D. They also examine the impact of such EPS-driven repurchases on the valuation of the purchasing firm. The authors note that the interpretation of their analysis of the impact on valuation is complicated because firms endogenously choose how to finance EPS-motivated share repurchases. Nevertheless, their analysis indicates that a number of firms in their sample are willing to sacrifice valuable investments to finance EPS-motivated share repurchases.

The research enriches the existing framework for assessing share repurchases. Specifically, it provides a more nuanced approach to assessing the impact of share repurchases for firms that just meet EPS expectations. Two key groups of practitioners will find the research very useful. First, investment analysts may be able to refine their assessment of the impact of share repurchases on investment returns. Second, management can use the framework to refine the assessment of their share repurchase financing options and the way markets react to their repurchasing decisions.

The authors test the causal effect of repurchases on investments including capital expenditures, employment, and R&D. They regress changes in investment on share repurchases, instrumented with an indicator of whether a firm would announce a negative EPS surprise without a repurchase.

The regressions compare firms that just miss the EPS consensus forecast (treatment group) with firms that just beat the consensus forecast (control group) and the relationship between changes in dividend policy (dependent variable) and stock price changes (independent variable). To ensure that the discontinuity in the likelihood of share repurchases is properly identified, the authors limit the sample to a small window around zero pre-repurchase EPS surprises.

The authors use Compustat as their main data source. They start with all firm-quarter observations in the database between 1988 and 2010. They exclude all utility firms and financial services firms as well as all firm-quarters with missing or nonpositive assets. They end up with about 390,000 firm-quarter observations. They merge these observations with stock-level data from CRSP and analyst forecast data from I/B/E/S.

There are currently high levels of cash on corporate balance sheets, especially in the United States. Such companies have faced pressure from activist shareholders, institutional investors, the government, and media for them to put their cash to use. Shareholders want cash returned through dividends or share repurchases or invested in high-return ventures. The government would like to see the cash invested in the real economy to help stimulate economic growth. With this context, the research is very timely. The analysis used is well structured and can be replicated by practitioners. The key insights are mostly intuitive and accessible to a whole range of readers. Nevertheless, as the authors admit, by focusing on firms that just miss or just beat their EPS expectations, the applicability of the insights is constrained. This constraint presents an opportunity for further research to be done to refine our understanding of the impact of share repurchases on real outcomes.