Do Short-Term Incentives Affect Long-Term Productivity?

Are US companies short-termist? Several recent papers identify actions that indicate short-termism, such as a reduction in long-term investment due to vesting equity and a reduction in R&D growth that allows firms to meet earnings forecasts. Hribar et al. (2006) and Almeida, Fos, and Kronlund (2016) show that firms that are just about to miss the consensus earnings-per-share (EPS) forecast have significantly higher repurchases than firms that just meet the EPS forecast without conducting repurchases. Almeida, Fos, and Kronlund (2016) also show a similar discontinuity in the growth of capital expenditures, employment, and R&D, which suggests that managers are willing to trade-off investments and employment for stock repurchases that allow them to meet analyst EPS forecasts. While this behavior is indicative of short-termism, it is not clear whether it leads to a deterioration of firms’ productivity in the long-run. On the contrary, one might expect firms to cut their less productive investments, which would not result in any adverse effect on the firm's overall productivity.

To shed further light on this question, it is crucial to look more closely into the characteristics of investments that get cut because of incentives to engage in EPS-driven repurchases. Are firms making cuts across the board, or are firms cutting less productive investments first? To answer this question, we use U.S. census data. These data allow us to examine changes in resource allocation and productivity at the plant level, and study how pre-existing plant characteristics are related to cuts due to these short-term incentives. Census data is crucial for our purposes, since firm-level data do not allow us to study within-firm changes in resource allocation.

Our identification strategy follows that in Almeida, Fos, and Kronlund (2016), which exploits a discontinuity in incentives to engage in repurchases when managers expect to just miss the analyst consensus EPS forecast. Consistent with that paper, we observe a significant decline in investment and employment expenditures in plants that belong to firms that have incentives to invest resources to boost short-term performance measures (those just to the left of the pre-repurchase zero EPS surprise).  Next, we focus on changes in total factor productivity (TFP), measured using the difference in TFP three years before to three years after the quarter in which we measure incentives to engage in EPS-driven repurchases. We find that firms with stronger incentives to engage in EPS-driven repurchases experience a significant deterioration in average productivity across its plants. In particular, firm-level TFP falls by about 1.4%. Thus, we find that stronger incentives to boost short-term performance lead to cuts in investments and employment as well as a drop in firm-level productivity.

We then investigate the reasons why EPS-driven repurchases may lead to a drop in firm-level TFP. First, we study whether cuts in investment and employment depend on the plant's productivity. It would be natural to expect that cuts in investment and employment are concentrated in relatively unproductive plants. However, we find evidence that firms make cuts across the board, irrespective of plant productivity. Specifically, we find that while cuts in employment are stronger for unproductive plants than for productive plants, firms appear to cut investment uniformly across productive and unproductive plants.

Importantly, we present evidence that this apparently inefficient reallocation of resources is at least partly a consequence of frictions in the labor market. Specifically, we investigate whether cuts in employment and investment are similar in states that have and have not adopted right-to-work (RTW) legislation. Labor unions may act as an impediment to doing relatively more efficient cuts, to the extent that labor rules constrain a firm's actions. Consistent with this hypothesis, we find that for plants located in RTW states (where the labor force is less organized), the only significant cuts are observed for employment in unproductive units, but we do not find significant cuts in investment across any type of plant, whether productive or unproductive, in these states. Moreover, we find no significant changes in employment in productive plants. These findings contrast with those for plants located in non-RTW states. Among the plants in these states with higher union power, firms make significant cuts in employment and investment not only for unproductive plants but also for productive plants. Thus, unionization of the labor force may prevent firms from doing efficient downsizing, which thereby enhances the negative long-term consequences of incentives to boost short-term performance.

Our findings suggest that plants located in non-RTW states experience less efficient cuts than plants located in RTW states, indicating that union power adversely affects the reallocation of resources. Do these differences lead to the negative changes in productivity? To examine this hypothesis, we investigate whether changes in firm-level productivity depend on the fraction of the firm's business located in non-RTW states. Our results do suggest that the reduction in TFP that is engendered by short-term incentives is concentrated in firms that have significant business in non-RTW states (strong labor unions). Firms with weaker labor unions undertake more efficient cuts, and thus experience no significant deterioration in firm productivity.

Overall, our evidence suggests that short term incentives lead to lower long-term productivity, but only if there are additional frictions that prevent firms from downsizing in an efficient manner. Firms that have incentives to engage in repurchases to meet EPS forecasts subsequently reduce investment and employment. If firms have most of their plants in states in which labor is weak, the bulk of the adjustment happens in unproductive plants, which minimizes the impact of the downsizing on productivity. But if their plants are located in states in which labor is strong, unionization of the labor force prevents firms from efficiently downsizing. These firms make cuts across the board, even in productive plants. As a result of this adjustment, productivity goes down in the long run.