Managerial flexibility, capacity investment, and inventory levels

Abstract

We study the effect of managerial time-horizon on two key operations decisions: inventory levels and capacity investment. On the one hand, we can expect operations decisions concerning capacity and inventory levels to be unaffected by any potential frictions between shareholders and managers, given that in theory they are both optimizable to meet the target service levels. On the other hand, expanding operations capacity requires significant financial resources, takes a long time to come to fruition, and faces demand uncertainty, hence investing in operations capacity may be deemed as prohibitively risky by short-term oriented executives facing shareholder pressure for higher earnings-per-share. For identification, we exploit a quasi-natural experiment (the staggered adoption of constituency statutes) that alleviated managerial short-termism by providing legal protection to executives adopting a long-term approach in their corporate decisions. Using a difference-in-differences design, we find that, after the reforms, inventory and capacity investment increased by 5.2% and 15.4%, respectively, for treated firms relative to control firms. We also find that these increases are gradual and persist over time, suggesting that they are structural in nature. We further show that the effect of the reforms on inventory levels and capacity investment are stronger for firms with ex-ante higher level of managerial short-termism, such as firms with low institutional ownership. Performance also increases relatively more for affected firms with higher ex-ante managerial short-termism. Our results pass a battery of robustness and validity tests. Interestingly, while constituency statutes are intended to protect executives from short-term oriented shareholder sanctions, our findings suggest that these statutes ultimately benefited not only executives, but also potentially the long-term interest of shareholders.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 15 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.