Is There Discretion in Wage Setting? A Test Using Takeover Legislation

Marianne Bertrand, Sendhil Mullainathan1998 · DOI

Abstract

Anecdotal evidence suggests that uncontrolled managers let wages rise above competitive levels. To test this belief, we examine the wage impact of antitakeover legislation passed throughout the 1980s in many states. Since many view hostile takeovers as an important disciplining device, these laws, by reducing takeover threats, potentially raised managerial discretion. If uncontrolled managers pay higher wages, we expect wages to rise following these laws. Using firm-level data, we find that these laws raised annual wages by 1% to 2%, or about $500 per year. The findings are robust to a battery of specification checks and do not appear to be contaminated by the political economy of the laws or other sources of bias. These results challenge standard theories of wage determination that ignore the role of managerial preferences.

Introduction

In classical economic theory, managers are typically modeled as agents who perfectly maximize profits for shareholders. However, agency theory suggests that managers possess their own preferences—such as avoiding labor conflict, enjoying a "quiet life," or building a loyal "empire"—which may conflict with shareholder interests. A long-standing anecdotal belief is that "uncontrolled" managers allow wages to rise above competitive levels to simplify their own roles.

This paper provides empirical evidence for this phenomenon by examining how managers use increased discretion when shielded from the threat of hostile takeovers. By analyzing state-level anti-takeover legislation in the 1980s, the study finds that protecting managers from market discipline leads to a 1–2% increase in average annual wages. These results suggest that standard profit-maximization models may overlook a critical human element: managerial preference.

Methodology and Data

The authors utilize the staggered passage of state anti-takeover laws as a "natural experiment" that reduced external pressure on managers. Specifically, they focus on Business Combination (BC) laws, which imposed moratoriums on certain transactions between target firms and hostile raiders, effectively granting incumbent managers veto power over takeovers.

Empirical Strategy

The study employs a differences-in-differences (DiD) framework, comparing firms incorporated in states that passed BC laws (treatment) to those in states that did not (control). The regression model includes:

  • Firm Fixed Effects: To control for time-invariant differences between companies.
  • Year Fixed Effects: To account for macroeconomic trends.
  • Time-Varying Controls: Firm characteristics such as log assets, log employment, log sales, and log market value are included, with their impacts allowed to vary by year to ensure results are not driven by shocks to specific firm types.

Data Sources

The analysis relies on COMPUSTAT data from 1976 to 1995. Because COMPUSTAT does not report individual wages, the authors calculate an average wage by dividing total labor expenses by the number of employees. The final sample consists of 877 firms and 9,305 firm-year observations. Data cleaning involved excluding extreme outliers (wage growth >75% or <-44%) to prevent results from being skewed by data errors or major corporate restructurings.

Key Findings

The empirical analysis yields several significant insights into the relationship between managerial discretion and wage setting:

  1. Wage Increases Following Protection: The passage of BC laws led to a statistically significant 1% to 2% increase in annual wages, amounting to approximately $500 per worker in 1998 dollars.
  2. Gradual Implementation: The wage effect was not immediate. Most of the increase occurred three or more years after the legislation passed, suggesting a gradual shift in corporate policy rather than an abrupt change.
  3. The Role of Firm Size: The effect was more pronounced in larger firms. This aligns with agency theory, as larger firms often have more dispersed ownership, making the threat of a takeover a more vital disciplining tool than in smaller, closely-held firms.
  4. Robustness and Placebo Tests: The results remain robust when excluding Delaware (a major incorporation hub) and controlling for state-specific trends. Furthermore, "placebo tests" showed no significant wage trends in the years leading up to the legislation, supporting the causal link between the laws and wage increases.

Discussion and Implications

The findings support the existence of a "range of indeterminateness" in wage setting. Rather than wages being a single fixed point determined solely by market competition, they appear to exist within a range where the specific point is influenced by managerial discretion or "slack."

Why Do Managers Pay More?

The authors suggest several motivations for this behavior:

  • Labor Peace: Higher wages may reduce the effort required for union negotiations or conflict resolution.
  • Worker Quality: Managers might use the discretion to hire or retain higher-quality workers to improve operational smoothness.
  • Social Preferences: Managers may simply prefer to be well-liked by their employees.

Limitations

While the study provides strong evidence of managerial discretion, it faces certain data limitations. The aggregate nature of firm-level data prevents distinguishing between broad-based raises and those targeted at specific groups, such as middle management. Additionally, it remains an open question whether these wage increases represent "pure waste" (rents) or a strategic investment in worker quality that benefits the firm's long-term stability.

Conclusion

This research serves as a cornerstone in labor economics and corporate finance by providing a rare empirical bridge between theoretical models of management and actual behavior. It demonstrates that when the "disciplining device" of the market is weakened, managers exercise their discretion in ways that prioritize their own preferences over strict profit maximization.

Summary updated

Abstract

Anecdotal evidence suggests that uncontrolled managers let wages rise above com-petitive levels. To test this belief, we examine the wage impact of antitakeover legis-lation passed throughout the 1980s in many states. Since many view hostile takeovers as an important disciplining device, these laws, by reducing takeover threats, poten-tially raised managerial discretion. If uncontrolled managers pay higher wages, we expect wages to rise following these laws. Using firm-level data, we find that these laws raised annual wages by 1 % to 2%, or about $500 per year. The findings are robust to a battery of specification checks and do not appear to be contaminated by the political economy of the laws or other sources of bias. These results challenge standard theories of wage determination that ignore the role of managerial preferences. 1.
BibTeX
@misc{bertrand1998is,
  author = {Marianne Bertrand and Sendhil Mullainathan},
  title = {{Is There Discretion in Wage Setting? A Test Using Takeover Legislation}},
  year = {1998},
  doi = {10.3386/w6807},
}
Figure