There is Value in Collaboration: Rethinking Pay-for-Performance

The following puzzle has bothered economists for decades; one of the most compelling ideas in compensation theory never quite made it into real-world pay contracts: relative performance evaluation, or RPE for short. The logic is elegant. Executives face risks they cannot control: a bad economy, an industry-wide downturn, or an oil shock. None of that should count when analyzing their effort as these are not their doing. Why not filter it out? Just pay your executives based on their performance relative to their competitors, not on whether the whole industry had a good or bad year. 

Holmström figured this out in 1982. It’s theoretically compelling, and yet decades of empirical research show that RPE is surprisingly rare in practice. Researchers have proposed various explanations: executives hedging their own wealth, boards captured by rent-seeking managers, uncertainty aversion that leads you to “pay for luck”. The puzzle, however, persists because we have been asking the wrong question. What if RPE is not failing but is simply inappropriate in many real-world settings?

The answer, we argue, lies in an overlooked but central feature of the modern economy: industry-wide value creation.

The Governance Unit Is Not Always the Firm

We tend to think of corporate governance as a firm-level problem: one principal, one agent, one board. But important value creation in the modern economy does not happen inside firms, but between them. Firms routinely invest in activities that benefit entire industries. Think, for example, of trade associations. The American Beverage Association lobbies for recycling legislation that benefits the entire sector; the New York Automobile Dealers Association produces the New York Auto Show, helping members coordinate EV infrastructure investments; the American Concrete Pavement Association runs campaigns that expand demand industry-wide. There are countless examples like those, involving standard-setting, regulatory engagement, shared R&D, and joint investments in talent and processes.

These are not activities that benefit one firm at the expense of another. They're genuinely collaborative. But collaboration requires effort—someone has to attend the meetings, build relationships, share information, and push the joint agenda forward—and this is exactly where the problem arises. 

If you pay executives under a classical RPE contract—where they are penalized when their peers perform well—you are effectively taxing collaboration. You are paying your CEO to avoid helping the very firms she needs to work with. Why invest in industry-level cooperation if success will be used against you? When firms care about collaboration, the logic of optimal incentives is different.

What the Optimal Contract Actually Looks Like

We formalize this intuition in a simple model building on classic incentive theory. The result is simple: when firms can collaborate to create industry-wide value, the optimal contract no longer penalizes executives when their collaborators perform well. In fact, it does the opposite—it rewards them. 

The “collaboration bonus” flips the sign of the RPE term to positive. Yet RPE does not disappear entirely. It remains optimal when applied to firms outside the collaborative network. The model is surgical: filter out the noise from non-collaborators, but reward performance among those you actually work with.

The Evidence Is Direct

We test these predictions using a large panel of executive-firm-year observations from 1999 to 2023, combined with a novel dataset of firm-level trade association memberships constructed from historical web archives. The results line up directly to the model’s predictions. 

Each additional trade association membership is associated with a 17% decline in RPE usage. But the effect is entirely concentrated where it should be: among trade association peers. RPE against non-members? Completely unchanged.

The same pattern appears when looking at realized pay. Executives are paid less when non-member peers perform well—the classic RPE effect—but more when trade association peers do well. The sign flips exactly where the model predicts. Firms are not abandoning RPE; they are applying it selectively.

How do firms implement this in practice? Largely by shifting toward compensation components that avoid explicit peer-based metrics—more discretionary bonuses, stock options, and restricted stock, and less reliance on formulaic performance measures. This also helps explain why we rarely observe explicit “collaboration bonuses”: tying CEO pay directly to competitors’ performance would raise red flags with regulators and proxy advisors as they resemble anticompetitive practices. Instead, firms deliver these incentives more quietly through compensation structures that preserve board discretion.

We also provide evidence that these incentives have real effects. The 2017 Tax Cuts and Jobs Act reduced the tax advantage of performance-based pay, leading many firms to scale it back. For firms without trade association ties, this primarily weakened firm-level performance, consistent with standard theory. But for firms participating in trade associations, it also reduced industry-level performance. Incentives do not just shape firm outcomes—they shape collaborative outcomes when firms are aligned to create industry-wide value.

A New Explanation for an Old Puzzle

The RPE puzzle looks very different once we recognize that some firms are playing a fundamentally different game. In many settings, the relevant governance unit is not the firm, but the industry. When value is created collectively, incentives must be designed collectively to reward industry success.

RPE did not fail. We have simply been applying it in the wrong world.

____________

Felipe Cabezón is an Assistant Professor of Finance at Virginia Tech Pamplin College of Business.

Gerard Hoberg is the Charles E. Cook Community Bank Professor of Finance at the USC Marshall School of Business.

Ekaterina Neretin is an Assistant Professor of Accounting at Bocconi University

This blog is based on a paper presented at the 2026 Corporate Governance Symposium and John L. Weinberg/IRRCi Research Paper Award Competition on 6th March 2026. Visit the event page to explore more conference-related blogs.

The ECGI does not, consistent with its constitutional purpose, have a view or opinion. If you wish to respond to this article, you can submit a blog article or 'letter to the editor' by clicking here.