Mingze Gao, Henry Leung, Buhui Qiu
6 min
This paper investigates the relationship between a firm's organization capital (OC)—the proprietary knowledge, systems, and processes that enhance productivity—and the performance-based incentives provided to non-CEO executives. The authors hypothesize that because OC enhances firm productivity, it reduces the marginal necessity for managerial effort, thereby allowing firms to reduce costly executive pay-for-performance sensitivity (PPS). To test this, the authors analyze a large U.S. sample from 1992 to 2015, using capitalized selling, general, and administrative (SG&A) expenses as a proxy for OC.
The empirical analysis reveals a robust, negative association between a firm's OC and the PPS of its non-CEO executives. A one-standard-deviation increase in OC is linked to a significant reduction in executive delta (the dollar change in wealth for a 1% change in firm value). To establish causality, the authors employ two identification strategies: an instrumental-variable approach using state-level unemployment insurance benefits, and a quasi-natural experiment examining exogenous CEO turnovers due to health-related issues. Both methods confirm that when a firm experiences a negative shock to its OC, it subsequently increases PPS for its non-CEO executives, supporting the substitution effect hypothesis.
This research contributes to the literature on corporate governance and executive compensation by identifying OC as a critical, previously overlooked determinant of pay design. It suggests that corporate boards do not set executive incentives in a vacuum; rather, they account for the firm's internal organizational efficiency. By demonstrating that OC can substitute for high-powered incentives, the paper provides a new perspective on why some firms may rely less on equity-based pay, potentially mitigating issues like managerial short-termism or earnings management associated with high-powered incentives.
Sam: They did. The sample is restricted to executives who remained in place throughout the event window with unchanged seniority. The effect persists even in that constrained sample—so it's not a story about promotions or organizational restructuring. The board is directly responding to the loss of the organizational autopilot. [[RP_SECTION:measurement-and-proxy-robustness|Measurement and Proxy Robustness]]
Alex: That's a meaningful identification strategy. What about the measurement side? Capitalized SG&A is the standard proxy for organizational capital, but it's a noisy one.
Sam: It is, and a careful referee would push on exactly that. SG&A captures a wide range of spending—advertising, administrative overhead, training—not all of which maps cleanly onto organizational capital. Measurement error here could attenuate the estimated substitution effect, which would actually bias against finding the result. The authors use industry-adjusted measures and run specification curve analyses across alternative constructions of the proxy. The substitution effect survives those checks consistently, which gives you some confidence that the signal is real even if the proxy is imperfect. [[RP_SECTION:shareholder-wealth-and-misalignment|Shareholder Wealth and Misalignment]]
Alex: So the noise is a conservative problem rather than an inflationary one. What about the shareholder wealth side—does the misalignment between incentive structure and organizational capital actually show up in returns?
Sam: That's where the paper moves from mechanism to consequence. The interaction between organizational capital and incentive pay predicts firm returns. Firms that maintain high-powered incentives despite having strong organizational capital appear to be over-paying for performance they don't need to buy. The implication is that misalignment isn't just theoretically inefficient—it has a measurable cost to shareholders.
Alex: Which reframes what a board is actually doing when it sets executive compensation. It's not just responding to the labor market for talent—it's making a structural decision about where performance comes from.
Sam: That's the deeper point. The conventional view treats compensation as a tool to drive effort. This research suggests boards are simultaneously managing internal architecture. The choice is between buying performance through risk-laden incentive contracts or building it into the firm's operating systems. Those are substitutes, and the optimal mix depends on how much organizational capital the firm has already accumulated. [[RP_SECTION:future-implications-of-automation|Future Implications of Automation]]
Alex: It also raises a question about where this goes as firms accumulate more algorithmic infrastructure. If AI systems take over the complex, judgment-intensive tasks that organizational capital currently handles, does the logic extend—executive incentives compress further as the marginal product of executive effort falls again?
Sam: That's a reasonable extrapolation from the mechanism, though the paper doesn't test it. The framework would predict exactly that: as automated systems absorb more of the value-generating work, the case for high-powered executive incentives weakens further. Whether current compensation models are equipped to track that shift is an open empirical question.
Alex: And one that probably needs better proxies than capitalized SG&A to answer.
Sam: Agreed. That's the binding constraint on the whole literature—the intangible capital measures are improving, but they're still catching up to the complexity of what firms actually build. For now, the core result stands: organizational capital and incentive intensity are substitutes, the relationship appears causal, and boards that ignore that tradeoff may be leaving money on the table. Thanks for listening to ResearchPod.