Risk Management Transparency and Compensation

Chang-Mo Kang & Donghyun Kim (김동현) — Journal of Corporate Finance, Vol. 75 (2022) 102245. doi:10.1016/j.jcorpfin.2022.102245

Research Question

When a manager must simultaneously create firm value and manage financial risk, how should shareholders design compensation contracts — and does their ability to monitor risk management practices change the answer? This paper addresses a fundamental tension in executive pay design: high-powered incentive compensation (e.g., stock options) induces productive effort, but it also provides incentives for managers to engage in risk-seeking behavior that inflates expected pay without enhancing firm value.

The central question is whether shareholders' ability to observe risk management choices resolves this tension. If shareholders can detect speculative or careless risk management — for instance, because accounting standards require transparent disclosure of derivative activities — they can condition compensation on that information, awarding more convex pay without fear that managers will exploit it to pursue self-interested speculation.

Theoretical Framework

The paper develops a multitask principal-agent model in which a risk-neutral manager with limited wealth chooses both a productive effort level and a risk management action. The risk management choice affects the firm's cash flow risk without altering its expected value. High-powered convex incentive pay increases productive effort but simultaneously tempts managers to take excessive risk — which increases the probability of extreme cash flow outcomes and raises expected managerial compensation without benefiting shareholders.

The key theoretical result: when shareholders can monitor risk management choices, they can induce both high effort and low risk-taking by offering convex incentive contracts. When they cannot monitor risk management, they must restrain the use of convex compensation to prevent risk-seeking. Shareholders' monitoring ability therefore directly enables higher-powered incentive pay. This prediction is novel relative to standard principal-agent models, where risk-incentive relationships are determined by exogenous firm risks rather than endogenous monitoring capacity.

Empirical Strategy

The paper tests this prediction using the issuance of Statement of Financial Accounting Standards No. 133 (Accounting for Derivative Instruments and Hedging Activities, FAS 133) as a natural experiment. Before FAS 133, firms could classify derivative holdings as either hedging or investment instruments at their discretion, using different accounting treatments and effectively hiding speculative derivative activity. FAS 133, issued in 1998 and effective by fiscal year 2001, mandated fair value accounting for all derivative holdings, eliminated discretionary classification, and required stringent documentation of hedging relationships — substantially improving shareholders' ability to observe the corporate use of derivatives.

The empirical design is a difference-in-differences (DID) model comparing compensation at firms that used derivatives before FAS 133 (users, n = 452) versus firms that did not (non-users, n = 360), before (1996–1997) and after (1999–2001) the standard's adoption. The dependent variable is Vega — the change in the fair value of newly granted options for a 1% increase in stock price volatility — which directly measures pay convexity. Data on executive compensation, job titles, and derivative holdings are drawn from Execucomp, Compustat, and 10-K filings.

Key Results

Main Effect: Pay Convexity of Financial Officers

After the issuance of FAS 133, derivative-using firms significantly increased the convexity of financial officer compensation relative to non-using firms. The DID coefficient on FAS133 × User is 4.204 (t = 2.92, significant at 1%), indicating that after FAS 133, the difference in Vega between users and non-users increased by approximately $4,457 — equal to 147% of the pre-FAS 133 gap of $3,032. The effect operates through option grants, not stock grants: option award values increase by approximately $0.4 million for user firms after FAS 133, while stock grant values do not differ significantly between groups (Table 3). This is consistent with options providing meaningful risk-taking incentives in a way that stock awards do not.

Falsification Tests: Only CFOs with Finance Expertise

A critical validation is that FAS 133's effect is confined to executives who are actually involved in derivatives decision-making. Among firms managed by financially expert CEOs (those with prior finance experience), the compensation Vega of these CEOs increases significantly after FAS 133 relative to non-users (Table 6, Column 1: coefficient = 24.50*, t = 1.75). By contrast, among firms with non-financial-expert CEOs, no significant effect appears. More importantly, FAS 133 has no significant effect on other executives, technical officers, or non-employee directors — groups not involved in derivatives program oversight. This pattern rules out the explanation that unobservable firm-level time trends drive the results.

Moderating Factors: Governance and Internal Power

The effect of FAS 133 is substantially stronger in governance environments where the monitoring problem is most acute before the new standard (Table 8):

• Board structure: The DID coefficient is 6.015*** in firms with classified (staggered) boards — which entrench management — versus an insignificant 1.385 in non-classified firms.
• Board co-optation: The effect is 6.037** in firms with high fractions of co-opted independent directors and insignificant in firms with low co-optation.
• Board attentiveness: Firms with inattentive directors (attending fewer than 75% of meetings) show a coefficient of 7.491**, versus an insignificant −0.825 in well-attended boards.
• Institutional monitoring: The effect is 8.067** in firms where institutional ownership concentration is below the median, versus an insignificant 1.759 above it.
• Financial officer influence: The effect is 7.875** in firms where financial officer pay exceeds 50% of CEO pay, and 7.234** in firms with long-tenured financial officers — suggesting that influential financial officers benefit most from the new disclosure regime.

Robustness

The core findings survive a battery of robustness checks: controlling for lagged Vega Holdings (to account for option holdings changes due to exercises or expirations), using kernel-weighted propensity score matching, excluding high-tech firms to address the dot-com bubble, and a parallel test around the issuance of FAS 161 (a 2008 amendment to FAS 133 that further enhanced disclosure granularity). Column 4 of Table 9 shows that FAS 161 increased pay convexity in firms with higher pre-reform unrealized cash flow hedge losses — precisely where the additional transparency was most informative. Pay-for-performance sensitivity (Delta) also increases after FAS 133 (by ~$5,247, Table 10), though the magnitude is smaller than the convexity effect.

Implications for Institutional Investors

This paper carries direct and practical implications for stewardship teams, proxy advisors, and ESG governance analysts.

• Disclosure quality as a governance prerequisite. The paper demonstrates that transparent risk management disclosure is not merely an information concern — it is a structural precondition for effective incentive design. Where derivatives accounting is opaque or standardization is lacking, shareholders are rationally constrained in their ability to offer performance-linked pay, regardless of their preferences. Engagement on disclosure quality (e.g., IFRS 9 / IFRS 7 compliance, ASC 815 hedge documentation) has direct downstream effects on compensation design.
• CFO compensation as a governance signal. The finding that FAS 133 selectively increased financial officer option grants — not those of other executives or directors — suggests that option-heavy CFO pay in derivative-active firms can function as a governance-enabling signal, rather than a rent-extraction mechanism. Voting policies that reflexively flag high CFO option intensity should consider the derivatives context of the firm.
• Board quality amplifies disclosure benefits. The heterogeneity results show that enhanced transparency is most effective in firms with weak board monitoring. Stewardship teams should therefore evaluate disclosure improvements in conjunction with board quality: a firm with stronger hedging disclosures but a co-opted or classified board is still more governance-constrained than a well-governed peer with equivalent disclosures.
• Engagement on derivatives oversight. The paper's subsample results — showing that FAS 133's effect is strongest in firms operating in multiple international regions and using currency derivatives — point to geographic complexity as a key risk management monitoring challenge. Investors should inquire specifically about board-level oversight of multi-regional derivatives programs, particularly in portfolio companies with significant FX or interest rate exposure.

Selected References

1. Gormley, T.A., Matsa, D.A., Milbourn, T. (2013). CEO compensation and corporate risk: Evidence from a natural experiment. Journal of Accounting and Economics, 56, 79–101.
2. Goldman, E., Slezak, S. (2006). An equilibrium model of incentive contracts in the presence of information manipulation. Journal of Financial Economics, 80, 603–626.
3. Zhang, H. (2009). Effect of derivative accounting rules on corporate risk-management behavior. Journal of Accounting and Economics, 47, 244–264.
4. Diamond, P.M. (1998). Managerial incentives: On the near linearity of optimal compensation. Journal of Political Economy, 106, 931–957.
5. Campbell, J., Khan, U., Pierce, S. (2021). The effect of mandatory disclosure on market inefficiencies: Evidence from FASB Statement No. 161. Accounting Review, 96, 153–176.