Should bonuses paid to CEOs and CFOs be linked to qualitative performance criteria such as displaying strong leadership, mentoring other executives, and having a strategic vision?
The answer is no, according to a recent study I conducted with my colleague Nancy Feng at Suffolk University. Our analysis of data for S&P 500 companies reveals that employee productivity, asset productivity, capital expenditures, and future stock returns are lower when the CEO’s bonus payment depends in part on qualitative criteria.
While our sample includes CEOs, we expect linkages in the case of CFOs, who report lower satisfaction with their compensation contracts when their bonuses are awarded subjectively.
To be clear, we do not argue that skills like leadership, mentoring, and strategic thinking are unimportant. Rather, we maintain that basing the CEO’s cash incentive payment on subjective, unstructured goals is not an effective strategy to enhance firm performance.
In 2017, compensation committees awarded large bonuses to the CEOs of numerous large U.S. corporations. Among high-profile companies that paid bonuses exceeding $10 million were CBS, Mondelez International, Morgan Stanley PepsiCo, and Walt Disney.
Despite the importance of performance payments in executive compensation, there is limited guidance on best practices for performance criteria selection in compensation contracts. While a majority of companies rely solely on quantitative performance metrics (e.g., operating income, earnings per share, stock returns, customer satisfaction) in determining CEOs bonuses, one in four S&P 500 companies uses a mix of quantitative and qualitative criteria.
Extant research on agency theory suggests that the optimal bonus compensation contract should include a subjective performance assessment. That’s because quantitative performance metrics can be influenced by managerial discretion and uncontrollable external factors.
In line with this observation, Robyn Dawes — a well-cited mathematical psychologist — had long argued that qualitative assessments in decision-making were irrelevant because one could make a reasonable judgment based on only quantitative criteria. His thesis has been proven to work in different settings, such as graduate school admissions and financial analysis.
In the context of CEO bonus contracts and performance evaluation, we identified three significant shortcomings of qualitative assessments.
First, qualitative performance criteria are ill-defined, providing the CEO with limited guidance regarding what exactly needs to be achieved going forward. For instance, the 2017 annual bonus of David Zaslav, CEO of Discovery, was based 50 % on the accomplishment of six qualitative goals.
Those included “further develop and integrate [the company’s] overall digital and ‘over the top’ strategy” and “develop robust succession plans for key operational roles, while continuing to attract, retain, mentor, and reward exceptional talent.”
Although those qualitative goals relate to important aspects of business (e.g., developing human capital), they involve a great deal of ambiguity.
Recent experimental evidence shows that managers perceive quantitative performance measures to be more scientific than qualitative measures.
Relatedly, a compensation consultant wrote about a board-level conversation she had on the need for clarity in executive rewards. On the one hand, a chief executive expressed a preference for a more structured and clearly communicated compensation package. On the other hand, the board chair said it was an executive’s role to manage ambiguity.
Our view is that CEOs are typically trained to manage ambiguity pertaining to operational decisions, not their compensation packages. Setting structured goals for the CEO or CFO is key to a company’s success.
Second, qualitative criteria and related performance targets are likely to be selected arbitrarily. No systematic supporting evidence can be collected and presented by compensation consultants and committees because of the idiosyncratic nature of the criteria under examination.
In contrast, relevant data can be collected and tested regarding various quantitative criteria used in CEO bonus contracts.
Third, whereas assessments based on the same quantitative data always lead to the same outcome, this does not hold true in the case of qualitative assessments. The subjective nature of such assessments may induce bias into performance evaluations, possibly reducing the effectiveness of the CEO’s incentive plan in driving better firm performance.
To test our prediction that firm performance is negatively associated with the use of qualitative performance metrics in CEO bonus contracts, we hand-collected data on CEO bonus payment criteria from the proxy statements of S&P 500 firms (excluding financial firms and utilities) for fiscal years 2007, 2010, and 2013.
We classified all criteria other than objectively quantifiable metrics (e.g., earnings per share) as qualitative criteria. This category includes such criteria as leadership development, succession planning, and strategic business reviews. Twenty-five percent of the sample firms included at least one qualitative criterion in their CEO bonus contracts.
After controlling for a set of CEO, firm, and industry characteristics, we found quite noticeable differences in the level of productivity between firms that base CEO bonus payments on only quantitative criteria and those that rely on both quantitative and qualitative criteria.
Firms in the latter category exhibited, on average, 18.5% and 10.1% lower employee productivity and asset productivity, respectively, than their counterparts in the former category.
Specifically, in terms of asset productivity losses, the results imply that for every $10 million invested in total assets, $1 million less revenue is generated when a qualitative goal is present in the CEO’s bonus contract.
Turning to investments in corporate infrastructure, we observed a similar pattern. Capital expenditures were lower by 12.2% among firms that linked CEO bonus payments to qualitative criteria than those that do not. Research shows that firms with low capital investment tend to underperform their peers.
Accordingly, we find that when CEO bonus payment is contingent on a qualitative criterion, firms earn six percentage points lower two-year-ahead stock returns (after adjusting for size and value risk).
Ancillary tests show that the documented negative implications of the use of qualitative criteria in the CEO’s bonus contract are more pronounced when bonus payments account for a large portion of the CEO’s total pay.
What do these findings mean for compensation committees? When determining the CEO’s or CFO’s bonus payment, it suffices to use quantitative performance criteria. These include a wide range of possible financial metrics such as return on invested capital, return on assets, and free cash flows, as well as nonfinancial metrics including customer satisfaction ratings, user growth, and employee turnover.
A subjective assessment based on qualitative performance criteria is unlikely to improve the quality of performance evaluations. Also, there are unintended consequences of adding qualitative criteria into a CEO’s bonus contract, namely reduced productivity and lower firm value.
An unwarranted conclusion from the present discussion would be that qualitative aspects of the CEO’s job are not relevant for firm success. Shareholders and the board of course expect that the CEO be an effective leader, mentor, and strategic thinker.
Nonetheless, compensation for such skills should not be a part of the CEO’s incentive payment package. It is unlikely that an effective leadership can be established through contractual arrangements. A better practice would be paying a higher salary or signing bonus to attract CEOs (and other executives) with desired skills and reputation.
Prior to the 2017 tax reform, performance-based executive compensation was tax deductible. However, to qualify for a deduction, firms had to specify objective performance criteria for payouts
With the elimination of the favorable tax treatment of performance-based pay vis-à-vis non-performance pay (beyond $1 million), some firms are deciding to increase the proportion of salary and other fixed payments in CEO compensation packages.
Another possible implication is that the use of qualitative performance criteria in CEO bonus contracts may become a more common practice (as there is no longer a binding tax deductibility incentive). We advise against doing so.
Ahmet Kurt is an assistant professor of accounting at Suffolk University’s Sawyer Business School in Boston.