The gender pay gap is a well-documented phenomenon in global labor markets, but this gap does not seem to apply to the pay of CEOs. However, upon further investigation, Massimiliano Tani, Keiran Sharpe, and Andrew Valentine find in new research that, compared to men CEOs, a higher share of the total compensation of women CEOs in English-speaking countries is tied to performance rather than fixed emoluments.

There is an intriguing anomaly in the pay of chief executive officers when analysed by gender: against overwhelming evidence that women are paid less than comparable men across the labor market, female CEOs do not seem to experience such penalty. How are we to explain this singular phenomenon? 

One possibility is that the CEO labor market is simply an exception relative to the rest of the labor market. This is an apparently plausible explanation, which has some support from empirical data. Specifically, evidence shows that the overall compensation payments made to female CEOs are equal to—or are even marginally greater than—the overall compensation payments made to their male peers. Case closed, one might think. It turns out, however, that matters are not as straightforward as a prima facie reading of the evidence might suggest.

To investigate this issue further, we looked at over 1,000 companies located across the globe. Most of the sample consisted of large as well as small- and medium-size enterprises operating in the consumer discretionary, financials, health care, industrials, and information technology sectors. In our initial investigation of the data, we found that it conformed with the consensus in the literature. This is to say, we found that newly appointed female CEOs are paid the same or marginally more than male CEOs, the ratio of fixed (non-performance) emoluments to overall compensation is the same between genders, and board diversity has no impact on either overall compensation or on the ratio of fixed-to-total compensation.

However, on further investigation based on better-matched comparison groups, we found an interesting wrinkle in the compensation payments by gender in the English-speaking countries—namely, Australia, Canada, New Zealand, the United Kingdom, and the United States. Notwithstanding that the value of the average compensation package is broadly equal for male and female CEOs, the ratio of fixed compensation to overall compensation is not. A larger percentage of male CEOs’ compensation packages is paid as fixed emoluments. Or, to put the same proposition another way: performance payments constitute a larger percentage of female CEOs’ compensation packages. Thus, in the Anglo-Saxon world, while female CEOs receive the same overall payments as their male counterparts, they must demonstrate their productivity to earn it. We also found that this result is marginally mitigated by board diversity—the greater the female representation on the board, the less is the “performance penalty” (i.e., the share of performance-based pay) applied to new female CEOs.

In contrast, we did not find any gender differences in total pay or the composition of CEO compensation in 11 other countries in Europe or Asia and other developing markets. Nor did we find that greater board diversity played a significant role affecting CEO compensation or the likelihood of appointing a woman CEO in these environments.

We suggest that differences in bargaining structures account for the phenomenon whereby male and female prospective CEOs are paid similarly in non-Anglo-Saxon countries but are paid differentially in the English-speaking world. The evidence suggests that the scope for bargaining in English-speaking countries is less constrained than it is in Asia and Europe, where social norms may put stronger bounds around any bilateral pay negotiations. Thus, the institutional constraints on bargaining over CEO pay combined with social norms that require equal pay regardless of gender imply that there will be no gender pay gap for CEOs, which is what we observe in non-Anglo Saxon countries.

In the English-speaking world, two salient factors seem to be in play. First, the scope for bilateral bargaining over CEO packages is greater than it is elsewhere. Secondly, women and men prospective CEOs negotiate with company boards for their pay package from different bargaining positions. These differences likely affect both the overall level and composition of compensation. However, the visibility of the CEO role, strong corporate governance practice, and the transparency of paid compensation in company annual reports for public companies may constrain boards and investors to pay particular attention to appear as not discriminating against women CEOs. As a result, boards may be open to offering similar total compensation, but then bargain resolutely on its cash vs. equity composition, especially when there is a lack of women directors. 

We formalise these hypotheses in a theoretical model of bilateral bargaining. On one side of the bargaining table is a profit-maximising firm trying to secure the services of a competent CEO by varying the proportion of cash-bonus offered: naturally, the greater is the proportion of a pay package paid as bonuses, the more aligned are the CEO’s and the company’s interests. On the other side of the table is a CEO candidate who maximises utility by choosing his/her effort for the package offered. Being risk averse and preferring leisure over work, the CEO would prefer to take a larger proportion of their pay as cash salary. The CEO’s gender affects the reservation wage, which is lower for women, as repeatedly found by a large literature based on secondary data and experiments in both the laboratory and in the field.

The bargaining problem confronting both parties is about how much of the overall compensation package should be paid as cash salary and how much should be paid as share bonuses. This is defined by two boundaries. The first sets the minimal acceptable position for the company. At that point, the proportion of the CEO’s income paid as share bonuses is at its lowest, the CEO’s effort is at its lowest, utility is at its maximum, and net profits is at the minimum level acceptable to the firm (the firm walks away from negotiations at lower levels of profits). The second boundary sets the minimal acceptable position for the CEO. At that point, the proportion of the CEO’s income paid as share bonuses is at its greatest, the CEO’s effort is at its greatest, net profits are at their maximum, and the CEO’s utility is at the minimum acceptable level (the CEO walks away from negotiations at lower levels of utility). 

As each party approaches their boundary of acceptability, their resistance to further movements to those boundaries increases. Conversely, as each party moves further from their own acceptable limit, their bargaining forcefulness is diminished. A bargaining equilibrium is attained when the bargaining forcefulness of the two parties are equalized. The diagram depicts the situation.

Two features of the model help explain the observations in the data. First, the better the new CEO’s outside options, the better is their starting point for negotiations and the better the final outcome is for them, all else being equal. Because women non-CEOs (which is the major pool from which new CEOs are drawn) earn less than their male counterparts, their bargaining position as newly minted CEOs is inherently weaker. This then leads to weaker bargaining outcomes for newly appointed women CEOs, which itself then weakens the initial bargaining position of other women CEOs down the track. 

Secondly, the model proposes that male-dominated boards are more forceful in resisting female interlocutors than they are in resisting male interlocutors, whilst more diverse boards are more even handed in their bargaining. This latter postulate helps explain why women CEOs do better with diverse boards than otherwise.

The compositional differences in CEO compensation between men and women is a unique characteristic of the Anglo-Saxon labor market. Although total compensation is equal for both men and women CEOs in these countries, the larger share of performance-based incentives in women CEO compensation packages reflects, to a degree, gender disparities within the broader labor market.

Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.