Research from the Federal Reserve Bank of New York finds evidence that women’s pay in banks is not just lower than that of men but the structure is different, in a way that best be explained by the view that boards decide pay, not based on economic efficiency, but on the basis of power, which is stronger for men, who have closer networks of influence than women bankers.
I was briefly on the European diversity committee at JP Morgan in London. I can personally testify that the bank takes diversity seriously and that executives who failed to promote and support women could find their careers damaged as a result. I believe other banks are similarly motivated in recruitment and retention. (My younger daughter recently attended a Goldman Sachs day for high school girls which left her impressed – Goldman, whatever you think of them, really wants to hire the best people whatever their gender, race or sexual orientation). But the biggest gap in female representation is at the top of banks (as it is in non-financial businesses – see this recent self-critical article from McKinsey). And the lack of powerful women in senior positions is a major reason for continuing pay inequality according to recent research from the New York branch of the Federal Reserve, the US central bank.
Three key findings
The NY Fed found three key facts about the structure of executive pay in banks. First, female executives’ pay packages typically have a much lower level of incentive compensation, which means payments such as bonuses and stock options, rather than salary. This difference results in 93% of the overall difference in pay between men and women.
Second, women’s pay is typically far less sensitive to changes in firm value than for men. A $1 million change in firm value on average results in a change in male compensation of $17,150 for men but only $1,670 for women.
Before examining the third fact, let’s consider why women’s pay might have a different structure. If women are on average more risk-averse than men then they might prefer a higher proportion of fixed pay and less variable or incentive-based pay. They would expect to get the same overall package, on average. They might also prefer more predictable pay because of being constrained by having to take a larger share of household responsibilities including childcare. Perhaps also they are less inclined to compete and to want to negotiate. These could be factors imposed on women involuntarily by society or in some way inherent or both.
Note that if these factors exist (whatever the justice of them) then conventional economic theory predicts that boards would rationally give women less incentive-based pay relative to men, reflecting the women’s preference for less risk. In a risky environment, that would mean the women accepting a lower average pay package in return for more certainty, because the women would be implicitly paying a risk premium (equivalent to insurance) to avoid unwanted volatility. Men, if they’re happy to accept the volatility, would get paid more.
None of this justifies the lower pay of women but it offers a market-based explanation of lower average pay, if we accept that women are on average more risk averse.
But the third fact is more remarkable. The NY Fed also finds that on average female executives’ pay is more exposed to bad firm performance and less exposed to good performance than for males. “A 1% increase in firm value generates a 13% rise in firm specific wealth for female executives, and a 44% rise for male executives, while a 1% decline in firm value generates a 63% decline in firm specific wealth for female executives and only a 33% decline for male executives.” The researchers also show that there is “no link between firm performance and the gender of top executives.”
Can an efficient market view of pay determination explain this last fact? No. Instead, the researchers invoke a competing theory of how pay is determined, the “managerial power” or “skimming” view. This means that boards are not independent but are chosen or captured by chief executives in a political process. Chief executives manage the boards to get both the highest compensation and to protect themselves as much as possible from falling pay owing to bad firm performance.
Executives skim, and men are better positioned to do it
The skimming view is consistent with the data because “female executives, who are younger, have lower tenure and are limited in accessing informal networks, are likely to be less entrenched than their male counterparts.” In other words they have less power to influence or capture boards in their favour. (Note that this assumes that female executives would behave in exactly the same way as men if they could, which might not be true.)
In sum, the “gender differences in the structure and level of compensation are not efficient.” You need to bear in mind that this research is from staffers at the US central bank’s main operational arm. As economists, their starting point on most question will be, is it efficient? They find in this case that bank executive compensation is not. That’s a strong finding, before we then consider questions of justice or fairness. The system fails a basic economic test, quite apart from wider moral ones.