California's Unconstitutional Gender Quotas for Corporate Boards
California's new law requiring corporate boards to have a minimum number of women is both unconstitutional and likely to do more harm than good.
California recently passed Senate Bill 826, a law requiring all publicly held corporations based in the state to have a minimum number of women on their boards. A corporation with four or fewer directors must have at least one woman on its board. If the board has five members, at least two women must be on it. If it has six or more, there must be at least three women. In his official signing statement, California Governor Jerry Brown praised the law, but also noted that "serious legal concerns have been raised" and that "these potential flaws may… prove fatal to its ultimate implementation." Governor Brown is right to worry. The law is clearly unconstitutional under current Supreme Court precedent. If it survives the nearly inevitable legal challenges, it is also likely to cause more harm than good.
The Supreme Court has long held that laws that discriminate on the basis of sex are subject to heightened "intermediate scrutiny" under the Equal Protection Clause of the Fourteenth Amendment. This test requires all such laws to be "substantially related" to an "important" state interest. In other words, the law must serve an important objective and there has to be a close fit between the discriminatory policy and the interest it supposedly advances, so as to prevent the state from engaging in any more sex discrimination than is actually needed to achieve its "important" objective.
In United States v. Virginia (1996), the Court arguably tightened up the standard still further, emphasizing that sex-discriminatory laws must have an "exceedingly persuasive justification." It is important to emphasize that this kind of heightened scrutiny applies even if the law is well-intentioned and not motivated by invidious prejudice or by a desire to subordinate one gender to the other. Indeed, Indeed, Craig v. Boren (1976), the case where the Supreme Court first ruled that gender classifications are subject to heightened judicial scrutiny struck down an Oklahoma law that forbade 18-20 year old men, but not 18-20 year old women to buy 3.2% beer. No one could plausibly claim that the Oklahoma state legislature in the 1970s was some kind of matriarchy seeking to persecute men. But the law was invalidated anyway.
California's gender-quota for corporate boards clearly discriminates on the basis of sex, and it also clearly flunks intermediate scrutiny. The most obvious interest it might serve is overcoming the (very real) history of discrimination against women in the corporate world. That likely qualifies as an "important" state interest. But the fit between the end and the means is far too dubious to pass muster.
The law applies to all corporate boards of publicly traded firms, regardless of whether there is any significant recent history of discrimination at the firm, or in the industry in question. It also applies regardless of underlying distribution of interest in serving on boards in a given industry. It is hard to deny that some industries attract disproportionate interest from men, as opposed to women, and vice versa. Some of these differences in preferences are themselves the result of a history of sexism. But by no means all. For example, data suggest declining sexism may actually increase the proportion of women who avoid STEM fields in favor of pursuing other careers.
Even in the best-case scenario, the California law functions as a meat cleaver, rather than the scalpel demanded by heightened scrutiny. Imagine a population where ability and interest in serving on boards is equally distributed between men and women and all firms are completely nondiscriminatory (that is, they always hire the best-qualified applicants, regardless of gender). Even so, 18.75% of five-person boards will have one or zero women on them just by random chance, and thereby be in violation of the law.
A board with six or more members must have at least three women. Here too, even with a completely equal distribution of ability and interest combined with a complete lack of discrimination, many firms will run afoul of the law just by random chance. For example, 34% of six-person boards, 14.5% of eight-person boards, and 5.5% of ten-person boards will have two or fewer women, and thereby be in violation. A law that forces companies to engage in sex discrimination just to avoid the risk of liability caused by random chance variation is surely unconstitutional.
A narrowly-focused system of gender preferences to combat discrimination at firms with a demonstrated history of bias against women might potentially pass intermediate scrutiny. But Bill 826 goes far beyond that.
Other possible rationales for the law fare no better. For example, the law might be justified as an effort to promote useful "diversity" on corporate boards, that can improve the governance of firms. But, again, the quota rigidly applies to all publicly traded firms in all industries, regardless of the evidence on whether increased gender diversity would be useful there or not. Duke law professor Kimberly Krawiec, a leading academic expert on corporate boards, points out that evidence that board diversity improves governance is highly uncertain at best, and that such effects are not a solid justification for Bill 826.
California state courts have ruled that the state constitution applies "strict scrutiny" to gender classifications, an even tougher standard than federal intermediate scrutiny required by the federal Supreme Court's interpretation of the Equal Protection Clause. If strict scrutiny applies, Bill 826 is even more likely to be struck down.
Unconstitutional sex discrimination may not be the only legal problem with Bill 826. Corporate law scholar Steve Bainbridge argues that it also violates the Dormant Commerce Clause. Stanford law Professor Joseph Grundfest outlines a variety of possible legal challenges to Bill 826 and urges affirmative action advocates to pursue other strategies to achieve their aims.
Even aside from constitutional problems, Bill 826 is an ill-conceived policy. Because it applies a rigid quota system to all firms across the board, it increases the chance that some will have to hire relatively poorly qualified candidates to meet the law's standards. As noted above, such a situation can occur even in an industry where the available talent is equally distributed between men and women, and hiring decisions are 100% free of sexism. To the extent that hiring less-qualified board members leads to a deterioration in the firm's performance, stockholders, employees, and consumers may end up worse off. To the extent that many of these people are women, Bill 826 could actually harm women far more than it benefits them. There are far more female stockholders, consumers, and workers who depend on these firms, than female (or male) applicants for positions on corporate boards.
In previous articles, blog posts, and amicus briefs, I have spoken out against forms of state-imposed sex discrimination favored by the political right, such as laws banning same-sex marriage and regulations excluding women from draft registration and combat positions in the armed forces. Bill 826 is favored by the opposite side of the political spectrum. Yet all of these laws share a common weakness: they discriminate on a large scale without anything approaching an adequate justification. Despite considerable progress, more remains to be done to open up traditionally male-dominated fields to women, and to establish a society where opportunities are no longer constrained by restrictive sex-role expectations. But rigid government-imposed quotas are not the right way to combat sexism.
UPDATE: As several readers have pointed out to me, the original version of this post included a calculation error with respect to the percentage of ten-member boards that, statistically speaking, would be in violation of the law in a world where talent is equally distributed by gender and hiring is completely free of sex discrimination. I apologize for this error (which I have now corrected) and thank those who caught it.