Gender diverse boards are less likely to use share buybacks to buy undervalued shares and achieve excess returns, than male-dominated ones, which may have better access to information networks.
The debate on gender diversity on boards has reached the highest echelons of business and policymaking. Many academics have also sought to investigate whether hiring more female board members increases shareholder value. But is diversity a path to value creation? If not, should metrics like gender diversity be dismissed?
A key problem with testing such hypotheses is always endogeneity: the fact that for example companies with a lot of female board representation have higher profitability could simply mean that more profitable firms hire more women. Endogeneity can lead to misleading conclusions hence also wrong decisions for such important matters.
One way to get around this key problem is to focus on exogenous shocks, such as the imposition of mandatory gender quotas in Norway. K. Ahern and A. Dittmar find that this regulation had a negative effect on shareholder value. Of course, mandatory quotas are not representative of voluntary decisions to have more women on the board.
A second approach is to focus on consequences of various corporate decisions and test whether board composition makes a difference. Two studies that focus on mergers and acquisitions find that stock prices increase more (or fall less) when there are more female executives on the board. The explanation is that 1) women are more risk averse than men, and 2) they are less overconfident than men. Hence as women are more careful with shareholders’ money they make better decisions. The fact that women are more risk averse than men has been used by Michel Barnier, the former EU Commissioner for Internal Market and Services, to justify mandatory quotas on bank boards. The argument is that the financial crisis could have been avoided if there had been more women on management boards.
Gender diversity and buybacks
In our recent paper, we test whether this conclusion also holds when companies announce share buybacks. The answer is no. First, ceteris paribus, companies are more, not less likely to announce share buybacks in high gender diversity firms. As buybacks increase leverage and therefore risk, this is inconsistent with the hypothesis that women are more risk averse. Second, in contrast to what has been observed in mergers and acquisitions, short-term as well as long-term post-announcement excess returns are smaller when there are more women on the board. While the lower short-term announcement returns disappear once we control for expected benefits from share buybacks such as signalling, agency cost reductions and other benefits from increasing leverage, the relative long-term underperformance after buybacks approved by high diversity firms persist.
Long-term performance after buybacks is usually down to market timing based on superior inside information: companies, possibly based on private information, believe their stock is undervalued and take advantage of this undervaluation to benefit long-term shareholders. Empirically, firms are more likely to be undervalued when they are smaller (and followed by fewer analysts, making potential information asymmetries between the market and the management easier), when they trade at low market-to-book ratios and if they have been beaten up in the six months prior to the buyback announcement.
Of course, one explanation for the gender effect on post-buyback announcement returns could be that firms with larger gender diversity “looked” less undervalued before the announcement – these tend to be larger firms, for example. However, even when we control for these firm characteristics the gender effect persists. Thus it seems that male-dominated boards may have better information to judge whether their stock is undervalued.
Better insider information
It is not because men are more overconfident, they are simply more confident because they have access to better information. We call this the “male information advantage hypothesis”. This male information advantage hypothesis was first proposed by A. Can Inci, M.P. Narayanan and H. Nejat Seyhun in the context of a study on insider trading. They find that female and male executives make profits from insider trading but men earn superior returns and trade more often. They argue that women have less access to high quality information because they are not part of the predominantly male information network.
Note that a share buyback can be considered a form of collective insider buying, to the extent that insiders own shares in the company and are long-term investors. Indeed, many firms apply the same blackout period (when all trades are forbidden) as they do for insider trading. This indirect insider trading is not illegal, although some may find it unethical. An alternative explanation for our results is that women tend to be more ethical than men. E. Scarlat, K.Shields and I. Clacher find that insider trading profits decline following switches from male to female CEOs and they argue that female executives change the corporate culture and encourage more ethical behaviour. This explanation, however, is difficult to reconcile with two other findings in our paper: the gender effect becomes less significant in the few cases (4 percent of our sample) when the CEO is a woman. Moreover, companies with significant female board representation tend to issue shares when the stock is undervalued. These two findings are more consistent with an information advantage hypothesis than with “ethical” considerations.
Perhaps the debate about diversity and about hiring more women in leadership teams is not well focused: it should concentrate on how to facilitate the access of women to the rich, predominantly male-dominated information (and social) networks – or, maybe, how to train men to be more ethical. In the context of our study, information creates value, and diversity matters as long as it allows for the aggregation of richer information. Problems should be solved at their actual source, not where they appear to be. Otherwise, value can be destroyed solving the wrong problems.
Theos Evgeniou is a Professor of Decision Sciences and Technology Management at INSEAD and is the Academic Director of INSEAD eLab.
Theo Vermaelen is a Professor of Finance at INSEAD and the UBS Chair in Investment Banking, endowed in memoriam Henry Grunfeld.