When a public company holds its annual meeting, shareholders cast votes on each director standing for election. Most of these contests are lopsided affairs, with nominees sailing to victory. But hidden inside those tallies is a subtler signal: the margin of support, or the share of “yes” votes a director receives compared with peers on the same ballot. That margin, it turns out, can reveal things about a company and even about the community where it is headquartered.
A new paper published in Applied Economics Letters asks whether the cultural character of a company’s hometown, its religious life, political leanings, and civic ties, shows up in how shareholders vote on women directors. The answer, according to the authors, is yes: women nominees generally receive stronger support than men, but that advantage shrinks in more religious counties, in more Republican-leaning areas, and in places with denser social networks.
The question behind the study
Women remain a minority on corporate boards, and a good deal of research has examined why they are nominated less often. Less attention has been paid to what happens after nomination, when shareholders actually vote. Because directors rarely lose these elections outright, researchers have often treated the voting stage as a formality. But the size of a director’s vote margin can matter for their standing on the board and future bargaining power.
Dave Carter of Oklahoma State University and colleagues at Central Washington University, Ohio University, Pennsylvania State University, and Miami University set out to see whether local cultural norms leak into these vote margins. Their reasoning draws on two ideas from the social sciences. One is that people tend to absorb the norms of the communities around them. The other is that decisions made by economic actors, including investors, often show a bias toward what feels locally familiar.
If a company is headquartered in a place where traditional gender roles are more deeply held, would shareholders be less enthusiastic about women nominees, even when those nominees are well qualified? That is the question the researchers designed their analysis to answer.
Building the dataset
The team assembled records of roughly 57,000 director elections at U.S. public companies between 2003 and 2017. Voting outcomes and director gender came from Institutional Shareholder Services. Director qualifications, an aggregate measure of experience and education, came from BoardEx. Financial variables came from Compustat.
To capture local culture, the researchers matched each firm’s headquarters county to three community measures. Religiosity was the share of residents adhering to any religion, drawn from the Association of Religion Data Archives. Political affiliation was the county’s Republican vote share in the most recent presidential election, from the MIT Election Lab. Social capital came from an index compiled by economists Anil Rupasingha and coauthors, which combines civic engagement, associational density, and community participation.
Their key outcome variable, borrowed from earlier work by Laura Field, Matthew Souther, and Adam Yore, was “Excess Vote For,” or the percentage of favorable votes a director received minus the median percentage received by their fellow directors at the same meeting. This effectively strips out anything specific to the firm or the meeting, isolating how a given nominee fared relative to the peers standing alongside them.
What the voting data showed
On average, women directors received somewhat more support than their male peers, a pattern the authors link to evidence that women nominated to boards often bring stronger qualifications. But that boost shrank, and sometimes disappeared, as the local cultural environment shifted.
In counties with higher religiosity, the added support for women nominees was significantly smaller. The same held for counties with higher social capital scores. The relationship with Republican vote share pointed in the same direction but was not statistically significant in the main regressions.
The social capital result caught the authors’ attention because it ran against their initial expectation. Social capital is often associated with trust and cooperation, qualities that might seem to favor a broader idea of who belongs in leadership. The researchers suggest two possible explanations. One is that dense social networks reinforce conformity, making shareholders in those communities more inclined to favor traditional director profiles. The other is more mechanical: the index draws partly on participation in organizations, such as business associations, golf clubs, and religious groups, that have historically been male-dominated.
Testing the causal story
An obvious concern is that something else about these counties, not their culture, could explain the pattern. To address that, the authors ran two additional analyses.
The first used an instrumental variables approach, which tries to isolate the parts of religiosity, political affiliation, and social capital that come from long-standing historical or demographic features of a place rather than from anything happening at the firm today. Religiosity was instrumented using its 1980 baseline level and lagged population. Republican affiliation was instrumented using state GDP growth and homeownership rates. Social capital was instrumented using racial homogeneity. The results largely matched the main findings.
The second analysis looked at companies that moved their headquarters during the sample period. If cultural norms in the surrounding community are actually shaping votes, then moving a company to a different kind of county should change the size of the woman-versus-man vote gap. The researchers found that when firms relocated to counties with higher religiosity or stronger Republican leanings, shareholder support for women directors declined relative to men after the move. The relocation effect for social capital pointed in the same direction but did not reach statistical significance.
What it means, and what it doesn’t
The authors interpret the pattern as evidence that cultural norms shape shareholder voting behavior even at a stage of the corporate governance process, ratification of nominees, where preferences are rarely thought to matter much. They stress that the gaps involved are modest in absolute terms; most directors still receive overwhelming support. But because women directors already start behind on measures like board leadership roles, differences in vote margins can affect their standing and influence.
Some caveats are worth noting. The study is observational, and the county-level measures are proxies for cultural norms rather than direct readings of shareholder attitudes. Many shareholders in a given election are institutional investors who are not themselves located in the firm’s headquarters county, so the mechanism by which local norms translate into voting outcomes is not fully pinned down. And the social capital finding, as the authors note, may say more about how social capital is measured than about community cohesion itself.
For investors and companies looking to expand board diversity, the paper offers a practical observation: nominating women is one step, but the reception those nominees get from shareholders can vary with where the company is based. In more conservative or more tightly networked communities, the authors suggest, active advocacy from investors who prioritize diversity may carry more weight.
