Brad Greenwood and Seth Carnahan discuss gender inequality in hiring and promotion at law firms.

If your instinct is to think that the political leanings of law firm partners have an impact on how female attorneys are paid and promoted, new research confirms it. Temple University’s Brad Greenwood, a professor of management information systems, and the University of Michigan’s Seth Carnahan, a professor of strategy, looked at a broad, multi-year database of promotions within law firms. It showed that women are at a disadvantage when their bosses — especially men — donate to Republican candidates but see less gender inequality when partners give to Democrats. They discussed the issue on the Knowledge at Wharton show on Wharton Business Radio on SiriusXM channel 111, covering what the data showed, whether women should be looking solely for liberals-run firms and what it might take to get the gender bias out of decision-making in the legal profession. Their study is titled “Managers’ Political Ideology and Gender Inequality in Hiring and Promotion.”

You can listen to the interview using the player above. An edited transcript of the conversation appears below.

Knowledge at Wharton: First, let’s get into how you two originally got the idea to research this.

Seth Carnahan: It’s a great question. Brad and I, in our research, are both broadly interested in gender diversity and gender inequality in organizations. And this was one of those topics that’s both socially relevant — people care a lot about how women get ahead in their careers — and also important for firms and managers, because there’s quite a bit of research that suggests that firms that are more diverse and have less gender inequality may perform better than other organizations. Within that broad context, this specific idea had both academic and practical motivations — real-world motivation.

On the academic side, there’s a lot of research that suggests that managers’ actions can influence gender diversity and inequality among subordinates. To be a little more concrete, there’s a relatively strong empirical pattern we see in data that female managers tend to hire and promote female subordinates at higher rates than male managers do. And there’s other research that looks at managers’ political ideology and correlates that with their investments and strategy. A big conclusion in that literature is that managers who are more liberal tend to have higher investments in CSR [corporate social responsibility] activities, things along these lines.

Knowledge at Wharton: Your research focused on law firms. That being said, is there a possibility that the results could correlate out to other sectors of the business community?

Carnahan: Yes. One of the motivations for this study actually was coming from the tech community. About a year ago, CEO Marc Benioff — he’s a well-known liberal political donor — came out and said, “I’m going to institute a new policy where I’m going to review the pay of everyone in the organization, and try to eliminate gender gaps in pay.” And then we saw Netflix, whose leadership tends to donate to liberal politicians, come out with a relatively large announcement about new family leave policies. So we saw these announcements, we had this theoretical motivation from academic research, and we said, let’s see if this pattern holds true. And we happened to have access to data on law firms.

Knowledge at Wharton: Given the type of people who are usually higher up within law firms, is there a little bit of a generational angle to this? The senior partner of a law firm is liable to have been in that firm for quite some time, and the people that are at the top of that firm may be in their 50s, 60s, possibly even 70s.

“The main differences between right- and left-leaning leading people are their views on equality and their views on the importance of social traditions.” –Brad Greenwood

Brad Greenwood: There’s two parts to that, to answer the question. First, empirically, we tried to control for things like age or other observable differences between partners. Our goal was to soak up all the heterogeneity that exists across these different partners and see really just what the effect of political ideology is in the end.

But honestly, I think this is less of a stereotype issue and more of what we would naturally conclude from some pretty deep streams of literature in political science. There’s some great work in [political science], specifically out of New York University, which suggests that the main differences between right- and left-leaning leading people are their views on equality and their views on the importance of social traditions. And one of those traditions in the American experience has been gender roles in employment.

Knowledge at Wharton: Obviously, the topic of gender inequality plays into the research in this area as well.

Greenwood: Exactly. If we think about the idea of gender roles, they’ve been rooted for better or worse in social tradition. As a result, the idea of women entering a powerful career like being a lawyer — or a lawyer or even a partner in Big Law — does mean we’re beginning to eschew those norms in some specific ways. We would expect — based on ideology and these views of the importance of social tradition — that different people are going to react to it in different ways. Conservatives may be more apprehensive about that sort of change, as opposed to someone with more liberal leanings.

Knowledge at Wharton: Outside of the lack of promotions, based on the data you collected from many of these firms, were you able to discern any other ways political leanings affect these lawyers, or even on the firm itself?

Carnahan: Yes. These are great questions. In the paper, we also explore two other outcomes. One is the extent to which partners select associates for their client teams. When you think about law firms, a very important way work is organized is, you have organizations, you have practice areas, and then you have client work come through the door, and partners and associates tend to coalesce in teams around those particular client matters. What we see pretty consistently is that more conservative male partners are less likely to select female associates for those teams, relative to moderate partners, while more liberal partners are more likely to select female associates for those teams relative to a more moderate partner.

Knowledge at Wharton: I was going to ask you whether you could see data that correlated the opposite point — whether or not there was a marked difference in promotions among those who were liberals or who donated to democratic candidates?

Carnahan: Yes, this is one of the interesting things about the paper. We do see that in practice areas that are predominantly liberal, women’s odds of promotion increase about 80% relative to those in practice areas that are relatively conservative. We do see this sliding scale along the ideology. But an important thing to note here is we still see high rates of gender inequality within these more liberal practice areas, such that women’s odds of getting promoted relative to men are still something like one-third to 50% lower.

Knowledge at Wharton: With all this data collected, what can be done with it?

Greenwood: One paper at a time.

Knowledge at Wharton: OK, I don’t want to put the cart before the horse here.

Greenwood: There are a lot of great benefits to this data because they allow us to control for a lot of the extreme minutia that might be influencing these results or might be causing bias in the estimates. But I think that there are some really great opportunities for future work that might come after this. … Can we actually estimate how views on gender inequality are changing over time? Are they getting better? Are they getting worse? For whom are they getting better? Under what situations are there differences in the evolution?

There’s also the opportunity to dive deeper into these constructs of liberal and conservative — break open that black box. Is it religious conservatives? Is it social conservatives? Is it fiscal conservatives? The same is true of liberals. Where are we beginning to see these effects to a greater or lesser degree? It’s an interesting way that we can potentially think about this stream of work evolving.

Knowledge at Wharton: For the firms where it’s harder for women to get promotions, how much is that a matter of the people at the top of the firms having been there for many, many years? And, with this data becoming more widely known, is it likely that you will see an immediate change, or even one within the next couple years, or will that not be the case because those who are at the top of those law firms will still be there? Might it be that you don’t start to see things change for another 10 to 15 years, when some of those partners a the top of the firm start to leave the firms and leadership starts to turn over a bit?

Greenwood: This is an issue of some amount of speculation, and I think you hit the nail on the head. On the one hand, there is a general shift towards people becoming less ingrained in social traditions over time. We’ve seen that over the last 100 years in American political discourse. So it is possible that, just through the natural evolutionary process of the market, these things will begin to get themselves ironed out.

But I think it’s also important to recognize that we’re talking about big, big dollars here. The National Law Journal pegged billing of associates for somewhere around $400 an hour — I think their estimates range between like $420 and $250. And that’s without a lot of firms reporting. These are big dollars. If these partners realize that they have an untapped resource out there, that there’s access to attorneys who are currently being underpaid, they’re going to begin to exploit that resource, and they’re going to begin hiring more women into their firms. You could actually see a rapid change if someone identifies that there’s this underutilized tool that could be brought into the firms and used.

Knowledge at Wharton: Is this, in some respects, a little bit of a wake-up call for female lawyers in terms of the type of firms that they’re choosing to go to — maybe even something they should add to the research they’re doing when they’re looking to join a firm?

Carnahan: This is a great question. For us, I think the first thing to note is that, in our conversations with lawyers, a lot of them, frankly, aren’t too surprised by our results. So I don’t know that it would necessarily shake folks up too much there. But I think it is something to keep in mind. Another thing that’s important to emphasize is that there are a lot of resources for lawyers who are trying to learn about firms before they join them, either as a law student or as an associate making a lateral move. They can talk to colleagues from law school. There are a lot of resources out there in terms of guidebooks that tell you about the culture of these firms. So I think we wouldn’t necessarily suggest that associates should look just at the political donations of the partners they might be working with, because there are so many other ways to get a handle on a firm’s culture.

“What we see pretty consistently is that more conservative male partners are less likely to select female associates for [client] teams.” –Seth Carnahan

Knowledge at Wharton: Is the available data on these firms general or specific in terms of the type of firm? Obviously, there are different segments of a practice. Firms practice different segments of the law. Is there a difference between one type of firm and another?

Carnahan: One of the powerful aspects of the data is that we have information on lawyers across all different kinds of law. Therefore, we’re able to do statistical things where we can control for those effects. We’re able to net that stuff out. Although one thing we haven’t really explored that might be interesting to look at, based on your question, is whether these effects are different across different types of law, say, IP law or environmental law, things along those lines.

Knowledge at Wharton: Was there any correlation in terms of where these firms were in the country — East Coast, West Coast?

Carnahan: We didn’t see that. We don’t specifically explore whether the effects are stronger, say, in the Southern United States, which tends to be more conservative, or in more liberal areas on the coasts. But we can say that our results aren’t really driven by those geographic factors. We also do some things where, for example, we exclude highly conservative areas, say the former states of the Confederacy, from the sample, and the results are similar. Or if we exclude highly liberal areas, say New York City or San Francisco, we see similar results.

Knowledge at Wharton: There are certainly areas of the country that seem to be more conservative or more liberal. I didn’t know if there was a direct correlation to that.

Carnahan: It would be interesting to look at whether those kinds of contextual factors, the ideology of the overall marketplace, influence these individual partners’ ideologies. It’s definitely an interesting question, and one we haven’t looked at yet.

Knowledge at Wharton: You also bring up the point that the types of bosses who may fall into this category are also less likely to sit on diversity committees.

Greenwood: Yes. I think the diversity committee — unlike the majority of this empirical investigation, where we’re looking at the likelihood of being hired, likelihood of turning over, likelihood of being promoted — really highlights a different attitude regarding even the importance of the issue. And this goes back to the differences on the sides of the political spectrum about the importance of inequality. As in, is this actually an issue in the minds of people? And a diversity committee, all else being equal, is not an extremely costly thing for people to sit on. So, just from first principles, these aren’t seen similarly by different employers.

Knowledge at Wharton: Are more women starting their own firms or working in firms run by women, to avoid this type of bias?

Greenwood: That’s a little bit outside the scope of the paper. We don’t necessarily see where people end up afterwards. But the idea of women and entrepreneurship is a super-hot space in research right now. There’s a lot of research that suggests that when people’s careers are stymied, they will select out of the organizations, either into an organization that will facilitate them to a greater degree, or they’ll start a firm on their own.

Moreover, there’s a lot of behavior recently that suggests you’re going to see social support groups emerge. Take Silicon Valley, where there are very few female VCs or female entrepreneurs: You have social groups like Women 2.0 and Women Who Code, places like the Hackbright Academy. These social structures are emerging to begin to support people as they progress in their careers.

Knowledge at Wharton: You also talk about who gets picked for client teams. If women are not being picked for these client teams, then the amount of money that they can bring into the firm is quite a bit lower, and that obviously has an effect on promotions as well.

Greenwood: Yes, and that’s another nice thing about this investigation, because it goes from soup to nuts. If you are less likely to be selected for a client team, or maybe on the margin you miss out on one particular opportunity, eventually, those things do add up and maybe you don’t get promoted to partner. And then eventually, maybe you do select out of the firm due to a lack of opportunity.

Knowledge at Wharton: Even though you weren’t looking specifically geographically, are there potential cities where this could possibly come up because of how some cities are more liberal than others?

Carnahan: We haven’t looked at this directly, but we could speculate on things here a bit, connecting back to what Brad was talking about before about how competition in the marketplace might reduce these effects. If you take that idea seriously, if you’re a conservative person in a market that has more liberals in it, it might be harder for you to impose your tastes when hiring or promoting your employees and still be competitive in the marketplace, because your competitors are less likely to be doing that. So I think that might be one reason where we might expect this to vary across cities, such that the effect could be stronger in more conservative cities.

“In our conversations with lawyers, a lot of them, frankly, aren’t too surprised by our results.” –Seth Carnahan

Knowledge at Wharton: Because we’re talking about gender gaps in general, from what you guys have seen, these biased effects can happen at any point within a woman’s career, correct?

Carnahan: That’s right. We see pretty consistently that the promotion effect is occurring later on in the woman’s tenure. Generally, promotion decisions are made from year five to year 12; that’s where we tend to see this effect. Whereas the turnover effect that Brad was talking about, where we see higher rates of gender inequality and turnover when associates are working for more conservative partners, those effects are consistent throughout the associate’s tenure. We don’t necessarily see a spike in exits early in women’s tenure, and then a decline in exits further on down the line. It tends to be pretty consistent throughout their time with the firm.

Knowledge at Wharton: Brad, in one article, you talked about what exactly might be the right level of diversity within a firm. That’s something that you really don’t have a good handle on, correct?

Carnahan: We have to emphasize that we don’t necessarily know what the right level of diversity is for particular organizations. And these partners could be making choices or tradeoffs that help them compete. Say a firm institutes a more aggressive, male-dominated culture that maybe in their particular client set or their particular city allows them to be successful. But it has this downside effect of not encouraging a practice area where women want to work. For us, it’s hard for us to necessarily say that the effects that we observe will ultimately result in lower firm performance, because these partners are making tradeoffs and optimizing on a lot of different choices.

Knowledge at Wharton: But isn’t there the potential to change over, say, the next 20 to 30 years based on the changes we’re seeing in other segments of the business world, where we’re seeing more women executives right now?

Carnahan: That’s exactly right. And something we do see — there’s been pretty good research on this by Christine Beckman at the University of Maryland, where Brad and I both did our Ph.Ds., and also Damon Phillips at Columbia University. What they show is that law firms are more likely to promote female associates as their clients have more female leadership. So we could expect to see these feedback loops that you’re describing over time, such that when large S&P 500 firms start promoting women, it puts pressure on law firms to also do the same. That could, over time, iron out some of these differences that we see.

Greenwood: That’s certainly going to happen. We hopefully will see a continued evolution in the workplace where equality comes more and more to the forefront. I think what’s also important to recognize … is that a lot of the discussions that we’re having in the popular press are almost from theological viewpoints on whether or not ideology is really an issue. They’re not based on concrete data where we’re getting reasonable estimates of the actual differences. If we don’t do that, if the conversation or discourse isn’t based on substantive ideas or really rigorously done studies, that’s a potential problem, because it just gets everybody off track.