Feifei Li: Alex, you’re very well known for your research into topics such as executive compensation and socially responsible investing. In 2011 you published your research on employee satisfaction and whether it matters in terms of corporate performance and stock market returns.1 Can you set up the broad framework underlying your research?
Alex Edmans: Sure. Employee satisfaction could matter for corporate performance, but it’s not clear which direction it matters. There are two schools of thought. One is that employee satisfaction is detrimental to firm performance. Why might that be? Well, one way you can make your employees happy is to pay them too much and not work them hard enough. And there are actually reasons why a manager might want to do that; for example, it takes a lot of effort to reduce unproductive employees’ wages, or management might simply enjoy better working relationships when workers are better paid. Nowadays, the ratio of the CEO’s pay to workers’ pay is being disclosed. A CEO who pays workers more might be able to justify higher pay for herself.
So, that’s the dark side. But there’s also the bright side, which is if you treat your workers well then you’ll be able to hire better people, they’ll stay longer, and they’ll be more productive and motivated. And even the result of that is not obvious. A manager might think the way to motivate workers is to count the number of widgets they produce and pay them per widget. But in the intangibles economy we now live in, a lot of the things managers seek from workers can’t be measured, such as mentoring subordinates. Therefore, the way to actually motivate today’s workers is to make them happy to be part of the firm and to be inspired by the mission of the company.
Feifei: Could you dive a bit more into the details of your research findings?
Alex: Yes. I wanted to look at the link between employee satisfaction and performance. So I needed to think about two things. First is how I could measure employee satisfaction. I used the list of the “100 Best Companies to Work For in America,” which has been available since 1984. It gave me tons of data to make sure my findings were solid. The list doesn’t look just at quantitative factors, such as pay and benefits, but it also considers qualitative factors, such as trust and management, pride in job, and camaraderie with colleagues.
The second thing I needed to measure was financial performance. I chose not to look at valuations or market share or profitability because of potential reverse causality: Does employee satisfaction lead to better performance or does better performance lead to employee satisfaction? Instead, I decided to look at future stock returns to mitigate—but not completely eliminate—some of the causality issues. For example, let’s say good performance causes workers to be happy. If performance is already good, the stock price should be high, and the future return shouldn’t be expected to be too much higher off of an already-high stock price. So that’s one way to mitigate some of the reverse causality concerns.
Feifei: What surprises me is how the market could miss that. Why didn’t investors realize employee satisfaction matters for corporations and start bidding up the prices for the firms listed in the “100 Best Companies to Work For in America”?
Alex: That’s a good question. I already mentioned the two hypotheses for why employee satisfaction might either be good for stock returns or bad for stock returns. And there’s a third option—that there is no effect. Let’s say even if employee satisfaction was good for stock performance, and if the market recognized that, it should already be priced in. But why might it not be priced in? One reason is the ambiguity around the impact of employee satisfaction: Is it positive or negative? And, indeed, some of the early research suggested that treating workers better was related to poorer stock returns.
A well-known study by Abowd,2 published in 1989, found that coincident with a pay increase, stock returns fell dollar for dollar, suggesting a zero-sum game. Maybe, as a result, the market is fixed on this zero-sum idea. Or perhaps the market does recognize that employee satisfaction matters, but just doesn’t pay attention to the information because it’s not very salient. A lot of research suggests that intangible factors are not incorporated into stock prices.
Gompers, Ishii, and Metrick (2003)3 showed, however, that better governance is associated with higher stock returns. You might think it’s obvious that if a firm is better governed, performance should be higher, similar to the situation that if employees are happy, performance should be higher. But despite governance measures being public information, the market doesn’t take them into account. The market is good at looking at metrics such as earnings and profits and dividends, but less good at incorporating intangible factors such as governance or employee satisfaction.
Feifei: In your study, have you conducted any subsample analysis to examine if the magnitude of the abnormal returns changes as the market becomes more aware of the potential benefits to socially responsible employers?
Alex: Yes. I looked at that because, as you say, the stock return performance depends not only on whether employee satisfaction is good, but the extent to which the market recognizes it. I looked at whether things changed after 1998. Before 1998, the Best Companies list was only published as a book. After that year, the list was published annually in Fortune magazine, which today has 20 million readers, both online and in print. We might expect that after 1998 the result should disappear because the information is so widely disseminated.
But interestingly, I found that the results were stronger after 1998, not weaker. This suggests that even if the market is aware of the information, it is not being taken into account—maybe because the prevailing view is that an employee-friendly company is just a lot of “fluff” and distracted from the bottom line. Therefore, the market doesn’t see the attribute as a positive investment criterion.
Feifei: Your research is supportive of the idea that socially responsible investing can be potentially beneficial for investors.
Feifei: But we all know socially responsible investing is a very broad topic that can be overwhelming and confusing to investors. Could you please comment on what investors should be looking for when they are evaluating ESG strategies?
Alex: One of the most challenging pieces of data for the socially responsible investing (SRI) movement is that the average SRI fund does not outperform the market. You might think it’s strange that my research finds a dimension of social responsibility that does outperform. I think it’s important to distinguish between socially responsible funds and socially responsible criteria. Some funds may not be outperforming because they might be using the wrong criteria. I chose employee satisfaction because employees are material to almost every company. Employees are a key asset, so there is a good reason for a link between them and a company’s future stock performance.
Other dimensions of social responsibility, such as Catholic values or animal rights, may not actually have a material impact on performance in some industries. If an SRI fund is engaged in a one-size-fits-all diagnosis of whether a company is socially responsible, then they are likely not looking at the correct screens.
A paper by Khan, Serafeim, and Yoon4 published in 2016 in The Accounting Review examined materiality. The authors looked at companies which performed well on social dimensions across the board and found these companies did not outperform the market. Then they looked at companies that outperformed on dimensions that were material in the company’s industry. Let’s say in healthcare, that dimension might be the environment, and in financials, it could be customer trust. These companies underperformed on immaterial dimensions, but still outperformed the market, which suggests that “more” social responsibility is not always better. The focus should be on the material issues in the company’s industry rather than on the immaterial issues. An improper focus may be the reason some SRI funds are underperforming.
Feifei: Is there value in translating your research into an investible strategy?
Alex: Yes. Indeed, the Parnassus Endeavor Fund is following a strategy that aligns with my findings. In 2017, Morningstar reported that the Endeavor Fund, which used to be called the Workplace Fund, was the best performing large-cap fund over the past 1, 3, 5, and 10 years. I believe since its inception in 2005 through 2017, it returned 12.5% per year, while the market returned 8.3%.
Understanding which SRI criteria matter is important. In terms of employee satisfaction, results are driven by two factors: One, employee satisfaction is good for firm value; it does not represent managerial slack. And two, employee satisfaction is not only good for firm value, but many investors do not recognize it’s good, and therefore it is the ideal criterion to invest on.
Feifei: What is the magnitude of the abnormal returns for the best-to-work-for companies in your study?
Alex: I compared the “100 Best Companies to Work For in America” to other companies in the same industry or those with similar characteristics. Depending on the benchmark, I found they outperformed by 2.3% to 3.8% per year over a 28-year period for a cumulative 89% to 184% outperformance.
Feifei: Thank you very much for sharing your thoughts with us, Alex.
Alex: Thanks very much for your interest.