Edmans opens up his research agenda from CSR to David Beckham by way of CEO pay.
LBSR: What are you working on?
Alex Edmans: One line of research that I’ve looked at recently is the link between corporate social responsibility (CSR) and firm performance. This is something not traditionally studied by finance people.
One argument is that companies should just care about their shareholders because shareholders are the ultimate owners of the firm. And this view is not as narrow-minded and as hard-hearted as it may sound. In order to be successful, a company needs to deliver good products for customers; it needs to treat its employees well, and so on. So you can simply focus on your shareholders, but doing so is going to make you care about these other stakeholders as a by-product.
But then there’s the second view, which is the CSR view that you need to explicitly care about other stakeholders. This is because it is difficult to see the shareholder implications of treating stakeholders well – for example, what is the ultimate profit of reducing your carbon emissions? Thus, you will only do so if you care explicitly about the environment, in addition to shareholders.
LBSR: And the debate here is: are companies that are socially responsible actually delivering greater value or is this something that is costly to them?
AE: The big problem is that it’s very hard to say what’s causing what. If I showed that companies that were socially responsible are also profitable, it could be that social responsibility causes profitability because workers are happy and customers are satisfied. Or it might be that only profitable companies can be socially responsible because they have enough money to spend on things like employee health benefits. You just can’t correlate it with profit.
This is why I come in as a finance person. We can use finance techniques to get around this issue. In finance, rather than looking at accounting profits, we can look at future stock returns. If there is indeed reverse causality – that CSR is being caused by high profitability – then the profitable firm will already have a high stock price today and so you should not expect the stock return to be high in the future. Thus, the future stock return will calculate everything that is over and above the current level of profitability. That’s a way of getting closer to causality.
LBSR: So you can measure performance against stock returns, but how do you measure social responsibility?
AE: I looked at the ‘100 Best Companies To Work For In America,’ published every year in Fortune. This is a narrow dimension of social responsibility as it just looks at employee welfare, but I use it as it is a particularly accurate measure. It doesn’t tell you how much companies spend on employee welfare programmes, but whether employees are satisfied and engaged as a result.
I found that companies on this list outperform peer firms significantly. I looked at a 26-year period, which includes both booms and recessions, and found that they beat peer firms by two to three per cent a year, which is highly significant. The nice thing with finance techniques is you can control for risk and peer performance. And it’s not that these companies are riskier, or doing well just because their broader industry is doing well. This is quite comforting – it shows the companies that treat their employees well deliver high shareholder returns.
LBSR: So goodness leads to greatness?
AE: Yes, at least when goodness is defined in terms of human capital or employee welfare.
LBSR: Were the results surprising?
AE: If you were to stop somebody randomly on the street and tell them this study shows that companies do better if their workers are happier, they might say that’s completely obvious, why do you need to do a study? But it’s actually not obvious at all because the way that we typically think about workers is just like any other input. Indeed, the assembly line was invented to work employees as hard as possible, ie. to reduce their welfare.
LBSR: Yet the percentage of people who are disengaged at work is substantial, isn’t it?
AE: Yes, it is. This is again because a lot of managers focus on the bottom line impact and something like this is unquantifiable. The other interesting thing in this study is while I found that these companies outperformed their peers, it took up to four to five years for the benefits of employee satisfaction to show up in the stock price. There are ways in which managers can act myopically in order to deliver on earnings targets.
LBSR: I suppose, in their defence, they would say that the churn of CEOs is increasing and that encourages short-termism.
AE: Think why is the churn of CEOs increasing? That might be because institutional investors themselves might only look at short-term performance metrics, so this problem is systemic. If equity analysts are only going to be studying companies by looking at their earnings announcements, then obviously managers will all look at these earnings announcements.
LBSR: Is there any sign that a systemic change is happening?
AE: Potentially. Now people are thinking about triple-bottom line reporting. Companies are trying to disclose levels of sustainability and there is increased interest in CSR from institutional investors. About ten per cent of investments are put into socially responsible funds. That’s not a lot, but it’s still not an insignificant part of the shareholder base. Any such change is always going to be slow, but at least the change is going in a positive direction. Indeed, major investment banks now have equity research teams focused on social responsibility.
Another thing that may help in terms of institutional investors is the size of the stakes they have. People typically emphasise the benefits of diversification – that shareholders should have small stakes in many firms. And in the US you have something called a Prudent Man Rule, which is that mutual funds should invest as a prudent man would, which is not to have too many eggs in a basket. But if that’s the case, you don’t have enough skin in the game in any particular company in order to look beyond the numbers, so it’s costly for you try to gather additional information. There are people like Bill Miller, who formerly ran the Legg Mason Value Trust in the US, and was known for having a focused fund of maybe 30 stocks. This means if you want to invest in a casino you go to the casino, see how the workers are treated, how customers are treated and glean additional information that way. Now that’s possible if you’ve got 30 stocks, but not if you’ve got several hundred stocks, which is what has often been encouraged.
LBSR: They have very little knowledge of the organisations they’ve invested in?
AE: Right, and also a reliance on numbers, and perhaps not sufficient appreciation of the limitations of the numbers. In the light of financial scandals, there are always calls for more disclosure. Again, this is not necessarily helpful, because that’s going to encourage people to focus on the disclosable matrices. If you’re just going to publish exam results from schools, then we’re going to skew efforts to just getting good exam results at the expense of extracurricular activities, a love of learning or respect for authority. So, maybe sometimes it’s better to disclose nothing than to disclose some things.
LBSR: So, if you’re going to measure a teacher’s performance, it’s got to be on a range of issues – not just focused on exams – and the same with managers, it can’t just be related to financials or just numbers?
AE: Correct. And we also can see this greater call for disclosure in terms of things like executive compensation; again, this is not necessarily going to be helpful, so one call, for example, is to try to force the disclosure of the ratio of the CEO’s pay to the median employee in the firm.
LBSR: But why is that a relevant metric?
AE: The CEO’s pay should perhaps be relative to other CEOs in the same industry or it should depend on their performance. We should not mind if the CEO gets paid 400 times the average worker if he has delivered huge returns. A £2m CEO salary seems a lot. But let’s take a £10bn market cap firm. If the CEO only adds 0.1 per cent of firm value compared to the next-best CEO, that is £10m in a £20bn firm.
LBSR: But the CEOs are not necessarily directly accountable for that increase in performance.
AE: People think that all CEOs do is play golf and fly corporate jets, but that’s a hugely caricatured view. There are many companies like Apple which would not be in the state they are today if it was not for an inspirational leader. People are willing to accept high pay for people like David Beckham, a movie star or a pop star, because they say, oh, well we can see this person’s talent, we can see this person take free kicks. You’d never compare David Beckham’s salary to the median salary at whatever football club he’s playing for, because an administrator of a football club and a footballer are different jobs. You compare him to other midfield players that you could have in the team.
LBSR: Can and should academics really play a part in this debate?
AE: The problem is that people are often angry about a single case. You can’t or shouldn’t make law on the basis of one case so that is where academic research is useful.
The reality is that the media is only going to report the one or two cases where compensation is egregious. The media only reports cases in which things break, just as it would never report that there was no traffic accident on a particular day.
There was a case in the US where a manager got a golden parachute and people said it was outrageous. And so a law was introduced limiting golden parachutes to no more than three times the salary, otherwise there would be tax implications. What happened is that companies that didn’t have golden parachutes adopted them and companies that had golden parachutes that were with one or two times salary began to think that three times salary was the right number. It backfired.
LBSR: Isn’t what makes people uncomfortable, a matter of fairness and risk? There is an idea that bankers in particular and people in the financial world were – and to some extent still are – rewarded but the risk was actually apparently minimal in that they couldn’t lose.
AE: It is worth noting that CEO pay has gone down by 40 per cent since the financial crisis, and rank and file pay has not come down by the same amount. The CEOs of firms such as Bear Stearns and Lehman lost literally hundreds of millions of dollars. True, they made extremely costly mistakes, but it is not clear that incentives were the problem. In fact, Lehman’s compensation model was perhaps closest to what practitioners recommend – employees had large stakes in their firm, and so a lot of skin in the game. Similarly, footballers miss penalties in important games, but these ‘mistakes’ are not the result of poor incentives.
LBSR: So are you saying that current pay practices are perfect?
AE: Not at all. Change is needed, but people are criticising the wrong things. A lot of the debate surrounds the level of pay, because this is politically the most controversial – a CEO earning £2m when the average worker in his firm earns £25,000 sounds outrageous. But, £2m is very small compared to a £10bn firm – it is 0.02 per cent. The more important question is not the level of pay, but the sensitivity of pay to performance – ie. the CEO’s incentives. If the CEO is paid according to short-term rather than long-term performance, he may underinvest and destroy two per cent, three per cent of firm value – far greater than the value lost by too high pay levels. Indeed, one of my papers shows that CEOs cut R&D in years in which significant equity vests. Thus, the solution is to lengthen the vesting periods of stock and options. Even though this will win fewer votes than ranting against the level of pay, it is ultimately more important for firms’ long-run growth.