No, don't leave executive pay to the shareholders

Alex Edmans has a blog on executive pay on HBR here. It’s wrong-headed in my view, so here’s a largely sarcasm and swearing-free response to it.

Politicians typically make two suggestions for pay reform. First, to cap, or at least force the disclosure of, the ratio of CEO pay to median employee pay. Second, to put pay packages to an employee vote, or as May suggests, put workers on boards.

In my experience of public policy around executive pay these claims are very wobbly. Actually politicians of various stripes have primarily focused on the disclosure of more information on the nature of executive remuneration and the introduction or extension of shareholder powers. If we look at the last few rounds of reform of executive pay the focus was on these two areas.

In the UK, I think I’m right in saying that the ideas of disclosing pay ratios and putting employees on rem comms, or giving the employees a vote, were part of the BIS narrative reporting consultation. But they weren’t part of the specific consultations on executive pay, which very much focused on disclosure and shareholder powers. It was the same with the original Directors Remuneration Reporting Regulations - nothing in there about pay ratios or employees on boards, or employee votes. And I don’t think I have seen a politician with a brief covering corp gov issues advocate a pay cap.

In reality it is only within the last five years that there has been some interest in both pay ratios and the role of employees in corp gov. But even now, for example in May’s proposals, there is emphasis on shareholders. It is inaccurate to make a claim about what suggestions politicians typically make without acknowledging that shareholder powers and greater disclosure are two most common recurring themes.

It is shareholders who bear the costs of paying the CEO, and so it is unclear whether the government should intervene.

I don’t think either assertion here can pass unchallenged. In terms of “bearing the costs” of CEO pay, as I’ve argued before I don’t think it is legally correct to say that CEO pay ever belonged to shareholders. In practice when an investor buys shares in a company the money they pay over goes to the previous owner of the shares, not the issuer. So there is no contribution of capital to the company that is then used to fund CEO pay.

As to whether the Government should intervene. Even from the perspective of a shareholder primacy advocate (which I am not) you can see this might be necessary. For example, if investors are unable by themselves to force the disclosure of pay ratios at all companies (which seems very likely). It is reasonable for the state to intervene to mandate disclosure of information if some investors think it is useful (as, for example, the Investment Association now does).

While it’s the level of pay that captures politicians’ (and the public’s) attention, it’s the structure of pay which matters more for firm value – for example, whether it vests in the short-term or long-term... The electorate will be more impressed by a politician who proposes a headline-grabbing law to halve a CEO’s salary than a politician who extends the vesting horizon from three years to seven years, even though the latter will have a far greater impact on long-term value creation.

The electorate may not believe that “firm value” has a meaningful impact on them, whereas they may believe that inequality is excessively high. I am personally skeptical that increasing vesting horizons will have a meaningful impact on executive behaviour or value creation. Is it really unreasonable therefore for a politician to focus more attention on a societal issue that angers many, as opposed to a technical reform that may have little, if any, real impact on the electorate?

Moreover, even if shareholders didn’t take into account the effect of poorly-designed contracts on CEO actions, it’s not clear why the government should regulate pay rather than these actions themselves – surely the most direct route to curtailing them.

This is a reasonable point, but doesn’t follow on from the last one. Again we need to be clear about what we are trying to deal with. Our much-mocked vote-grabbing politician may believe that there is more than one issue to be tackled in reform – both inequality and perverse incentives regarding R&D spend, for example. So, yes, they could intervene directly there as well as seeking to put downward pressure on pay levels.

A second motivation to lower pay is to reduce inequality. However, attempts to curtail pay through regulation may backfire. Kevin J. Murphy describes how the entire history of executive compensation regulation is filled with unintended consequences.    

It is certainly true that the history of executive pay policy is filled with assertions of unintended consequences, but I personally believe this often based on a misunderstanding of social action. And this also raises big problems for shareholder primacy advocates.

On the first point, is it really the case that increased CEO pay and perks are an “unintended consequence” of intervention? Do you mean that remuneration committees unwittingly awarded CEOs more pay and perks? Because to me it looks like CEO pay and perks increased because companies intended to increase them. So what we have to isolate here are the actors, their intentions and the outcomes.

Policymakers have typically enacted reform to increase disclosure around executive pay with the expectation that this information would shame companies into restraint and/or encourage shareholder engagement over pay that enforces restraint. The immediate consequence has indeed been the intended one – the disclosure of more information.

However, then we move into the second piece of action – the reaction of companies. Rem comms have reviewed the information available and concluded (presumably) that their CEOs have needed more pay and perks. They have then awarded more pay and perks to CEOs, and that was the intended consequence of their action.

I don’t see unintended consequences – I see unrealized expectations. The first of which is the failure of rem comms to exercise restraint.

This is the process: Actor A is trying to achieve Outcome 1 through Policy X, but Actor B does not want Outcome 1, and prefers Outcome 2. Actor A successfully implements Policy X, but Actor B undertakes Response Y that achieves Outcome 2. In this case is Outcome 2 an “unintended consequence” of Policy X, or is it an “intended consequence” of Response Y? When we are dealing with two sets of agents who have different desired outcomes these issues are not as clear cut as some would have us believe. 

Similarly, the other “unrealized expectation” of executive pay reform is that shareholders would respond to disclosure by exerting some pressure for restraint. But they have rarely done so. This raises all sorts of questions – not least do they really have the right incentives to act, and might shareholders approve of pay that society considers unacceptable? I will come back to this, but in my opinion those who promote both the “unintended consequences” theory of pay and shareholder primacy need to account for the failure of their favoured stakeholders to act.

A CEO might reduce his company’s pay ratio by firing low-paid workers, converting them to part-time status, or increasing their cash salary but reducing their non-financial compensation (such as on-the-job training and superior working conditions).

Really? We can all think of possible negative outcomes of policies we don’t favour, but they have to pass the bullshit test, and this one doesn’t for me. If CEO was seriously willing to fire low-paid workers or radically restructure their remuneration just to manipulate a pay ratio figure then I hope the other board members would fire them. Risking alienating the workforce over something so trivial would clearly not be in the company’s interest and I don’t believe that CEOs are that dumb.

A cap could also lead boards to focus on the “optics” of pay (e.g. a low ratio) and ignore more important dimensions, such as performance targets being long-term rather than short-term.

Again I perhaps have a more optimistic view of board directors’ mental capabilities. I will go out on a limb here and say I think they could manage to consider both how a pay ratio looks AND make sure targets for incentive schemes are long-term.

And we need to unpack all the assumptions in this claim. Are long-term targets "more important"? Perhaps if you think that inequality doesn't matter, and that boards can't learn anything useful from looking at their own internal pay structures. And you also have to believe that making targets more long-term would have a meaningful impact on executive behaviour that would have positive impacts that go beyond those from looking at pay across the business. This in turn relies on implicit beliefs about how people will respond to incentives and targets (particularly those several years ahead) that might be highly questionable if spelled out. (they very rarely are in exec pay policy discussions though.)

The reality is that what is being asserted here does not take account of the fact that there are different interests and objectives in executive pay policy. 

So if capping the pay ratio wouldn’t work

Not convinced this has been proven by the above, but go on…

what about putting workers on boards, or submitting CEO pay packages to an employee vote? There are reasons to be skeptical here as well. An employment contract is an extremely complex issue and cannot be whittled down to a simple number such as a pay ratio, which the vote might focus on. It covers topics such as the optimal vesting schedule, the appropriate mix of stock vs. options vs. salary vs. pensions vs. bonuses, whether industry performance be filtered out and, if so, how (indexed options? indexed stock? options on indexed stock? Stock with indexed performance vesting thresholds?) Confused? Well, so might employees be, should CEO pay contracts be put up for a vote.

First, we’ve conflated two things here – employees on boards (and presumably rem comms) and an employee vote on exec pay. If we are talking about the first case, I don’t see why employees (plural!) on a rem comm need be inherently less capable of forming a view than any other committee member. If they don’t understand they can ask (as more rem comm members should) and other board members or the rem consultant can explain. If they still don't understand in my opinion they shouldn't agree to what is being proposed.

If we are talking about a vote, I don’t see why a union or some other representative body if no union is recognized couldn’t analyse the proposals and make a recommendation to employees on how to vote. Yes these arrangements are complex, but that is a result of lots of attempts to redesign performance pay, with the knock-on that many execs don’t understand the system either. In fact, this is a good argument for scrapping most of the existing structures and going back to fixed pay as the main element, but that’s for another day.

As an aside, it seems to me that underlying this section is the implication that employees are just too thick to understand exec pay.

Companies already have natural incentives to treat employees as valued partners to the business – one of my own studies showed that firms with high employee satisfaction beat their peers by 2-3%/year. 

Hang on a minute, you just said that chief execs would be willing to fire low-paid employees or fiddle their pay just to skew their internal pay ratio. Which is it – are employees valued partners, or viewed as a line of data to manipulated? I find it hard to think the answer is both. And might not employees feel themselves valued if they were allowed some involvement in the direction of the business through board membership?

Indeed, Gary Gorton and Frank Schmid found that worker representation on German boards is associated with lower profitability and firm value.

I won’t go into a full review of the literature, since this is a big topic. Only the other day I saw a Bloomberg article that said firms with employee representation outperformed, for example. But in any case, again we don’t just argue this in relation to firm value and profitability, it is a much bigger argument.

Is the message to do nothing? Far from it. It’s to leave the decisions to major shareholders, who have the expertise and incentives to get these decisions right.

Once again, I think what’s being asserted here is hugely contestable. Do shareholders have the “expertise” to get it right? In an ideal world, executive pay would be about recruiting and motivating the best people. For shareholders to have “expertise” here I think would require them to be a mixture of recruitment agents and behavioral psychologists, and clearly they are neither. Instead they apply very general rules of thumb – pay the going rate, tie pay performance – that both easily to manipulate (former) and of questionable effectiveness (latter).

The question of whether shareholders have the incentives to act is equally troublesome. There are obvious conflicts of interest – asset managers may manage money for the pension fund of the target company, or be part of a bigger financial services group with a client link. There are biases – asset managers are well-paid, certainly relative to most beneficiaries, and thus not as likely to share concerns about high pay (because they don’t consider it high).

Then there’s the collective action problem: why bother engaging over pay if the benefits of doing so, which are likely to be small, accrue to all investors (including other asset managers who are overweight in the stock relative to you)?  

Even if our asset manager does have an incentive to intervene, because they have a big overweight position in a particular company, we can't assume the outcomes would be better, certainly not in terms of addressing high pay. If the asset manager believes in the “war for talent”, aren’t they going to want to pay as much as possible to secure that talent? For example, here's what one institutional investor said when M&S awarded Marc Bolland a large upfront payout:

"If the company is going to hire the best people, if they are going to take them out of the company they have been working for, then they have to buy them out. That is the way the world works."

And what if another investor overweight in a competitor, thinks that same? Won't they push for "pay what it takes" too? And isn't the likely result an exec pay arms race why may result in increasing both the cost of executives across the board and the pay gap within firms?

Therefore we should be absolutely clear that there a) there are sound reasons for shareholders to not act effectively and b) even if shareholders do act there is no guarantee that the pressure will be downwards.

Unlike regulation, which is one-size-fits-all, shareholders can decide what the optimal pay package is for that particular firm.

But in reality does it result in pay packages that meet the needs of the firm, or does it just replicate the type of pay that asset managers are used to? David Sainsbury made a good point in the largely disappointing Progressive Capitalism that the directors of PLCs now have financial markets style pay. Most pay structures seem to look pretty similar in essence: smallish base pay, short-term cash bonus, various share schemes. Is that really right for all companies? Isn’t that “one-size-fits-all”?

And, again, there is no guarantee that the impact will be preventing executive pay going further up overall. Here’s what BlackRock said a few years back –

One of the difficulties we face as investors is that we (rightly) assess a company’s arguments for pay changes or increases in light of that company’s circumstances. Generally, companies do present strong arguments for the changes they wish to make. But our assessment tends not to take into account the impact it will have on the trend overall.  

This is what I think happens in practice. What looks right firm by firm can bid up the price across the board.

Moreover, when pay is inefficient, it is often a symptom of a more underlying corporate governance problem, brought on by conflicted boards and dispersed shareholders. Addressing pay via regulation will solve only these symptoms; encouraging independent boards and large shareholders will solve the underlying problem. That will improve not only pay, but other governance issues.

So, a minute ago we were being told that we need to leave pay to the shareholders because they have the expertise and incentives to act. But now it turns out that that both boards and shareholders face challenges that result in poor pay practices. And the solution to this is to “encourage” different boards and different shareholders to be put in place. If we have to change the nature of the two principal actors (as shareholder primacy advocates see corp gov anyway) in order to get better outcomes that seems like an indirect way of tackling the problem.

The bottom line is that, to the extent pay is a problem, it should be shareholders, not politicians or employees, who fix it.

Based on my experience of public policy relating to executive pay, I would say we have tried to rely on shareholders to fix it, and the results aren’t great. The reality is that shareholder oversight is a weak tool and has not restrained executive pay. If it has had any impact, it has been to make executive pay (regardless of sector) look more like pay at the top of financial services companies.  

Relying on shareholders to fix executive pay requires us both to accept that the views and interests of shareholders (in reality asset managers) are the only ones that count, and to make some heroic assumptions about how they will react in practice. In my opinion it's a very partial take on executive pay as an issue, and in practice leads to monomania in terms of policy ideas, always returning to the same old same old: better disclosure, better targets, better shareholder powers. If we yet again choose this path, we will get more of the same: exec pay will simply look like what asset managers think is reasonable (both in structure and scale). But that is a very long way away from where the public is at. 

Policy “solutions”, if there can ever be any that are more than temporary fixes, have to take account of the reality that there are different interests, and they may pull in different directions. You may not be able to make executive pay seem more reasonable to the public AND structure it in a way that asset managers like. To say “leave it to shareholders” is an assertion of the primacy of asset managers' views and interests. It is not unreasonable for politicians to conclude, after various attempts to make shareholder oversight work, and facing the reality of both high executive pay and an angry public, that other options are worthing looking at. This isn't making the issue political, it is acknowledging the nature of the problem that politicians have to deal with. 

The only positive I can see from agreeing to a “leave it to the shareholders” approach is that (while it will waste a few years) it will fail to tackle the problem, which makes more radical action further down the line more likely. It is the “anti-capitalist” option, provided that you are measuring performance over the long term…