How should CEOs be paid?

Executive pay is a high-profile topic and almost everyone has an opinion on it. Many shareholders, workers and politicians believe the entire system is broken and requires a substantial overhaul. But, despite being well-intentioned, their suggested reforms may not be targeting the elements of pay most critical for shareholder value and society.

The level of pay

Much of the debate is on what I call a Level 1 issue – the level of pay. For example, in September 2013, the US Securities and Exchange Commission voted in favour of a measure that would require public companies to disclose the ratio of the CEO’s pay to the median salary of its employees. The European Commission is contemplating going further by requiring a binding vote on this ratio. Separately, proposals to increase taxes – most prominently made by Thomas Piketty – are a response to seemingly excessive pay levels.

While high taxes or ratio caps would indeed address income inequality (an important topic, but beyond the focus of this article), it’s unclear whether they would do much to improve shareholder (or stakeholder) value. The levels of CEO pay, while very high compared to median employee pay – and thus a politically-charged issue – are actually very small compared to total firm value. For example, median CEO pay in a large US firm is $10 million – only 0.05% of a $20 billion firm. That’s not to say that it’s not important – a firm can’t be blasé about $10 million – but the other dimensions may be more important.

The sensitivity of pay

Instead, what matters for firm value isn’t the level of pay, but the incentives that it provides to CEOs: as Jensen and Murphy famously argued: “It’s not how much you pay, but how.” Level 2 thinking studies the sensitivity of pay to performance. Specifically, it looks at how much of a manager’s total pay is comprised of stock and options (which are sensitive to firm value) rather than cash salary (which is less so). As the thinking goes, greater stock and options make the CEO more aligned with shareholders and thus provide superior incentives. Jensen and Murphy bemoaned the low equity incentives at the time as evidence that CEOs were “paid like bureaucrats”.

However, while seemingly intuitive, there is no evidence that better-incentivized CEOs perform better. Of all the banks, it was Lehman Brothers whose compensation scheme most closely matched what Level 2 thinkers argue is the ideal – very high employee stock ownership. Using a larger sample, Fahlenbrach and Stulz found “some evidence that banks with CEOs whose incentives were better aligned with the interests of shareholders performed worse and no evidence that they performed better”. Indeed, the European Commission has recently capped banker bonuses at two times their salary, seemingly reducing bankers’ incentives to perform well – but also reducing their punishment if things go badly.

The structure of pay

The concern with high equity incentives is that they encourage CEOs to pump up the short-term stock price at the expense of long-term value – for example, writing subprime loans and then cashing out their equity before the loans become delinquent. But the root cause of this problem isn’t the amount of stock and options that a CEO has, but their vesting horizon – whether they vest in the short term or long term, and thus whether they align the CEO with short-term or long-term shareholder value. Level 3 thinking thus focuses on the structure of pay.

In a 2014 paper, my co-authors and I studied how CEOs behave when they have a significant number of shares and options vesting. CEOs typically sell their equity upon vesting to diversify, and so vesting equity makes them particularly concerned about the short-term stock price.

We found that in those years where CEOs have significant equity vesting, they cut investment in many forms – R&D, advertising and capital expenditure. Moreover, in those years, they’re more likely to exactly meet or just beat analyst earnings’ forecasts. If the forecast is $1.27 per share, they report earnings of $1.27 or $1.28. Indeed, the magnitude of the investment cuts is just enough to allow CEOs to meet the target. Thus, vesting equity induces CEOs to act myopically – to cut investment to meet short-term targets. These results control for other equity incentives, such as unvested equity and voluntary holdings of already-vested equity.

In another 2014 article, my co-authors and I show that, in the months where CEOs have vesting equity, they release more news. This is an easy way to pump up the short-term stock price, as news attracts attention to the stock. This attention also increases trading volume, which allows CEOs to cash out their equity in a more liquid market. Indeed, we find that these news releases lead to significant increases in the stock price and trading volume in a 16-day window, but the effect dies down over 31 days, consistent with a temporary attention boost. The median CEO cashes out all of their vesting equity within seven days, so within the window of inflation.

The increase in news releases only relates to discretionary news (such as conferences, client and product announcements, and special dividends), which are within the CEO’s control, and not non-discretionary news (such as scheduled earnings announcements). Moreover, CEOs reduce discretionary news releases in both the month before and the month after the vesting month, suggesting a strategic reallocation of news into the vesting month and away from adjacent months. In addition to releasing more news items the vesting month, CEOs releases more positive news – media articles immediately following these news releases contain significantly more positive words than normal.

Why do we care?

Both consequences of vesting equity are important. Investment is critical to the long-run health of a company. In the 21st century, most firms compete on product quality rather than cost efficiency, for which intangible assets – such as brand strength and innovative capabilities – are particularly important. Building such intangibles requires sustained investment, particularly in R&D and advertising. Moving to news, many stakeholders, such as employees, suppliers, customers and investors, base their decision on whether to initiate, continue or terminate their relationship with a firm on news, or on stock prices that are affected by news. In addition to these efficiency consequences, news also has distributional consequences by affecting the price at which shareholders trade. Indeed, Regulation FD aims to level the playing field between investors by prohibiting selective disclosure of information. Public news releases to all shareholders achieve this goal – but CEOs may delay news until a time when they have vesting equity.

What can be done?

One solution is to lengthen vesting periods. While increasing vesting horizons from, say, three to five years may not be as politically alluring to voters as a rant about the level of pay, it will likely have a much greater effect on shareholder value and society. For example, such a change would incentivize CEOs to engage in a long-term investment with a four-year horizon.

Can clawbacks achieve the same thing, i.e. pay out a bonus for good short-term performance and then claw it back if long-term performance lags? Despite attracting much attention, the legality of clawbacks is unclear, and successful implementation is very rare. The CEO may have spent the money, or transferred it to a spouse or a relative. Trying to claw back a bonus that you have prematurely paid (based on short-term performance) is like shutting the barn door after the horse has bolted. The best solution is not to pay out the bonus until later on.

Is the lengthening of a vesting horizon simply kicking the can down the road? All equity has to vest at some point, so extending it by a few years won’t necessarily persuade CEOs to think and act in the long-term interests of a company. I have some sympathy with this concern – one of the other implications of the papers I’ve co-authored is that boards of directors, and other stakeholders, should scrutinize CEOs in the months (or years) when they have significant equity vesting. Since most of the current focus is on Levels 1 and 2 of the CEO’s contract, most stakeholders don’t pay attention to vesting horizons. But the main benefit will be on the behaviour of CEOs today – such a lengthening would encourage them to take that four-year project.

What all this research has shown is that by structuring CEO pay in a particular way, we can ensure they have the long-term interests of the firm at heart.

Author: Alex Edmans is Professor of Finance at London Business School and Wharton.

Image: A participant walks past a poster at the Asian Financial Forum in Hong Kong January 20, 2010. REUTERS/Bobby Yip

Further reading:
Edmans, Alex, Vivian W. Fang, and Katharina A. Lewellen (2014), “Equity Vesting and Managerial Myopia,” Working Paper, London Business School
Edmans, Alex, Luis Goncalves-Pinto, Yanbo Wang, and Moqi Xu (2014), “Strategic News Releases in Equity Vesting Months,” Working Paper, London Business School