Why has CEO pay exploded?

In the first chapter of Predictably Irrational, Dan Ariely explains how in 1976 the total compensation for the average chief executive officer (CEO) in the United States was about 36 times that of the average worker in their companies. By 1993, it was 131 times and public outrage was growing.

So the SEC issued a new rule requiring companies to disclose CEO pay. The bureaucrats believed that as soon as the disparity was visible to shareholders, the market would fix the problem and excessive pay would be competed away.

It didn’t quite work out like that. By 2007, the average CEO made 369 times what the average worker made. Although there’s far less data available in Australia, a 2009 Productivity Commission report claimed that in 2008/9 the CEOs of Australia's Top 20 companies earned about 110 average weekly earnings. The trend is similar, if less pronounced.

But why even worry? If an increase in CEO pay accords with the increase in value of the businesses they manage, a King’s ransom for a CEO is still a tidy sum for shareholders, right?

The problem is that the rise in CEO pay vastly exceeds the increase in the value of the companies they run. Which means that executive remuneration, to give it the fancy term, has increased at the expense of shareholder value, not in the service of it.

How can this be so?

To put it bluntly, CEOs are rent seekers using flawed economic theories to justify huge pay packets. [Don’t feel so outraged. The same economic theories predict we’d do the same thing in their shoes—Ed]

The reality, according to academics Lucian Bebchuk and Jesse Fried, authors of Pay without Performance: The unfulfilled promise of executive compensation, is that directors have ‘various economic incentives to support, or at least go along with, arrangements favorable to the company’s top executives’.

In Australia, CEOs play a big role in director reappointment. Why would they want to reappoint someone that has tried to restrict their pay packet? Directors that haggle over CEO pay tend not to be invited back and their ‘awkward’ reputation soon travels.

Directors know they have to work with the CEO come what may, so why ruffle feathers? And approving a pay rise for the CEO makes an increase in director remuneration that much more likely. The arm’s length theory takes no account of these all-too-human and very powerful motivations.

Any examples?

Penrice Soda chairman David Trebeck, after 38% of shareholders voted against the company’s remuneration report, offered up a recent example when he complained that senior executives had not had a pay rise for two whole years.

The fact that his company’s share price had fallen by more than 90% over the period seemed not to matter. A clearer example of the separation of pay from performance is hard to find, which is probably why Penrice directors recently called for the abolition of the ‘two-strikes’ rule, where a company board can face re-election if shareholders disagree with remuneration policies.

Research also backs up the proposition that boards support CEOs over shareholders. US research by Core, Holthausen, and Larcker found that CEO pay tends to be higher when the board is large because it makes organising against the CEO more difficult.

It’s also higher when directors are appointed by CEOs because it creates a sense of obligation, and when external directors serve on three or more boards (they’re too distracted).

Cyert, Kang, and Kumar found that CEO pay is between 20% and 40% higher if the CEO is also chairman of the board. Take note investors in News Corp, where Rupert Murdoch holds both positions. ASX guidelines say that the same person should not hold these roles, a view lost on the late David Coe and Ken Lay at Enron.

The same academics also found that there is an inverse relationship between CEO pay and the share ownership of the board’s compensation committee.

Why doesn’t the publishing of CEO pay push wages down?

Because everyone knows what everyone else is earning, and boards want to be seen to pay for top talent, appointments tend to be made at above average salaries, pushing salaries up further. There is a market impact, but not the one predicted by the SEC.

Known as the Lake Wobegon effect, named after broadcaster Garrison Keillor’s fictional town where ‘all the women are strong, all the men are good looking, and all the children are above average,’ it explains how, under public ownership, QR National paid its CEO about $1m per year but two years later the publically listed Aurizon CEO now pockets $6.8 million. That’s the price of keeping up with the Joneses when you can peer into their pay packets.

Is that the only explanation?

Not entirely. The idea that the interests of a CEO should align with those of their shareholders is quite plausible, so tying CEO pay to share price performance makes apparent sense.

But Roger Martin, Dean of the Rotman School of Management at the University of Toronto and author of Fixing the Game, argues it’s not sensible at all.

Martin draws a parallel between business and sport. Both operate in two worlds; the real one, where products are made and sold and games are won and lost; and the world of expectations, where bets are placed on who will win and which company’s stock will rise or fall.

The problem is that CEO pay, through the issue of options, is actually tied to the expectations world, not the real world. All too often stock prices aren’t based on actual products or profitability but on expectations of future company performance.

That offers all sorts of incentives to CEOs trying to get their options ‘in the money’, explaining, for example, when a company’s stock is floundering huge writeoffs are more likely and why CEOs rarely dampen unrealistic expectations around high share prices.

Tying CEO remuneration to a stock price, argues Martin, is like a sportsman betting on his own games. In most sports, that’s illegal. In business, it’s called ‘aligning incentives’.

Any more examples?

Cochlear CEO Chris Roberts already owned $50 million in shares before the board granted him 231,161 options at a price of $62.78 each, exercisable in August 2015. By the time shareholders got to vote on the package, the shares were trading at $72.32 and Roberts was sitting on a $2 million profit.

A product recall in the previous year meant that the earnings per share hurdle that triggered the issue would almost certainly be met. Outraged shareholders voted down the company's remuneration report.

Chairman Rick Holiday-Smith had something to offset the embarrassment of being responsible for the deal. He took home $453,775 in director fees for the year, plus another $326,694 as chairman of market regulator ASX.

What’s the real solution?

Whilst Australian companies suffer from the same problems of growing CEO pay and a breakdown in the link between pay and performance, according to the Productivity Commission the problem isn’t as bad here as it is in the US or UK.

The ‘two-strikes’ rule also gives shareholders more power to resist outlandish remuneration, although it requires reluctant fund managers to be more active and engaged.

In the end though, existing policies don’t target the root cause of the problem. As Roger Martin says, ‘The true key to long-term sustainability is building customer and employee bases that enable long-term profitability. If we are to emerge from the current mess, executives must switch their focus entirely to the real market and completely ignore the expectations market. Management should not, and in fact cannot, protect the interests of those who buy shares on the open market at prices that are purely a function of expectations.’

Note: The Growth portfolio owns shares in Cochlear and News Corp, while the Income portfolio owns shares in ASX.