Valuing bosses

Steve Jobs’ announcement that he’s taking medical leave led to an 8.3% drop in Apple’s share price. This suggests he is worth $26.5bn to Apple shareholders.
This is vastly more than even the best paid CEOs get, and suggests that Mr Jobs’ $1 salary is one of the great bargains of all time.
Which poses the question: could it be that bosses are actually underpaid?
Of course, it would be silly to infer anything from an outlier such as Mr Jobs. But we don’t need to do so. Matthew Sinclair has directed me to an interesting study which looks at share prices’ reaction to the sudden death of a top executive. The thinking here is that the shares’ reaction is equal to the contribution the boss is expected to make to the company in future, minus his pay. A price fall upon the death of a boss therefore indicates that the market thinks he generates more wealth for shareholders than he is paid, whilst a rise indicates that he destroys shareholder value. So, what do the figures say?:

Compiling a sample of 149 executives who suddenly died in the United States from 1991 to 2008, we find that, following death, stock prices drop by 1.22% on average. Since the average capitalization of firms in our sample is $1.5 billion, average firm value decreases by almost $18.8 million

Granted, there is variation around this average: in 67 of the 149 cases, share prices rose, indicating that the executive subtracted shareholder value. And overall, most of the value generated by bosses - four-fifths - flows to bosses, not shareholders.
Neverthless, this suggests that, on average, the managerial labour market is tolerably efficient. If anything, bosses are a little underpaid. However, I have three quibbles here.
1. Are bosses who drop dead suddenly really representative of average bosses? Possibly not. It could be that they die early because they work harder and neglect their health, or worry more about their company and so succumb to heart disease. If it is the case that the good die young, then the sample of dead bosses overstates the contribution that all bosses make to the firm.
2. Share prices might not fall merely because shareholders lose the boss’s skills. The death of a boss creates uncertainty about the future of the company, and shares must fall to reflect this. Because a man’s sudden death naturally unsettles us and makes us anxious, this uncertainty effect might be large.
3. This study only tells us about the market’s opinion of a boss’s contribution. But what if this opinion is systematically biased upwards? If investors believe in the myth of leadership, they will collectively overstate the value of bosses. In such a case, share price reactions merely capture the extent of belief in this illusion.
So, I’m not convinced that bosses are underpaid. But then, I’m not sure what sort of evidence would be convincing here. Whether this is because my prior antipathy towards bosses is strong, or because the evidence is inherently elusive, I’m not sure. 

Comments

[Andrew]

I haven't read the paper, but doesn't this need to consider who is likely to replace them? In the case of the positive impacts, it might not be that the CEO was bad, just that that markets think the heir apparent is better and are glad he's got the chance earlier.

[TheOneEyedMan]

Claim three should be testable. If it holds then you should be able to get out-performance over some holding horizon by buying the stocks of the dead CEOs. If you got the dates and the tickers that would be pretty easy to calculate.

In many ways I treat this finding as a lower bound on the value of the CEO. For one, I expect CEOs to capture much of their benefits to the firm in the form of higher compensation. For another, the replacement CEOs are also likely to be good, if not as good on average for firms where the last CEO was good. This number is really just the measure of the value lost by the firm by getting their second best CEO instead of their first.

[Funfoid]

The BELIEF in bosses (Myth of Leadership) is a critical variable. But business bosses are too often (unconsciously) conceived as people leading a war, holding the fort, making market skirmish, brandishing the brand flag etc.

I don't see any easy way to subtract the belief from the reality because of the obfuscation that the myth itself breeds.

In one case you may succeed but later cases will adjust to your prior success. Its an ART of obfuscation--a game of dodgems.

[Pete]

I think (2) looks like the decisive factor here. A man suddenly dying or disappearing at short notice is a very, very different thing to a an orderly transition.

I'm also unsure about the methodology here. Surely the problem isn't so much that whatever value the boss was providing has now been lost, but that no-one is doing a job that might be quite important to the smooth running of the organisation.

As an analogy, suppose some important piece of equipment unexpectedly fails at a firm. We wouldn't expect the economic impact to equate to the overall value of the equipment, but to the total value of everything they couldn't do whilst they replaced it.

[Harriet]

Share prices are all about belief. The experiment looking at how share prices changed after a change of bosses would tell you more about the beliefs of the participants in the market than it would about the actual impact of the particular boss. This is why extremely profitable companies sometimes have much lower market capitalisations and enterprise values than companies that have never made a profit and have no expectation of doing so in the foreseeable future.

[william]

The pay of a boss is like price - it's exactly what you pay for it, not some "intrinsic value". There's no high or low about it. If you can get a good boss for low pay, that's great. But you're more likely to get a poor boss for high pay.

[John H]

Another factor is that the sudden death of a CEO not only causes the loss of that CEO, but in the immediate term creates uncertainty - which markets abhor above almost all else, it seems.

So the drop in share price must be attributed in fairly substantial part to that short-term uncertainty rather than just the loss of the CEO.

That's backed up by the reference on p.15 of the paper to stock prices turning back up once an interim or replacement CEO is appointed, restoring a degree of stability and certainty.

With Steve Jobs, though, I'm sure a large part of his perceived value arises from the recollection of what a basket-case Apple became between his original departure and his return, and the fear that the same will happen again. In other words, it's a concern about the fragility of Apple's market position and the quality of its wider management.

[BenSix]

I still feel it must depend on how the bosses are perceived. If Carly Fiorina had stepped down from HP or, heck, John Sculley from Apple their shares would have taken off like Curd Jürgens after seeing Brigitte Bardot.

[Ted]

Apple's shares recovered the losses within a few hours (before the profits announcement). It's just noise.

[Adam]

There's some survivor bias there. You'd expect bad CEOs to be ousted quicker (or have their company taken over) and serve less time than good ones, so a CEO dropping dead is more likely to be a good one.

[charlieman]

"The thinking here is that the shares’ reaction is equal to the contribution the boss is expected to make to the company in future, minus his pay."

Isn't there a logic flaw there? The company needs to have a CEO. So the share reaction should, by that hypothesis, compare current CEO's contribution minus CEO's cost, versus replacement CEO's anticipated contribution minus replacement cost.

[Abdullah]

The interesting thing about my job – advertising – is that many CEOs often have their smarter, younger minions unearth insights about the business which they then tell to credulous hacks who repeat them, unchallenged, in their business columns. And thus the illusion of omniscience from a single Godhead is perpetuated. Share price might be affected if these CEOs left but no one within their companies would miss them. Funny, isn't it?

[McDuff]

Are we not also failing to consider that "the market" might not, in fact, know what the fuck it is doing from one moment to the next?