3 ms·
This may be a correlation-vs-causation effect. It's possible that companies that are in poor markets and likely to do poorly in the next few years have to pay
by nostrademons 12y ago
This may be a correlation-vs-causation effect. It's possible that companies that are in poor markets and likely to do poorly in the next few years have to pay more to attract qualified CEO candidates, because good executives realize the shit the company is in and don't want to take the risk of a failure on their resume without being compensated significantly for it.
This would also explain why the negative correlation is strongest at the extreme high ends of the pay scale - these are the companies that are in the worst shape - and why these CEOs relied heavily on acquisitions, because they had no talent or useful products in the organization itself. It also fits with Warren Buffett's observation that "When a management with a reputation for brilliance tackles a business with a reputation for bad economics, it is the reputation of the business that remains intact."
One way to confirm or deny this hypothesis would be to look at companies that previously were high-performing, and then see what the correlation between CEO pay and the change in market performance is within that group.
- scott_s 12y agoI thought of this as well, and I was bothered that the authors did not. (At least the author of the Forbes piece - I could not access the research paper.) The author of the Forbes piece immediately jumped into explaining the effect assuming causation, which I was not convinced has been established. To be clear, I find it reasonable that there could be causation. But we still need evidence to support that conclusion.
- notahacker 12y agoThe paper itself covered a lot of ground (including "firm size and prior stock performance, especially over the past three years, are significant predictors of both excess cash and incentive compensation"; the results confirm than recent stock performance is positively correlated with CEO pay). It didn't however give any detailed consideration to the obvious other hypotheses: (i) Far from being overconfident about their decisions, top paid CEOs (and the boards that negotiate their pay packet) are actually more conservative than the market about the future prospects for the stock, so are willing to agree bigger stock-based incentive packages as a result. CEOs generally should have more insight into the workings of their firms than Wall Street analysts, of course... (ii) If a CEO's pay packet at time t predicts good stock performance since t-3 and bad stock performance up to t+3 might it be that their ability to prevent the firm's stock price reverting to the mean is limited (even if they're actually still doing a good job... and receiving acknowledgement in the form of pay packets more focused on that difficulty) There's also [iii] The decisions the CEO makes is not even slightly affected by their pay structure but high CEO compensation packages is a good proxy for lack of board oversight but that's a lot closer (and to a large extent compatible with) the authors' hypothesis about overconfidence
- rfergie 12y agoMaybe regression to the mean also explains it? Firm has a good three years so CEO pay goes up. But that performance was just random chance and the true mean (and future performance) is lower
- phaus 12y ago>This may be a correlation-vs-causation effect. It's possible that companies that are in poor markets and likely to do poorly in the next few years have to pay more to attract qualified CEO candidates, because good executives realize the shit the company is in and don't want to take the risk of a failure on their resume without being compensated significantly for it. I was also thinking that it could be the correlation vs causation effect, but I think it may have more to do with the fact that the highest paid CEOs are usually going to be found working at gigantic companies that are struggling to continue to grow.
- greendestiny 12y agoYes I think this is a simpler correlation. Bigger established companies have more cash to pay the CEO but less room for growth. The companies themselves are probably more concerned with minimizing risk and holding position - it might be a better study to compare CEO pay with board evaluation of performance.
- Ygg2 12y agoI'd argue it's not. People that are paid most to do a job that isn't repetitive manual labor have worst performance scores, time and time again. So unless CEOs are paid to manually bolt things or throws ball into a hoop, they probably aren't doing well because, in part, they are given money. Money introduces fear (i.e. fear that you won't make enough money or lose your position with lots of money) and that leads to tunnel vision. People give more performance if you give them more autonomy, mastery and purpose. https://www.youtube.com/watch?v=u6XAPnuFjJc https://www.youtube.com/watch?v=u6XAPnuFjJc
- Fuxy 12y agoI would argue the opposite with so much money on at your finger tips you are more prone to bad decisions because you are not invested enough in the outcome. You ear more in a year than you can spend in a life time what are the chances of you looking a every deal very attentively? You do the absolute minimum to maintain your position but you're not driven enough to put more effort into it.
- mcv 12y agoI also immediately though of that talk. It makes sense that it also applies to CEOs.
- netcan 12y agoI agree. It's unfortunate that whenever you see media coverage of a study like this, the safe assumption is bad interpretation of the data, bad data, small sample sizes and the like. It could also be true. We're such apes. You feel like you need to interrogate them to see what they're missing or hiding. Is this just a product of pay in certain industries? The article mentions poorly performing banks. Bank CEO markets are competitive. There are a lot of them. They move and they have options in government, think task and such. Banks also happen to be on a bad run. Tech OTOH is doing well. We're on a founder-CEO trend and most of them are rich of founder stack, not salaries. Stock is more common as a compensation. You're more likely to find $1 salaries in tech companies. Could the story really be more about industries, their compensation norms and recent performance? Could it be about founders vs hired guns? Could your (sounds plausible) reasoning be true, bad companies need to pay CEOs higher salaries. I'm not claiming it is, just pointing out how unfortunate it is that we can't say "Forbes, University of Utah. They must have eliminated those. Their conclusion must be a likely one. "How could this be? In a word, overconfidence. CEOs who get paid huge amounts tend to think less critically about their decisions." - really? On second though, your reasoning must be correct. Stock based compensation is much more attractive in a company/industry that's making gains. Warren Buffet or Larry Ellison can be compensated by owning stock, options or similar. If a stock is flat, those kinds of incentive packages are worth less so it needs to be made up in salary.