Fixing Executive Compensation

Executive compensation was under heavy fire a while ago, and came up today in a debate with a friend.

The level of executive compensation has been rising over time. It was argued that this is a leech of society, just as much as unions are (which I'll blog about soon). The issues with executive compensation actually parallel those of unions, a little bit, so I'll reference unions near the end.

The first point to make about executive compensation is the more general, less problematic one: why are they so high, and why are they rising?

Executive compensation is subject to positional externality. Most commonly, this is the "keeping up with the Joneses" effect. If the Jones family buys a Volvo, you need a Mercedes to be the nicest car on the street. When you buy a Mercedes, they need a Maserati to be the nicest car on the street. In order to take the number 1 position, the stakes keep getting higher and higher until it's no longer worth it to compete.

This is also observed in higher education. It used to be the case that well over 60% or 70% of jobs could be gotten right out of high school; now, a much larger percentage require a college degree. Part of this is the "technocratization" of society - everything is more complex, so smarter workers are more important - but part of it is also a positional externality effect. If you and I both apply for a job, and you have an associate's degree and I don't, then you get the job. To top you, I need a bachelor's. To top me, then, you'll need a master's.

This also happens with firms. As firms compete to take the number one spot in a market, they need the best management. In order for me to have the best management, I need to pay the most. Then my competitor has to respond by paying even more to get even better management (or lure mine away!). And I need to respond, etc. until it's no longer worth it to pay more for better executives.

Thus, it should be reasonably apparent that the more competitive the environment, the higher executive pay climbs. The US is generally considered to be the most competitive marketplace in the world, thanks to free market principles, a high level of technology and overall education. To succeed in America is much more difficult than succeeding in most other countries, because there are so many people with talent. This is a good thing, generally, and results in innovation. There's a reason why the much-more-competitive US has outgrown a number of lower GDP per capita countries, flying in the face of every major macroeconomic model that predicts country convergence (you slow down your growth as you get bigger). Setting aside the fact that I think country convergence is a very limited theory that's used in a number of inappropriate contexts, the US' competition level has benefited us all immensely.

Competition has led to a number of things. The higher level of competition means we have the highest executive compensation in the world. We do, in fact, get what we pay for. The US scores highest in the world on the majority of metrics related to management performance. This is especially due to the tails - we have a lot more superior managers, and we have almost no genuinely incompetent CEOs, because their companies die or fire them way too fast. Compared to other countries, this lack of a lower tail has been very advantageous for stability and growth.

The US has also almost certainly been getting more competitive over time as people acquire more skills, which means you should see management performance rising as superior CEOs have to stand out even more and the benefit of having a superior CEO continues to rise. Compensation rises in lockstep. This seems, empirically, to have happened.

Thus, the idea that CEOs in America should be paid exorbitant sums of money is not intrinsically a bad thing. CEO can be a very, very hard job at a large company, both in terms of how difficult it is to succeed at it and also the brutal lifestyle (the hours, the travel, the scrutiny, the legal liability, etc).

CEO is not a particularly secure job - the average CEO tenure is about 8 years. You compete, or you get fired. You're paid by your shareholders to perform, and you rarely have another shot at a top job if you don't succeed. Your pay in your tenure has to last you. This is another reason why compensation ends up high - it's a job with absolutely no security.

This is complicated slightly if it's hard to tell who is a good CEO and who is a bad CEO, and there are asymmetric returns to a good CEO and a bad CEO. If talent isn't obvious but you gain more from a good CEO than you lose from a bad one, you see pay rise as people pay extra for an extra shot at "winning" by finding a good one. This explains a lot of Wall Street pay - a rainmaker can make your firm, and if you're wrong it's not catastrophic. Conversely, if you lose more from a bad talent than you gain from a good one, pay gets deflated. Music or movies come to mind, where it's hard to see who has the necessary charisma to do well, but finding a new star isn't as helpful (you could just hire someone you know is good for 10 million), as flopping is bad (hundreds of millions of dollars lost), then you don't pay very much to people who you're not sure about, which is why there are so many starving artists.

Everything we've talked about until now is highly compatible and perhaps even a prerequisite for a dynamic society - you WANT competition, competition is good, and you want talented people working hard to succeed, because people benefit from that. Even for CEOs who don't have large ownership stakes, talent is good. Alan Mullaly comes to mind - he took over Boeing, turned it around, and it went from serious distress into the largest and most important aircraft manufacturer in the world. He is now doing the same thing for Ford. We all benefit far more from Mullaly's success than we lose by paying him - in employment, in technology, in economic progress, etc. As part of a loosely-defined social contract, everything up until now is a good thing.

(EDIT: I've been thinking about this more. Classical positional externalities are generally considered negative things - keeping up with the Joneses by having both of you spend on things that don't matter is clearly welfare destroying. However, this case is not a typical positional externality, because if high wages widen the pool of potential CEOs and raise the expectations for what they will be able to accomplish, then the positional externality effect can actually be positive, in the same way that having another superior salesman in an organization can make his/her teammates work harder and become a better overall sales staff- a status that is no longer positional. If better people want to become CEOs - actually a concern, because of the long hours, travel, high stress, short job tenure and legal liability of the CEO position - and they work to reach a higher set of expectations, that's a good thing. Especially given that it's almost impossible to preserve rank-ordering with any externally-administered "solutions", and it's a side effect of a very beneficial process - competition - it's on net a good thing.)

However, job security has value, so if you gain job security, it's not a good thing if your pay stays as exorbitantly high. This is where executive compensation has a problem, and also how it compares to unions.

You see, executive compensation is also set by corporate boards. Corporate boards are ostensibly elected by shareholders, but the way corporate boards work right now in the US, executives themselves can serve on boards, executives can influence the salary given to boardmembers, boardmembers don't need to think like owners, and any shareholder who does not vote his or her shares has his or her shares voted by the CEO and the board.

Needless to say, this creates a reciprocity between the CEO and the board that can be very destructive for shareholders. The board can set whatever "targets" it likes to help the CEO get paid. Shareholders largely care about the price of their stock, which means that as long as the stock is doing well, the shareholders won't demand the CEO be gone. Thus, a CEO doing a bad job but who is the beneficiary of a rising stock price can remain in power. Thus, you're adding a job security to being a high-paid CEO that isn't good - it keeps bad CEOs in power, removes accountability and victimizes shareholders who let themselves be lulled by short-term results.

THIS is a destructive situation - a bad CEO should be fired as soon as a better prospect is there to replace them. The effects of competition get delayed by a number of years, which hurts everyone. It's not neverending - a bad CEO will eventually be fired when the company runs into trouble - but it can take a while. Society pays for this delay.

Unions are the same way - good employees should be paid well, and probably have a number of perks. Bad employees, however, should not be protected at the expense of taxpayers, consumers and shareholders.

The union problem tends to be bigger for the simple reason that the CEO's hold on the board is looser than a union's hold on a company or government organization. CEOs will eventually get fired, while union concessions have a tendency to grow until the company starts dying - which, in the case of a government organization, is an almost-never kind of event. So unions swell as a problem over time, while executive compensation is a much more constant problem.

That said, the board-executive reciprocity situation is still a negative one. I've listed a number of factors here: http://tfideas.blogspot.com/2010/03/perfect-government-in-trevors-world.html.  The biggest factor is to force the board to think like owners instead of like the CEO's buddies. The CEO should not be allowed to be the board's chairman, the boardmembers should probably not be paid and they certainly should be forced to have large personal positions in the company. Proxy voting (anyone who doesnt vote is voted by the board) should not be legal. Elections should be frequent, with publicity for board challengers, and compensation formulae should be a) public and b) very long-term. If this means paying the CEO after he or she is gone, fine, but the CEO should get paid based on the results of that year's accomplishments over the next few years, instead of just a snapshot-in-time approach.

I'll write more about unions later, but for those looking for a primer on the executive compensation debate, this hopefully is helpful.