The Emergent Fool

Executive Compensation

Posted by Rafe Furst in Alternative Institutions, Economics, Incentives, Interventions, Management on April 4th, 2009 | View Comments

The main problem with executive pay is not that they are compensated too highly, but that there’s not enough pain for them personally when they do a bad job. I propose that the top three executives in all public companies be required to invest 100% of their salary in their own stock each year, with a decaying lockup period before they can sell.

Why 100%? Let’s face it, if you are one of the top three executives in a pubic company, there’s no reason, financially speaking, you should need guaranteed cash compensation. Having a safety net only serves to misalign your incentives with those of your employees and your investors, most of whom are at the mercy of your decisions. If you do need a guaranteed salary to pay the bills, go get a real job! Coffee is for closers.

How long of a lockup? The point of the lockup is to discourage short-term decision making. Long-term value investors typically have a minimum 3 year horizon, and illiquid venture-backed companies typically have 3-5 year vesting schedule on contingent compensation. In the past, you could have argued for an even longer lockup period, but with the pace of innovation today, I suggest a 5 year decay with 20% of salary-purchased stock becoming unlocked after each year.

What about benefits? Any non-contingent compensation (healthcare, company car, vacation pay, etc.) should not be allowed; it just serves to undermine the goal here. Again, being a CEO of a public company is a privilege you should earn by your past performance, not a God-given right. Remember all those years before you made it to the top when you were accumulating significant personal wealth? Put some of that aside to bankroll you through your tenure as top-dog. If you are good at your job, you will be hansomely rewarded via your equity.

What about other contingent compensation? You want extra stock or stock-option grants beyond your “salary”? That’s fine as long as it’s negotiated at arms length with the board of directors, and the grants are vested over a similar 5-year period. The key is that there should be no safety nets where you get paid while the other company stakeholders suffer.

What happens year after year? Every year, you are up for a renewal or change in salary. The board makes you an offer, if you don’t like it you leave. Over the course of time, you will have stock that is being released according to the 5-year clock but on staggered schedules. If you leave, you still get your locked up stock, but there is no change in schedule for unlocking it. You have to reap what you sow, but now your long-term fate is in someone else’s hands.

What about pre-IPO companies? What about them? Private companies wouldn’t change under this proposal. But once you file for IPO, the top three execs have to agree to the new compensation rules, including the 5-year decaying lockup for all equity and options previously granted.

Why 3? Seems like a reasonable number. Most of the decision making power is conferred to the top three executives. I can see the number being tied to market cap, so that while the minimum might be 3, GE might have it’s top 10 executives paid this way.

Isn’t this plan still too soft? Personally, I think it is. One could argue that the risk-reward ratio should be turned on its head so that it’s much less risky to start a new venture* which has huge value-upside and much more risky to take the helm of a established “blue chip” ship. With this line of reasoning, simply requiring executive salary to be invested seems like too much of a freeroll to me; I’d like to see some real skin in the game and require the top execs to reach into their savings and buy lock-up stock as well. But I’m happy to try the basic plan and see how it goes.

* I’ve put my money where my mouth is on this one [REDACTED 05/08/2009: see here].

Related posts:

Daniel

I'm definitely in favor of stochastic regulatory elements to prevent gaming the system, but as Kevin says, I don't think this it is ever going to be politically feasible.

@Ace, I think (some level of) BOD stock holding requirements are relatively standard now (at differing degrees.) But at the end of the day, all you have to do is buy stock with the money they give you, so I don't think it means much.
ace

Here is a real world example of one company's thinking...although this is for a board position not a top 3 job.

I have a buddy who has been invited to join the board of a large publicly traded Canadian Corporation.

He is required to buy shares of the company with a value of 5 times his annual compensation from the board. He is given 5 years to achieve that level and there is some mechanism for relief if you are too destitute to achieve it (perhaps you don't want someone on the board who is in this position though).

Now if the CEO were required weekly to do the same, as Rafe suggested in the top post it could get interesting. Instead of a paycheck, every two weeks he gets a statement of how many shares he acquired for that period. So if his "salary" was 20k per week and the share price averaged $20 during that week he would have earned 1000 shares. Due to dollar cost averaging he would be incented to keep the share price as low as possible until his last week on the job when he would want it to soar...and that might be a self fulfilling prophecy as shareholders would rejoice that he was finally leaving!
kevindick

@Rafe. Always with the algorithms :-) That would probably work in theory. But from a practicality point of view, I don't see your random draw method ever being politically feasible. I might as well wish for a truly libertarian state.

Because we've got a history of overlapping regulations from different agencies and jurisdictions, I think that's the most practical way to inject randomness into the process. If you could potentially be regulated by any one of 7 different agencies in 3 jurisdictions, it's hard to effect regulatory capture or figure out all the rules (given a reasonable background rate of change and regulator discretion).

@Schaef. Geddy who?
Scott Schaefer

Kevin should be so lucky as to one day be compared to the great Geddy Lee.
rafefurst

@Kev, I like your idea of "stochastically evolving" regulation that is one step ahead of the gamers, but I'm not so sure about messy. Is there a way to input a set if incentive objectives and algorithmically derive a diverse set of regulations that achieve those objectives? If so, we could have a yearly random draw lottery to see which companies get which regulation scheme. Every year new schemes are added to the mix and old ones are removed...
rafefurst

Whatever. You compare me to Marx, I compare you to McCarthy, you compare me to Hitler, and soon I'll be comparing you to Geddy Lee. Where does it end?!!!

:-P
kevindick

I'm just bustin' your chops Rafe. I think in a Platonic world, no regulation is better. However, I stipulate that at our current level of civilization, regulation is at least inevitable and may well be helpful on average. But I think the best policy is to have messy diversified regulation that changes a lot.
Scott Schaefer

I've had enough of posting on your damn blog --- why should Rafe get credit for all of MY insight?

Here are some thoughts on Kevin's "gaming" point:

http://utah-economist.blogspot.com/2009/04/fix-...
Scott Schaefer

I think we hit a nerve.
rafefurst

@"Shaef McCarthy" and "Kevin Bachus": burn me in Salem, shame on you, burn me in Congress twice, neat hat trick!

Let me put my argument in terms that an economist can understand. Instead of "assuming a CEO" and having companies and compensation packages compete for their business, how about "assume a compensation package for public companies" and let would-be CEOs compete to get hired. It's still free market, and there's no teat involved, just responsible oversight.
kevindick

Ahh, I get it now! I believe that you are right. I knew there was some crucial distinction between the discrete decision analytic view and the continuous portfolio view, but couldn't put my finger on it.

I thought you were saying that the CEO's diversification implied a difference in the shape of his utility curve in general versus shareholders. But it doesn't. It's the fact that the CEO is undiversified in his own company relative to shareholders that puts him on a different point in his utility curve with respect to discrete decisions made about his company (because he can't diversify the idiosyncratic risk). That sounds absolutely correct.

Let me just say that it's a real treat to have this discussion with a real economist. I can feel the economic sections of my brain growing once more.

And yes, Rafe is a pinko. Always wanting to suckle at the teat of the nanny state :-) (Shh! Don't tell Laura)
rafefurst

ROFL!

I agree with your normative descriptions, I just reject that we have to accept them as prescriptions. If the prospective CEO can't handle the risk and accept a package the aligns her incentives with the shareholders, then she can move to France. I don't see what's more capitalist and American than that.
Scott Schaefer

Rafe has always been something of a lefty pinko. Move to France. (jk)
Scott Schaefer

Kev, I agree that we're agreeing generally.

I also think I'm not being sufficiently clear about the point I'm trying to make.

In re-reading your April 8, 2009 at 8:46 post, I'm now thinking that we're mixing up beta and idiosyncratic risk I'm talking about the latter, and you're talking about the former.

Here's an attempt to clarify:

Let's suppose shareholders are well diversified. Then the only risk measure the shareholders care about is the firm's beta because that's the risk they can't diversify.

Now, consider an action the CEO could take that increases the firm's expected return, doesn't change beta, but increases the variance of the firm's return. That is, this is a change to the firm's strategy that boosts expected return but increases idiosyncratic variance.

How do risk-averse shareholders feel about this change? They _love_ it. Why? They can diversify the increased idiosyncratic risk, no problem. And they'll benefit from the increase in expected return. This boosts their expected return without changing their risk premium at all.

How might a risk-averse manager feel about this? The problem is that the risk-averse manager isn't diversified. You can't be diversified if so much of your wealth (not to mention human capital) is tied up one firm. So, a risk-averse CEO will dislike the firm's idiosyncratic risk. The new strategy increases the CEO's expected return but also increases his risk premium.

As a result, you can have situations where the shareholders want the CEO to do something that boosts idiosyncratic risk. But the CEO doesn't want to do it.

This can happen even if CEOs are far more risk tolerant than the average person (which I agree is likely).

If shareholders are more risk averse than the CEO, then they'll hold less market risk than the CEO. That is, the beta of the portfolio they hold will be less than the beta the CEO will hold. And I think this is what you're saying and it's 100% right. But the agency problem I'm identifying isn't a beta problem; it's an idiosyncratic risk problem. If shareholders are really risk averse, they'll really dislike beta, but they'll still want the firm to do anything that increases expected return at the cost of boosting idiosyncratic risk.

As a result, the shareholders might want to write a contract that shields the CEO from idiosyncratic risk, because doing so improves his decision-making. Such a contract might look a lot like a call option.

Does this sound right?
rafefurst

I agree with the gamability of non-optimal compensation schemes. However, if the alternative is to do nothing, that's worse.

Not all gaming has the same impact. As you point out, the better your model of incentive and risk profile, the harder it will be to game effectively, which is to say there will be diminishing returns for such activity. Additionally, the more direct loopholes you close, the more bizarre and detectable the gaming looks.

I am sympathetic to the argument that says we'd be chasing our tails and creating more and more baroque regulatory/legal structure (ala the tax code), however I look at it as the lesser of two evils. And it's not as tough a problem as taxation. The diversity of incentives and risk profiles of would-be CEOs of public companies is practically zero. And what I'm suggesting is somewhat of a forcing function to keep that diversity (and hence gameability) to a minimum.
kevindick

Well Rafe, a corollary to Schaef's objection is that any non-optimal compensation scheme (which will be all of them) will eventually start to be gamed. So you have to constantly improve them. This is not something I trust the government to do. It also argues for having a diversified portfolio of compensation schemes so they don't all fail in the same way at the same time. This argues against standardization.
rafefurst

And my view is that if measures like these can be shown to work, they should be mandated. The reason being that boards are notoriously conflicted with the top execs (and are often ineffectual in making bold big decisions, better at rubber stamping), and shareholders don't wield enough de facto power.
kevindick

I understand about the equilibrium implications of being risk neutral. But all the evidence I ever saw was that every agent is risk averse once the outcomes get large enough. Therefore, we're comparing the degree of CEO risk aversion to average shareholder risk aversion. Under these conditions, I believe my statement about portfolio betas applies.

Now, one could argue that the range where risk aversion comes in is vastly different for the CEO and an institutional investor. But that's not the end of the question. Rather, from a decision theoretic perspective, the question is how discrete probabilistic events affect each actor. I would argue that the impact of most events we care about scale with the actor's holdings. If that's the case, the _scaled_ utility curves may actually have very similar first and second derivatives.

We're saying the same thing about contract theory. What I meant to say was I was never impressed with what it had to say about constructing a precise optimal contract in a given situation. As you pointed out, there are a lot of factors to consider and in practice we can't possibly determine the values to use. Clearly, the qualitative results are useful.

Because I don't think you read the blog in general, I should point out that I'm anti-government intervention in general for exactly the reasons you point it. I was more agreeing that these sorts of measures were things shareholders and Board members should consider.
Scott Schaefer

@kd (some more)

With regard to how to use contract theory here... I think the value of it is in giving tools to assess the _costs_ of any intervention ex ante.

That is, suppose Congress passes a law extending vesting periods. That might be a fine idea. It would clearly be a potentially sensible response to what has happened. But, Congress is notoriously bad at anticipating adverse consequences of interventions. (Examples are legion, but see for instance my work on the Civil Rights Act of 1991.) There's probably a reason why vesting periods weren't longer to begin with, and mandating longer vesting periods would amount to requiring firms to incur whatever costs are associated with longer vesting. It may well be that the benefit is bigger than that cost, but we at least need to think hard about the cost before we proceed.

Contract theory can't tell us things like "Cost = $7." But it can tell us things like "Variance of CEO wealth with rise, and this risk needs to be compensated. Variance of CEO wealth will rise, which may change CEO choices when facing risky strategic decisions." These are side effects we should at least be thinking about. And that's what I was pointing out with my original reply.
Scott Schaefer

@kd

Some definitions: Risk aversion means you're willing to give up expected value in order to reduce risk. If you're risk neutral, then you make decisions thinking only about expected value and you don't think about risk at all. If you're risk seeking, then you're willing to give up expected value to get risk.

Suppose you're risk neutral and everyone else is risk averse. Because everyone else is risk averse, there is (in equilibrium) a premium for holding risk.

This is why the high-beta stocks earn high expected returns. To see why, suppose high and low beta stocks had the same expected return. High beta means high non-diversifiable risk, and so all the risk-averse investors would prefer holding the low beta stocks to high. This would bid the price of low beta stocks up (depressing their expected return) and push the price of high beta stocks down (boosting their expected return). The prices adjust until the marginal investor is indifferent between holding the high and low beta stocks.

What does this mean for a risk-neutral investor? A risk-neutral investor would want to hold the portfolio with the highest possible risk. The reason is that this portfolio earns the highest possible expected return (due to the fact that nobody else wants to hold much of that asset because they don't like the risk), and if you're really risk neutral then all you want to do is maximize expected return.

A truly risk neutral investor would want to put 100% of wealth in the riskiest possible asset, because that asset has the highest expected return.

This is why I was arguing that the fact that CEOs diversify their personal portfolios means they're at least a little risk averse; that is, willing to give up at least a little expected value to reduce risk.
kevindick

@Schaef. We're getting rather off topic, but I wanted to explore your point about the implications of CEOs diversifying.

Admittedly, it's been a long time since I seriously examined portfolio theory in the context of decision theory (it was probably back when you were always the 3 Man :-). However, I don't think your interpretation is right. Diversification only implies an optimizing agent--getting the best return for a given risk tolerance.

The implied risk tolerance would be the portfolio's beta. In fact, my recollection is that we tried to estimate utility curve derivatives using betas collected from the portfolios of various demographics.

Therefore, we'd want to look at the betas of CEO portfolios versus those of their companies' stocks versus those of the average shareholder. That actually seems like something we could do fairly easily. Of course, the philosophical thinking about risk could have changed in the last 20 years. If so, "I'm willing to learn," in the immortal words of Bill Murray.

I do take your point about CEO compensation in general. I was never impressed with the practical applicability of optimal contract theory. But we could probably figure out what the first 3 to 5 most sensitive terms are and do better than what we've got now.
rafefurst

@Jay, I guess what I would say is that if your company loses money, you don't get paid, period. It's your job as CEO to turn the ship away from the storm. If that means diversifying and/or getting out of a bad industry, or if the long term seems like too long for you, tough noogies. Somebody's going to be willing to step up to the challenge. And if not, that's a good sign that everyone should abandon that particular ship.
Jay Greenspan

The thing that kills me about Blankfein and others is that they only managed to come to the realization that executive pay incentives often don't match those of shareholders a year into this global crisis. The fact is that institutional investors have been clamoring for changes for years. Accountability for CEOs and boards (and the relationship between the two) has been a regular complaint of pension funds, some mutual funds, and large investors like Karl Icahn. Look at this quote form the article. It's absurd:

"Much of the past year has been deeply humbling for my industry," Blankfein said, acknowledging it could take years to rebuild the investor confidence lost in the crisis caused partly by industry practices that appear "self-serving and greedy in hindsight."

Hindsight. Right.

This is one of these things that just drive me up a wall. So much of conversation among the elites on topics of finance were carried with a specialized vocabulary and a condescending view of anyone who doubted their brilliance. I'd been in more than one conversation with bankers, where the underlying message was crystal clear -- "poor dear, he just doesn't get it. It's just beyond him."

It turned out that in this case, like so many, while there might be all kinds of complexity in the system, some concepts are pretty simple. Such as: Perverse incentives lead to crappy decisions.

Guess the above was more an unhelpful rant than. Sorry about that.

@Rafe: I think you're pretty much on the right track, but feel any compensation package would need to be tied to industry and corporate specifics as well as the general economic climate. We can easily think of examples where a CEO performed heroically while the stock lost significant value. (Look around. Everyone's lost, but someone must have done a good job over the last year.) The converse is true as well.

How would you compensate the CEO of a large pharmaceutical concern? Most of the big ones are expecting big drops in revenue as patents expire. There's nothing to take the place of these drugs in the pipeline. It can take 10+ years to bring a new drug to market.
rafefurst

Very interesting, sound like Blankfein was reading this blog:

"For senior employees, most of the compensation should be in deferred equity."
"Individual performance should be evaluated over time to avoid excessive risk taking. All awards of stock should be subject to future delivery or a deferred ability to exercise them over at least a three-year period."
"Senior company officials should be required to keep the bulk of the stock they receive until they retire."

While I agree with Scott that what people do and what they say they will do are often different, it seems promising that a CEO came up with the same recommendations that appeared here by non-CEOs.
Jay Greenspan

Woops. Here's the AP report on the same talk:

http://tinyurl.com/cxrkqk
rafefurst

@Jay Greenspan, you have to be a member and pay to read that...
Jay Greenspan

Some thoughts on compensation from the CEO of Goldman (also notes on Too Big to Fail)

http://online.wsj.com/article/SB123911343792496...
Scott Schaefer

Kev, you're for sure right that CEOs are likely more risk tolerant than the population at large. But I think there's evidence that CEOs still diversify their personal portfolios --- and diversification only makes sense if you're risk averse. And any amount of CEO risk aversion will lead to at least the possibility that CEOs would behave in a risk averse manner when that's not what shareholders want.

I'm not arguing that "options are a good thing" but I am arguing that "it's not immediately obvious that options are a bad thing."

It's very hard to assess risk preferences. We can ask people about decisions they'd make, but it's well documented that what people _predict they'll do_ and what people _actually do_ are often different. Have you seen the risk aversion studies that use "Deal or No Deal?" --- There was one in the AER recently. That's a nice high-stakes experiment, and we'd need to do the same thing on a sample of CEOs to answer this question. (We'd need a _seriously_ large research grant.)

My general thought on CEO pay --- and I have been meaning to write this down formally --- is that our theories don't pin down the data enough for us to be able to make sensible statements about what the "right" form of pay is. Pay should depend (according to theory) on CEO outside options, the marginal productivity of the CEO's effort, the mean/variance properties of the firm's investment opportunities, the CEO's degree of risk aversion, the second derivative (yes it's true) of the CEO's disutility of effort function.... and so on. And we don't observe any of it.

And what this means is that for just about any statement along the lines of "Pay should look like X not Y," it's very easy to come up with a combination of those unobservables for which Y is a better way to pay executives (for shareholders and the economy at large) than X.

The limits of social science are depressing, I know.
kevindick

Schaef, I take your point about the time horizons. If an executive's expected tenure is small enough and consumption discounting is large enough relative to the expected inter arrival time of a bursting bubble, there may be no way to solve this problem through timing of payments. You'd need some sort of large potential penalty assessed ex post to overcome the problem you describe.

However, I quibble with the options risk alignment point. I think a corollary to the tournament theory of compensation is that CEOs get selected for unusually high risk preference. Options would accentuate this potentially undesirable feature. A behavioral study of CEO risk preference would be very interesting. I couldn't find anything after a few minutes of Googling.
Alex Golubev

I love the suggestion, but the systemic risk is caused by leverage (i completely agree that shareholders get screwed over by call options in executive comp). So i would be more concerned with having this implemented in high leverage entities (all banks and hedge funds...maybe utlities?). I wouldn't exclude private partnerships. Otherwise that's where the system will reestablish itself. any entity with more than 100 individuals or certain size should be subject to this. no company profit sharing networks either (those financial engineers will work so damn hard to beat the comp system!).
Daniel

I think your plan may be missing some key variables. Should we not take into account the executives prior networth? If I have 100 million prior to taking the job, then I may not feel so much pain upon losing ("freeroll.") My prior income has guaranteed me a comfortable living for the foreseeable future. I do not need this income, and I will not necessarily feel tied to the future success of the company. Since I don't need the money, if things are bad, I might just dip out after a year or two. This compensation has marginal utility no matter how you slice it. So how does expected compensation/networth ratio affect performance? And, on the other side of the spectrum, how do we view people with little or no networth? They seem to have little to lose and everything to gain, but can someone with little to lose be trusted? It seems that expected comp/networth should be taken into account. 1 mil for someone with 1 mil is different than 1 mil for someone with 10 mil and *extremely* different for someone with 100 mil.

Seems like one of the solutions you mention, to require a significant investment or wager of one's existing networth is necessary to further tie one to the fate of the company, and allow for a better risk/reward profile. If the CEO is willing to wager/invest 25 million of his own money in the company/stock, then we can conceivably offer him more significant rewards (or not. The rewards will come as he increases the stocks value.) Either way, this serves to better align incentives. (One must bet a portion of their past success on their future success)

I also think one should be able to draw a minimum salary and elect to receive certain benefits (deducted from salary, up to something really small like 1-2%) as this confers tax advantages that are not available otherwise. But I agree, since benefits are normally estimated to be 10-30% of "compensation," they should be eliminated, and any benefits should be directly and accurately deducted from compensation. I'd rather rewrite the tax codes, but this seems unlikely.

Also, five years seems about right as over time (holding compensation constant) we reach a 2-1 unvested/vesting ratio. And as the world becomes more extreme longer vesting periods become less and less practical.

@Scott, whatever the vesting period is, there will always be execs who have been working long enough to be vesting a significant amount of comp that CDO/CDS are appealing. I think the best solutions to this are to force the exec to separately invest their own money, which in the case of Stan O'neil would have become worthless. This would offset the gains he made from risky behavior and possibly discourage it. We might also require execs to leave some of their own money invested after they leave the firm. (So they don't burn the firm to the ground before leaving)
Scott Schaefer

Hey Kevin!

I haven't seen it yet --- but then I haven't been to a theater in years... But I did wear my Geddy Lee Henhouse t-shirt (which I caught when thrown from the stage in Boise last summer) on Friday.

It's not clear that options "misalign." Options clearly reward risk-taking, but we'd probably expect that execs would be more risk averse than well diversified shareholders. As a result, insulating the exec from downside risk can serve to counteract the exec's natural risk-aversion, and get him to look at risk more like shareholders do.

This is why it's not immediately obvious that forcing execs to share the pain when things go bad is such a good thing. It's clearly good for motivating the exec to do things that increase the expected value of the share price but don't change the variance. But it's potentially bad for motivating a risk-averse exec to do things that increase both expected value and variance.

Here's where my horizons thinking is coming from. If we ask about the specific mistakes that were being made at the top of the big banks, I think it's this: Bank CEOs knew that a housing bubble pop would mean the loss of billions in shareholder wealth. And they knew there was some probability of this happening. But they also knew that they had bonuses and equity lockups that were coming due in the near term. And these were huge in dollar terms. So, a bank CEO could either get out of the CDO/CDS game --- which would mean falling earnings relative to peers, falling stock prices, an increased likelihood of getting fired, and a much smaller personal payoff coming from the bonuses and near-term equity lockups. Or, a bank CEO could continue to play, in which case he just hopes that he can cash out with his zillions before the music stops.

It seems to me that Rafe's suggestion wouldn't have led to behavior that was terribly different from what we saw. Think about the equity lockups that would be ending in 2006 for Merrill's Stan O'Neill, under Rafe's plan. He'd have had stock from 1/5 of his 2001 salary, 1/5 of his 2002 salary, 1/5 of his 2003 salary, 1/5 of his 2004 salary, and 1/5 of his 2005 salary becoming vested. Salary plus stock appreciation over that term --- we're talking about tens of millions of dollars. Again, there's a strong temptation to stay on the CDO/CDS path, and take your profits.
kevindick

Hey Schaef. I was just thinking of you given the major role of Rush in "I Love You Man" :-)

I would argue that Rafe's proposal is quite a bit different from the status quo. First, as you know, options are definitely very different from stock in terms of the incentive for risk taking. They misalign executive and shareholder interests.

Second, stock on top of salary _may_ also be different from straight stock. I hypothesize the endowment effect is much stronger with the latter. It would be an interesting thing to test. If my hypothesis is true, Rafe's plan will decrease risk taking. Not sure if that's good or bad but certainly brings the executive's psychological position with respect to the company more in line with that of shareholders.

Third, I'm not sure if your analysis of the time horizons is correct. From a decision theoretic perspective, it seems to me that the salient question is not whether the executive has the incentive to look beyond the horizon of the potential next bubble. This could be infinitely long of course.

It seems to me the paramter that will determine the executive's behavior is the probability that his compensation will be materially affected by a bubble bursting (relative to his expected tenure in "good" times). Once an executive reaches a point where he gets yearly payouts the chance is perhaps 1 in 10. But Rafe's plan could push this probability to 3 in 10 or even 5 in 10.

I'd love to see a professional analysis of what this change in probability of being affected by a bubble does to expected behavior.
Scott Schaefer

I agree with your general diagnosis that short-termism is the root of the problem.

But this problem is very hard to fix.

First off, 3 years => too short. This wouldn't have prevented the financial collapse, since the gestation period of the housing bubble (and the dot-com bubble before that) was about 5-6 years.

Plus this isn't terribly different from the status quo, where the big payouts are coming from stock and option grants that have 3-5 year vesting requirements.

On top of that, any attempt to lengthen vesting periods or put in clawback provisions will increase risk on executives and risk must be rewarded so overall levels of pay must rise. And Joe the Plumber --- not mention House Republicans --- will be miffed when he sees that!

Further, any time you increase an exec's downside risk, you limit his incentives to take risk in the first place. And I'm not sure how that fits in with "encouraging innovation."

I'll give some different thoughts on this problem my blog sometime in the next couple of weeks. I have some different but related ideas.
rafefurst

I like this tweak as long as the paycheck is delivered on the regular schedule in hard-copy form to the CEO's desk and not automatically deposited :-)
kevindick

I'd only make one suggestion. The CEO _would_ be allowed a guaranteed salary. But it would be limited to the median salary at the company.

Because this is probably a relatively small amount of money, I don't think it would impact the performance incentive much. But it would make him/her pay closer attention to the overall compensation posture of the company which I think is a reasonable goal for shareholders to have.