The Power of Incentives

October 2017

There are two ways you can get a mule to move: with a stick or with a carrot. You could punish the mule by hitting it with a stick until it moves. Alternatively, you could encourage the mule to move by dangling a tasty carrot in front of it. Either way, if you want it to move you need to provide it with an incentive to do so.

The same line of thinking applies to how CEOs are compensated. Sure, a CEO must be paid some base salary in order to encourage him or her to go to work every day, but if a board wants to encourage exceptional performance, then some sort of variable bonus is often employed.

The following chart shows just how strongly boards feel about variable compensation. For CEOs of companies on the S&P 500, only 10% of total compensation takes the form of a base salary. In contrast, almost one-third of compensation is derived from long-term incentive plan (“LTIP”) performance, while one-fifth is short-term incentive plan (“STIP”) performance.

On the surface, this type of incentive structure would seem to do a good job of encouraging CEOs to work harder for the benefit of the company. But all too often these incentives simply allow executives to get richer while company performance falters. In other words, the mule has figured out how to eat the carrot without moving at all.

The following chart, compiled by Goldman Sachs, recently caught our attention. On the horizontal axis, various LTIP metrics are shown. For example, EPS refers to earnings-per-share targets that CEOs must achieve in order to be paid a portion of their LTIP bonus. On the vertical axis, average total shareholder return (“TSR”) compound annual growth rates (“CAGR”) are shown. Basically, a positive value means that the share prices of a particular category of companies have outperformed the shares of a broader peer group.

This chart shows that there is only one performance metric that has correlated with share price outperformance: return on capital. CEOs who are compensated based on any of the other metrics in that chart collectively saw their companies’ shares underperform the shares of companies whose CEOs' compensation was not based on the same metrics.

This finding came as no surprise to us. A CEO’s job is to allocate capital to projects that generate profits. The greater the profits, the greater the return on this allocated capital. Thus, measuring a company’s return on capital is an excellent way to measure a CEO’s effectiveness.

In contrast, all of the other metrics in that chart can be increased in a manner that does not benefit shareholders over the long-term. Want to increase revenue? Simply acquire another company without worrying about the price paid. Want to increase cash flow? Just stop investing in machinery and buildings such that they fall into disrepair.

Unintended Consequences

This isn’t a criticism of CEOs. This is simply an observation of how incentives can shape human behaviour. As the author Upton Sinclair wrote, "It is difficult to get a man to understand something, when his salary depends upon his not understanding it!" Biologist Richard Dawkins points out this behaviour isn’t even limited to humans: “The rabbit runs faster than the fox, because the rabbit is running for his life while the fox is only running for his dinner.”

Because incentives are such an incredibly-powerful motivator, great care must be taken in designing them. Second-order effects and unintended consequences must be contemplated. In our personal experience, one of the more striking examples of this came during a trip to Peru. In order to encourage home construction, the government doesn’t charge property tax until a building is deemed to be finished. That seems like a reasonable policy! Unfortunately, the unintended consequence of this incentive is illustrated in the following picture.


As the picture shows, some homeowners who live in otherwise-completed homes leave rebar sticking out of the roof for many years after construction has ended. This allows the owner to declare that the building is still unfinished, and avoid paying property taxes.

How it affects our portfolio

We pay close attention to management compensation structures whenever we analyze a company. While we are occasionally shocked by just how much money some CEOs are paid, we have a hard time criticizing generous payouts that reward an executive who does an exceptional job for shareholders.

But doing an exceptional job is not the same thing as growing a company at all costs. To us, it means generating an exceptional return on capital over the long term. Unfortunately, not all CEOs and boards of directors see it the same way: many are preoccupied with empire building and have incentives in place to encourage this detrimental behaviour. Despite the fact that a CEO is ultimately responsible to act in the best interests of shareholders, you can be sure that the CEO’s own interests will always come first in his or her mind. The goal of a well-designed incentive structure must be to align the CEO’s selfish interests with the interests of the shareholders providing the capital.