Max Hodak Writings

Perverse incentives are endemic

January 2023

There are many ways to run a company, but I often find it helpful to view a large part of my job as designing the incentive structure of the organization so that, to the extent possible, the right things happen automatically. While there is a lot to be said for directly managing things you have a clear vision for, you can’t always be in the room, and more to the point, you shouldn’t be; excellent coworkers are sources of new judgment and creativity, not just more hours in the day for you.

So how do you encourage the right things to happen at scale as a matter of course? You design incentives that reward it.

Equity in a company is the classic way of incentivizing employees to help build value generally but it provides limited practical help over short timescales, and worse, maxiziming short-term valuation growth is often a pretty bad number to focus on; it is an insufficiently good gradient for day-to-day things.

On the other hand, perverse incentives — things that actually incentivize worse performance — are strangely common. For example:

There is always some nice-sounding explanation for why things should be work out and with enough goodwill it’s tempting to believe it. But fundamentally there is a true fact — the presence of the incentive — which no amount of good faith can overcome. You may get lucky sometimes, but that is not a scalable strategy.

These are of course just three examples of many, but even these span a huge range of frequently encountered things. Basically the entire construction industry runs on cost-plus contracting; I believe this is one (of several) root causes of why construction in America is so overwhelmingly slow and expensive.1 An underrated, key part of the success stories for SpaceX and Anduril, aside from and enabled by the quality of their engineering, was the business model innovation of using their own money for R&D and then charging value-based pricing to customers. Arguably the whole story of why housing is expensive in the Bay Area is that homeowners have captured local government and are using their power to restrict further supply so as to maximize their own property values. Despite how different all of these things are, what ties them together is the power of incentives. This may sound obvious, but empirically there are huge amounts of value left on the table every day by misunderstanding these things.

Edsger Dijkstra popularized the idea of the “Buxton Index”, which he defined as “the length of the period, measured in years, over which the entity makes its plans”:

For the little grocery shop around the corner it is about 1/2, for the true Christian it is infinity, and for most other entities it is in between: about 4 for the average politician who aims at his re-election, slightly more for most industries, but much less for the managers who have to write quarterly reports. The Buxton Index is an important concept because close co-operation between entities with very different Buxton Indices invariably fails and leads to moral complaints about the partner. The party with the smaller Buxton Index is accused of being superficial and short-sighted, while the party with the larger Buxton Index is accused of neglect of duty, of backing out of its responsibility, of freewheeling, etc.. In addition, each party accuses the other one of being stupid. The great advantage of the Buxton Index is that, as a simple numerical notion, it is morally neutral and lifts the difference above the plane of moral concerns. The Buxton Index is important to bear in mind when considering academic/industrial co-operation.

I love this encapsulation of how the time horizon over which one has accountability leads to significant changes in how one is incentivized, and the ripple effects this can cause when you don’t anticipate a potential mismatch.

The fact that rational agents will tend to act in their own self-interest prompts the idea that effective leadership depends on finding ways to align your interests with those on whom you depend; and not in vague, long-term ways, but in practical, today ways. (Relatedly, it is important not to hold someone accountable for something not within their power to control; i.e., software engineers should almost never be given revenue targets to hit.) Being able to set the incentives of those around you is a superpower.

Hourly rates are ubiquitous because they are simple. In many cases, management is applied at timescales that are short relative to the total costs incurred, and so it works out fine. Trying to do this when the timescale of effective management (by the customer) is long relative to the rate of spending is how you get things like billion-dollar government websites.

There is no general algorithm for designing effective incentives2; it is a creative exercise which depends heavily on the specifics of the situation. But all else equal, the simpler the better and selected goals should to the extent possible be ends in themselves. Market-based mechanisms like competitive auctions can be powerful, but need to be used thoughtfully. At the end of the day, all of this is about the reality of human nature and solutions must be things people find motivating, not alienating.

Insisting on aligned incentives may sometimes limit your choice of vendors or partners, but I’ve learned to let that become an important signal of mutual cultural fit. Working with others who are willing to commit to well-aligned incentives and push for the rewards that makes possible is, I believe, one of the most underrated sources of alpha, and has the benefit of automatically lowering the risk you get caught up in market irrationality when such epidemics come along from time to time.

1. It is very common, for example, for a general contractor or an architect be paid a percentage of total construction costs as “standard.” I cannot sufficiently emphasize the degree to which this is insane. First of all, when dealing with large amounts of money and sophisticated parties, there simply is no such thing as a “standard term” in a contract. Certainly they may choose not to work with someone asking to do business in a way they to which they unaccustomed, but nevertheless, agreeing to a contract that contains materially perverse incentives is a recipe for failure, no matter the intentions.

2. Theoretically this is part of the field of mechanism design, and there are specialists whose business is coming up with practical compensation structures for complex business goals and avoiding unintended consequences, which is often harder than it sounds. But there’s nothing magic that prevents you from getting most of the way there by just spending the time thinking carefully. Full disclosure, my father is an executive compensation consultant with Farient Advisors, and the phrase “perverse incentives are endemic” was said a lot in our house while I was growing up.