My blog posts and reading material have both been on a decidedly economics-heavy slant recently. The topic today, incentives, squarely falls into the category of economics. However, when I say economics, I’m not talking about “analyzing supply and demand curves.” I’m talking about the true basis of economics: understanding how human beings make decisions in a world of scarcity. A fair definition of incentive is “a reward or punishment that motivates behavior to achieve a desired outcome.” When most people think about economic incentives, they’re thinking of money. If I offer my son $5 if he washes the dishes, I’m incentivizing certain behavior. We can’t guarantee that he’ll do what I want him to do, but we can agree that the incentive structure itself will guide and ultimately determine what outcome will occur. The great thing about monetary incentives is how easy they are to talk about and compare. “Would I rather make $5 washing the dishes or $10 cleaning the gutters?” But much of the world is incentivized in non-monetary ways too. For example, using the “punishment” half of the definition above, I might threaten my son with losing Nintendo Switch access if he doesn’t wash the dishes. No money is involved, but I’m still incentivizing behavior. And there are plenty of incentives beyond our direct control\! My son is *also* incentivized to not wash dishes because it’s boring, or because he has some friends over that he wants to hang out with, or dozens of other things. Ultimately, the conflicting array of different incentive structures placed on him will ultimately determine what actions he chooses to take. ## Why incentives matter A phrase I see often in discussions—whether they are political, parenting, economic, or business—is “if they could **just** do…” Each time I see that phrase, I cringe a bit internally. Usually, the underlying assumption of the statement is “if people would behave contrary to their incentivized behavior then things would be better.” For example: * If my kids would just go to bed when I tell them, they wouldn’t be so cranky in the morning. * If people would just use the recycling bin, we wouldn’t have such a landfill problem. * If people would just stop being lazy, our team would deliver our project on time. In all these cases, the speakers are seemingly flummoxed as to why the people in question don’t behave more rationally. The problem is: each group is behaving perfectly rationally. * The kids have a high time preference, and care more about the joy of staying up now than the crankiness in the morning. Plus, they don’t really suffer the consequences of morning crankiness, their parents do. * No individual suffers much from their individual contribution to a landfill. If they stopped growing the size of the landfill, it would make an insignificant difference versus the amount of effort they need to engage in to properly recycle. * If a team doesn’t properly account for the productivity of individuals on a project, each individual receives less harm from their own inaction. Sure, the project may be delayed, company revenue may be down, and they may even risk losing their job when the company goes out of business. But their laziness individually won’t determine the entirety of that outcome. By contrast, they greatly benefit from being lazy by getting to relax at work, go on social media, read a book, or do whatever else they do when they’re supposed to be working.  My point here is that, as long as you ignore the reality of how incentives drive human behavior, you’ll fail at getting the outcomes you want. If everything I wrote up until now made perfect sense, you understand the premise of this blog post. The rest of it will focus on a bunch of real-world examples to hammer home the point, and demonstrate how versatile this mental model is. ## Running a company Let’s say I run my own company, with myself as the only employee. My personal revenue will be 100% determined by my own actions. If I decide to take Tuesday afternoon off and go fishing, I’ve chosen to lose that afternoon’s revenue. Implicitly, I’ve decided that the enjoyment I get from an afternoon of fishing is greater than the potential revenue. You may think I’m being lazy, but it’s my decision to make. In this situation, the incentive–money–is perfectly aligned with my actions. Compare this to a typical company/employee relationship. I might have a bank of Paid Time Off (PTO) days, in which case once again my incentives are relatively aligned. I know that I can take off 15 days throughout the year, and I’ve chosen to use half a day for the fishing trip. All is still good. What about unlimited time off? Suddenly incentives are starting to misalign. I don’t directly pay a price for not showing up to work on Tuesday. Or Wednesday as well, for that matter. I might ultimately be fired for not doing my job, but that will take longer to work its way through the system than simply not making any money for the day taken off. Compensation overall falls into this misaligned incentive structure. Let’s forget about taking time off. Instead, I work full time on a software project I’m assigned. But instead of using the normal toolchain we’re all used to at work, I play around with a new programming language. I get the fun and joy of playing with new technology, and potentially get to pad my resume a bit when I’m ready to look for a new job. But my current company gets slower results, less productivity, and is forced to subsidize my extracurricular learning. When a CEO has a bonus structure based on profitability, he’ll do everything he can to make the company profitable. This might include things that actually benefit the company, like improving product quality, reducing internal red tape, or finding cheaper vendors. But it might also include destructive practices, like slashing the R\&D budget to show massive profits this year, in exchange for a catastrophe next year when the next version of the product fails to ship.  Or my favorite example. My parents owned a business when I was growing up. They had a back office where they ran operations like accounting. All of the furniture was old couches from our house. After all, any money they spent on furniture came right out of their paychecks\! But in a large corporate environment, each department is generally given a budget for office furniture, a budget which doesn’t roll over year-to-year. The result? Executives make sure to spend the entire budget each year, often buying furniture far more expensive than they would choose if it was their own money. There are plenty of details you can quibble with above. It’s in a company’s best interest to give people downtime so that they can come back recharged. Having good ergonomic furniture can in fact increase productivity in excess of the money spent on it. But overall, the picture is pretty clear: in large corporate structures, you’re guaranteed to have mismatches between the company’s goals and the incentive structure placed on individuals. Using our model from above, we can lament how lazy, greedy, and unethical the employees are for doing what they’re incentivized to do instead of what’s right. But that’s simply ignoring the reality of human nature. # Moral hazard Moral hazard is a situation where one party is incentivized to take on more risk because another party will bear the consequences. Suppose I tell my son when he turns 21 (or whatever legal gambling age is) that I’ll cover all his losses for a day at the casino, but he gets to keep all the winnings. What do you think he’s going to do? The most logical course of action is to place the largest possible bets for as long as possible, asking me to cover each time he loses, and taking money off the table and into his bank account each time he wins.  But let’s look at a slightly more nuanced example. I go to a bathroom in the mall. As I’m leaving, I wash my hands. It will take me an extra 1 second to turn off the water when I’m done washing. That’s a trivial price to pay. If I *don’t* turn off the water, the mall will have to pay for many liters of wasted water, benefiting no one. But I won’t suffer any consequences at all. This is also a moral hazard, but most people will still turn off the water. Why? Usually due to some combination of other reasons such as: 1. We’re so habituated to turning off the water that we don’t even consider *not* turning it off. Put differently, the mental effort needed to not turn off the water is more expensive than the 1 second of time to turn it off. 2. Many of us have been brought up with a deep guilt about wasting resources like water. We have an internal incentive structure that makes the 1 second to turn off the water much less costly than the mental anguish of the waste we created. 3. We’re afraid we’ll be caught by someone else and face some kind of social repercussions. (Or maybe more than social. Are you sure there isn’t a law against leaving the water tap on?) Even with all that in place, you may notice that many public bathrooms use automatic water dispensers. Sure, there’s a sanitation reason for that, but it’s also to avoid this moral hazard. A common denominator in both of these is that the person taking the action that causes the liability (either the gambling or leaving the water on) is not the person who bears the responsibility for that liability (the father or the mall owner). Generally speaking, the closer together the person making the decision and the person incurring the liability are, the smaller the moral hazard. It’s easy to demonstrate that by extending the casino example a bit. I said it was the father who was covering the losses of the gambler. Many children (though not all) would want to avoid totally bankrupting their parents, or at least financially hurting them. Instead, imagine that someone from the IRS shows up at your door, hands you a credit card, and tells you you can use it at a casino all day, taking home all the chips you want. The money is coming from the government. How many people would put any restriction on how much they spend? And since we’re talking about the government already… ## Government moral hazards As I was preparing to write this blog post, the California wildfires hit. The discussions around those wildfires gave a *huge* number of examples of moral hazards. I decided to cherry-pick a few for this post. The first and most obvious one: California is asking for disaster relief funds from the federal government. That sounds wonderful. These fires were a natural disaster, so why shouldn’t the federal government pitch in and help take care of people? The problem is, once again, a moral hazard. In the case of the wildfires, California and Los Angeles both had ample actions they could have taken to mitigate the destruction of this fire: better forest management, larger fire department, keeping the water reservoirs filled, and probably much more that hasn’t come to light yet. If the federal government bails out California, it will be a clear message for the future: your mistakes will be fixed by others. You know what kind of behavior that incentivizes? More risky behavior\! Why spend state funds on forest management and extra firefighters—activities that don’t win politicians a lot of votes in general—when you could instead spend it on a football stadium, higher unemployment payments, or anything else, and then let the feds cover the cost of screw-ups. You may notice that this is virtually identical to the 2008 “too big to fail” bail-outs. Wall Street took insanely risky behavior, reaped huge profits for years, and when they eventually got caught with their pants down, the rest of us bailed them out. “Privatizing profits, socializing losses.”  And here’s the absolute best part of this: I can’t even truly blame either California *or* Wall Street. (I mean, I *do* blame them, I think their behavior is reprehensible, but you’ll see what I mean.) In a world where the rules of the game implicitly include the bail-out mentality, you would be harming your citizens/shareholders/investors if you didn’t engage in that risky behavior. Since everyone is on the hook for those socialized losses, your best bet is to maximize those privatized profits. There’s a lot more to government and moral hazard, but I think these two cases demonstrate the crux pretty solidly. But let’s leave moral hazard behind for a bit and get to general incentivization discussions. # Non-monetary competition At least 50% of the economics knowledge I have comes from the very first econ course I took in college. That professor was amazing, and had some very colorful stories. I can’t vouch for the veracity of the two I’m about to share, but they definitely drive the point home. In the 1970s, the US had an oil shortage. To “fix” this problem, they instituted price caps on gasoline, which of course resulted in insufficient gasoline. To “fix” this problem, they instituted policies where, depending on your license plate number, you could only fill up gas on certain days of the week. (Irrelevant detail for our point here, but this just resulted in people filling up their tanks more often, no reduction in gas usage.) Anyway, my professor’s wife had a friend. My professor described in *great* detail how attractive this woman was. I’ll skip those details here since this is a PG-rated blog. In any event, she never had any trouble filling up her gas tank any day of the week. She would drive up, be told she couldn’t fill up gas today, bat her eyes at the attendant, explain how helpless she was, and was always allowed to fill up gas. This is a demonstration of *non-monetary compensation*. Most of the time in a free market, capitalist economy, people are compensated through money. When price caps come into play, there’s a limit to how much monetary compensation someone can receive. And in that case, people find other ways of competing. Like this woman’s case: through using flirtatious behavior to compensate the gas station workers to let her cheat the rules. The other example was much more insidious. Santa Monica had a problem: it was predominantly wealthy and white. They wanted to fix this problem, and decided to put in place rent controls. After some time, they discovered that Santa Monica had become *wealthier and whiter*, the exact opposite of their desired outcome. Why would that happen? Someone investigated, and ended up interviewing a landlady that demonstrated the reason. She was an older white woman, and admittedly racist. Prior to the rent controls, she would list her apartments in the newspaper, and would be legally obligated to rent to anyone who could afford it. Once rent controls were in place, she took a different tact. She knew that she would only get a certain amount for the apartment, and that the demand for apartments was higher than the supply. That meant she could be picky. She ended up finding tenants through friends-of-friends. Since it wasn’t an official advertisement, she wasn’t legally required to rent it out if someone could afford to pay. Instead, she got to interview people individually and then make them an offer. Normally, that would have resulted in receiving a lower rental price, but not under rent controls. So who did she choose? A young, unmarried, wealthy, white woman. It made perfect sense. Women were less intimidating and more likely to maintain the apartment better. Wealthy people, she determined, would be better tenants. (I have no idea if this is true in practice or not, I’m not a landlord myself.) Unmarried, because no kids running around meant less damage to the property. And, of course, white. Because she was racist, and her incentive structure made her prefer whites. You can deride her for being racist, I won’t disagree with you. But it’s simply the reality. Under the non-rent-control scenario, her profit motive for money outweighed her racism motive. But under rent control, the monetary competition was removed, and she was free to play into her racist tendencies without facing any negative consequences. ## Bureaucracy These were the two examples I remember for that course. But non-monetary compensation pops up in many more places. One highly pertinent example is bureaucracies. Imagine you have a government office, or a large corporation’s acquisition department, or the team that apportions grants at a university. In all these cases, you have a group of people making decisions about handing out money that has no monetary impact on them. If they give to the best qualified recipients, they receive no raises. If they spend the money recklessly on frivolous projects, they face no consequences. Under such an incentivization scheme, there’s little to encourage the bureaucrats to make intelligent funding decisions. Instead, they’ll be incentivized to spend the money where they recognize non-monetary benefits. This is why it’s so common to hear about expensive meals, gift bags at conferences, and even more inappropriate ways of trying to curry favor with those that hold the purse strings. Compare that ever so briefly with the purchases made by a small mom-and-pop store like my parents owned. Could my dad take a bribe to buy from a vendor who’s ripping him off? Absolutely he could\! But he’d lose more on the deal than he’d make on the bribe, since he’s directly incentivized by the deal itself. It would make much more sense for him to go with the better vendor, save $5,000 on the deal, and then treat himself to a lavish $400 meal to celebrate. # Government incentivized behavior This post is getting longer in the tooth than I’d intended, so I’ll finish off with this section and make it a bit briefer. Beyond all the methods mentioned above, government has another mechanism for modifying behavior: through directly changing incentives via legislation, regulation, and monetary policy. Let’s see some examples: * Artificial modification of interest rates encourages people to take on more debt than they would in a free capital market, leading to [malinvestment](https://en.wikipedia.org/wiki/Malinvestment) and a consumer debt crisis, and causing the boom-bust cycle we all painfully experience. * Going along with that, giving tax breaks on interest payments further artificially incentivizes people to take on debt that they wouldn’t otherwise. * During COVID-19, at some points unemployment benefits were greater than minimum wage, incentivizing people to rather stay home and not work than get a job, leading to reduced overall productivity in the economy and more printed dollars for benefits. In other words, it was a perfect recipe for inflation. * The tax code gives deductions to “help” people. That might be true, but the real impact is incentivizing people to make decisions they wouldn’t have otherwise. For example, giving out tax deductions on children encourages having more kids. Tax deductions on childcare and preschools incentivizes dual-income households. Whether or not you like the outcomes, it’s clear that it’s government that’s encouraging these outcomes to happen. * Tax incentives cause people to engage in behavior they wouldn’t otherwise (daycare+working mother, for example). * Inflation means that the value of your money goes down over time, which encourages people to spend more today, when their money has a larger impact. (Milton Friedman described this as [high living](https://www.youtube.com/watch?v=ZwNDd2_beTU).) # Conclusion The idea here is simple, and fully encapsulated in the title: incentives determine outcomes. If you want to know how to get a certain outcome from others, incentivize them to want that to happen. If you want to understand why people act in seemingly irrational ways, check their incentives. If you’re confused why leaders (and especially politicians) seem to engage in destructive behavior, check their incentives. We can bemoan these realities all we want, but they *are* realities. While there are some people who have a solid internal moral and ethical code, and that internal code incentivizes them to behave against their externally-incentivized interests, those people are rare. And frankly, those people are self-defeating. People *should* take advantage of the incentives around them. Because if they don’t, someone else will. (If you want a literary example of that last comment, see the horse in Animal Farm.) How do we improve the world under these conditions? Make sure the incentives align well with the overall goals of society. To me, it’s a simple formula: * Focus on free trade, value for value, as the basis of a society. In that system, people are always incentivized to provide value to other people. * Reduce the size of bureaucracies and large groups of all kinds. The larger an organization becomes, the farther the consequences of decisions are from those who make them. * And since the nature of human beings will be to try and create areas where they can control the incentive systems to their own benefits, make that as difficult as possible. That comes in the form of strict limits on government power, for example. And even if you don’t want to buy in to this conclusion, I hope the rest of the content was educational, and maybe a bit entertaining\!