What are incentives in economics? Why people respond to them

Incentives are the rewards and penalties that make one choice more attractive than another: a higher price, a bonus, a fine, praise, shame or a clear conscience. In economics, they explain why behavior changes when circumstances do. Raise the price of something and people tend to use less; pay for something and people tend to supply more. The hard part is that people respond to the incentive as it is actually written, not as its designer meant it.

In short

• An incentive changes the costs or rewards of a choice, and people adjust what they do in response, often in ways nobody planned.
• Incentives can be financial, social or moral, and adding a money incentive can weaken the social or moral one already there.
• When a reward is tied to a measure, people improve the measure, which isn't always the same as improving what it measures.

What counts as an incentive?

Anything that changes what a choice costs or pays is an incentive. Economists usually start with money, because it is easy to see and count: wages, prices, taxes, subsidies, fines, interest. But money is only one kind. In Freakonomics (2005), the economist Steven Levitt and the journalist Stephen Dubner sorted incentives into three:

Kind	How it works	Example
Economic	Money gained or lost	A tax on cigarettes, a bonus for hitting a sales target
Social	How other people see you	Being seen to recycle, or seen to litter
Moral	How you see yourself	Not cheating because it would feel wrong

Most real choices carry all three at once. A driver who doesn't speed may be avoiding a fine, a neighbor's disapproval and a guilty conscience together. Incentives matter in economics because resources are scarce: every choice gives something up, so anything that changes what a choice costs changes which choices people make. That is why the overview of economics lists incentives among its few core ideas.

How do incentives change behavior?

Mostly in the direction you'd expect. When gasoline gets dearer, people drive a little less and, over years, buy more efficient cars. When a city pays a bounty on a pest, people catch more of it. When a job pays more, more people apply. Incentives work through millions of small decisions, which is why a tax or a subsidy can shift a whole market without anyone being told what to do.

Incentives also explain outcomes that look irrational from outside. A fishing boat that takes more than its share of a shared stock is responding to the fact that any fish it leaves will be caught by someone else, the logic of the tragedy of the commons. A factory that pollutes a river is responding to the fact that it doesn't pay for the damage, which economists call an externality. Two suspects who would both be better off silent each confess, because each does better by confessing whatever the other does, the trap in the prisoner's dilemma. In each case the people involved are acting sensibly given what they face; the problem is the incentive.

Why do some incentives backfire?

Because people respond to the rule as written. The best-documented case comes from Hanoi in 1902. The French colonial government, afraid of bubonic plague, paid a bounty of one cent per rat, and to claim it, a catcher had to hand in only the rat's tail. Officials soon noticed tailless rats in the city: catchers were cutting off tails and releasing the rats to breed, according to the historian Michael Vann's research, summarized in Wikipedia's account of the Great Hanoi Rat Massacre. The scheme was dropped. The rule paid for tails, not for fewer rats, and that is what it got.

A more famous story is weaker. In the "cobra effect," British officials in Delhi paid for dead cobras, people bred cobras for the reward, and when the bounty ended the breeders let them go, leaving more snakes than before. The German economist Horst Siebert gave the effect its name in 2001. But no contemporary records of cobra breeding in British India have been found: a 2025 investigation, summarized in the Wikipedia article on perverse incentives, traced the story to an 1873 newspaper article that said only "it was alleged." Vann has argued the Hanoi case should replace it, because it can be documented. The idea holds; the cobra story is best treated as a parable.

A modern case is documented in detail. In September 2016, the US Consumer Financial Protection Bureau fined Wells Fargo $100 million after employees, under sales goals and pay tied to them, opened roughly 1.5 million deposit accounts and applied for roughly 565,000 credit cards that customers may not have authorized, by the bank's own analysis. Pay tied to reported numbers has driven bigger failures too, among them Enron's collapse.

Can paying people make them do less?

Sometimes, when a money incentive pushes out a social or moral one. In 1998 the economists Uri Gneezy and Aldo Rustichini studied 10 private day-care centers in Haifa, Israel, where some parents collected their children late, keeping a teacher after closing time. After 4 weeks of counting late pickups, six centers, picked at random, began fining parents NIS 10, about $2.70 then, for arriving more than 10 minutes late. The other four carried on as before.

Late pickups at the fined centers rose, and within a few weeks settled at almost twice the starting level, the authors reported in a paper titled "A Fine Is a Price." When the fine was dropped in week 17, lateness didn't fall back. Their reading: before the fine, parents felt they were imposing on the teacher, a moral cost. The fine turned that into a small fee for extra childcare, which many were happy to pay. A penalty meant as a deterrent worked as a price. Economists call this crowding out: a financial incentive displacing a social or moral one.

The study doesn't show that fines never work. Larger fines, or fines framed differently, often do. It shows that an incentive lands among the ones already in place, and can change what a choice means to people.

What is Goodhart's law?

Goodhart's law is the measurement version of the same problem: once a number becomes a target, people improve the number by whatever route is easiest, and it stops telling you what it used to. It is named after the British economist Charles Goodhart, who wrote in a 1975 paper on monetary policy that "any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes." The shorter wording most people quote, "when a measure becomes a target, it ceases to be a good measure," was popularized by the anthropologist Marilyn Strathern in a 1997 paper, building on a 1996 chapter by the accounting scholar Keith Hoskin, as Wikipedia's article on the law records.

The pattern is easy to spot once named: rat tails counted instead of rats, accounts opened instead of customers served. The same risk sits in any waiting-time target that can be met by changing when the clock starts, or any school ranking that can be raised by teaching only the tested subjects. None of it requires everyone to be dishonest, only that the reward follow the number.

How do policy designers test an incentive?

Because incentives so often surprise their designers, careful ones are tested before they spread. The questions designers ask make a useful checklist for anyone setting a reward or a rule:

1. What exactly is rewarded? Name the action that earns the reward, and compare it with the outcome you actually want. Any gap between the two is where gaming happens.
2. Who could earn it without doing the thing? Picture the cheapest way to qualify, as the Hanoi rat catchers did.
3. What incentives are already there? A money reward can push out pride, duty or guilt, as in the Haifa day-care centers.
4. Can it be tried small, against a comparison? Gneezy and Rustichini's design, a change in some places and not others, chosen at random, is the standard way to see what an incentive does rather than what it was meant to do.
5. What happens when it ends? Behavior doesn't always return to where it was, as Haifa's parents showed.

Still argued

How far money incentives crowd out other motives is argued among economists and psychologists. A 1999 review of experiments by the psychologists Edward Deci, Richard Koestner and Richard Ryan concluded that tangible rewards tied to doing a task tend to reduce people's own interest in it. Other researchers, among them Judy Cameron and David Pierce, read the evidence as showing the effect is small and limited to particular kinds of reward, and many incentive schemes plainly work as intended. The Haifa result is widely cited and has been discussed for decades, but how well it generalizes beyond small fines in close-knit settings is not settled.

Check yourself

• Why did paying for rat tails fail to reduce the number of rats in Hanoi?
• Explain how a fine for late pickup could make lateness more common.
• What does Goodhart's law say, and what would it predict for a sales team paid by the number of new accounts?

Sources

• A Fine Is a Price — Uri Gneezy and Aldo Rustichini, Journal of Legal Studies, 2000
• CFPB fines Wells Fargo $100 million for widespread illegal practice of secretly opening unauthorized accounts — Consumer Financial Protection Bureau, 2016
• Great Hanoi Rat Massacre — Wikipedia
• Perverse incentive — Wikipedia
• Goodhart's law — Wikipedia