Incentives, Hidden Motives, and Why People Do What They Do

The Central Claim

The animating premise of Freakonomics is not particularly original within economics — it’s the disciplinary default. Assume that people respond to incentives. To understand behavior, find the relevant incentives. Where behavior is surprising, the incentives are probably different from the obvious ones.

What made the Levitt-Dubner presentation novel was the application: they applied the incentive lens to domains where economists hadn’t typically operated — crime, education, sumo wrestling, drug dealing — and found that stated explanations of behavior were often wrong because they assumed the stated incentives were the real ones. The book’s signature move is the reveal: the conventional story is X, but the real incentive structure is Y, and Y predicts the behavior much better.

The drug dealer who still lives with his mother. The real-estate agent who sells your home quickly at a lower price because her marginal incentive (a fraction of the higher price) is less valuable than freeing up her time. The sumo wrestlers who collude to preserve each other’s crucial match results because the marginal value of winning a make-or-break match differs from the marginal value of winning an inconsequential one. These examples work because the gap between the stated incentive (“maximize commissions,” “compete to win”) and the actual incentive structure (commission fractions, the value of each additional match win) produces measurably different behavior.

Robin Hanson’s Deeper Version

Robin Hanson and Kevin Simler’s The Elephant in the Brain (2018) extends this into a more philosophically radical claim: most of our professed motivations for social behavior are rationalizations of hidden evolutionary motives that we can’t easily observe because evolution has made us systematically blind to them.

The evolutionary premise: humans are social animals for whom status, mate acquisition, coalition formation, and reputation management have been crucial to survival and reproduction for millions of years. Natural selection would have favored individuals who pursue these goals effectively. But direct status pursuit and reputation management are socially costly if perceived — people who are obviously trying to appear impressive rather than genuinely helping are punished socially. Selection therefore favors individuals who pursue these goals in ways that look like something else, including to themselves.

The elephant in the brain is the self-interested motive that drives behavior but that we don’t acknowledge — often because we’ve genuinely convinced ourselves the prosocial story is the whole story. Hanson and Simler argue this explains puzzles in education (we claim to pursue knowledge but exhibit behaviors — grade-chasing, credential-signaling, high-status institution attendance — more consistent with status acquisition), medicine (we claim to pursue health but overconsume expensive treatments beyond what evidence supports, in ways more consistent with conspicuous care-seeking than with maximizing health outcomes), and politics (we claim to pursue good policy but exhibit behaviors more consistent with tribal loyalty and group identity than with maximally accurate policy beliefs).

Information Asymmetry and the Principal-Agent Problem

The formal economic version of the hidden motives problem is principal-agent theory. A principal (employer, client, voter) hires an agent (employee, professional, politician) to act on their behalf. The agent has private information — they know their own effort level, their own interests, and details about the task the principal can’t observe. The agent’s incentives are not perfectly aligned with the principal’s.

George Akerlof’s “The Market for Lemons” (1970, Nobel Prize 2001) is the foundation. In the used-car market, sellers know more about the car’s quality than buyers. Buyers, knowing they can’t distinguish good cars from bad ones, offer the average quality price. Sellers of above-average cars find the price unattractive and withdraw from the market. This lowers average quality, which lowers the average price, which drives more above-average sellers out — an adverse selection spiral that can collapse the market entirely. The information asymmetry destroys the gains from trade that would otherwise exist.

The lemons problem generalizes to any market with information asymmetry: insurance (people who know they’re high-risk want insurance more than those who know they’re low-risk; insurers can’t fully distinguish them), financial markets (issuers of securities know more about their quality than buyers; rating agencies were supposed to solve this and failed notoriously in 2008), professional services (clients can’t fully evaluate the quality of legal, medical, or financial advice they’re receiving).

Incentive Design and Its Failures

The practical application of incentive economics is mechanism design — designing institutional rules, contracts, and compensation structures to align agents’ incentives with principals’ goals. The difficulties are well-documented.

Teaching to the test is the canonical example in education: if teachers are evaluated by standardized test scores, they optimize for test scores. Test scores are proxies for learning; optimizing the proxy undermines what it’s a proxy for. Goodhart’s Law again — the measure becomes a target and ceases to be a good measure.

Financial sector compensation design produced the 2008 crisis in part through incentive misalignment. Investment bankers were paid bonuses based on deal volume and short-term performance, not long-term outcomes. Mortgage originators were paid per origination, not per loan quality — and loans were sold off to securitization vehicles rather than held, removing the skin-in-the-game that would have aligned incentives with loan quality. The entire chain from origination to securitization to rating to sale had incentives pointing toward quantity over quality, and the losses from poor quality were diffused to parties (pension funds, governments, ultimately taxpayers) who had no ability to monitor the origination quality.

When Incentives Backfire

Adding explicit incentives to activities that were previously governed by intrinsic motivation or social norms can reduce the target behavior — a phenomenon called motivation crowding-out.

The Israeli daycare study (Gneezy and Rustichini, 2000): a daycare introduced a fine for parents who picked up children late. Late pickups increased. The fine converted a social obligation (be punctual, don’t impose on the teachers) into a market transaction (pay for the additional time). The social norm was stronger than the financial penalty; replacing the norm with the fine removed the social cost while adding only a modest financial one.

Similar effects appear in blood donation (paying for blood donations reduces supply in some studies, because it converts a prosocial act into a commercial one and attracts the wrong kind of donors), volunteer work, and civic participation. The crowding-out effect suggests that incentive design needs to account for the social and psychological context in which the incentive operates — adding a financial incentive on top of social norms can undermine the norms without the financial incentive being strong enough to compensate.

The Common Thread

Across Freakonomics’s reveal-format examples, Hanson’s evolutionary hidden motives, and the formal information-asymmetry and principal-agent literature, the common thread is that stated motivations are unreliable guides to behavior. The relevant variables are:

What are the actual payoffs from different actions, measured in the currency the agent actually cares about (not necessarily money)? What information does each party have that others don’t? What behaviors can be observed and contracted on, and what can’t? What social norms and identity considerations shape the response to explicit incentives?

The gap between the stated story and the answer to these questions is where the interesting economics lives.