How Economists Think: Incentives, Margins, and Models

A French colonial government once tried to rid Hanoi of rats by paying a bounty for each rat killed. Within months, the city was full of tailless rats, and some entrepreneurs were quietly farming rodents in their backyards. The rat problem got worse.

That story isn’t a quirk of history. It’s a window into how economists actually think. Economics is less a list of facts than a way of thinking - a handful of mental moves you can point at almost any decision. Learn the moves, and you’ll start to see why your gym membership goes unused, why a “free” road still costs you something, and why good intentions so often produce the opposite result.

Why this matters

Most people make daily decisions on autopilot - finish the movie because you paid for it, buy in bulk because it “feels” cheaper, assume that if you win a deal someone else must lose.

Economic thinking is a quiet upgrade to that autopilot. It gives you a small set of questions that cut through fuzzy reasoning:

• Where will this incentive really push people, including the paths I didn’t plan?
• Is the next unit worth more than it costs, or am I clinging to money already spent?
• Am I treating this trade as a fight when it could make both sides better off?

You don’t need math for any of this. You need a few sharp habits. Here are the ones that do the most work.

People respond to incentives and not just the way you intended

An incentive is anything that rewards or penalizes a choice and so makes it more or less likely. A discount rewards buying. A fine punishes a behavior. A tax makes you do less of something.

The bedrock claim of economics is simple: when the reward or cost of an action changes, behavior changes too - usually toward the reward. Economist Steven Landsburg put it bluntly: “People respond to incentives; the rest is commentary.”

Here’s the part beginners miss. An incentive acts on every path to the reward, not just the path you had in mind. So it helps to split the response in two:

• The intended response - the behavior you wanted. Raise the cigarette tax, people smoke less.
• The side response - a path nobody planned, which can swamp the first one. Raise that tax steeply, and a black market in smuggled cigarettes appears.

Think of an incentive as water poured on a slope. You aim it at one plant, but water flows down every channel gravity offers. A careful designer studies the whole slope first. A careless one floods the basement.

When incentives backfire

History is stuffed with rules that delivered the exact opposite of their goal. The pattern is almost always the same: the rule rewards a measurable stand-in instead of the real goal, and people optimize the stand-in.

The Hanoi rats

Back to those rats. The bounty was paid on proof: hand in a rat tail. Catchers soon realized they could clip the tail, collect the money, and release the rat alive so it could breed and grow more tails. Others just farmed rats outright.

The true goal was fewer rats. The proxy was tails. People maximized tails. The rats won.

The cobra effect

A widely told story (less well documented, so treat it as a parable) describes British-run Delhi paying a bounty for dead cobras. People started breeding cobras to cash in. When officials scrapped the scheme, the now-worthless snakes were set loose and the cobra population climbed. Economist Horst Siebert named this the cobra effect: a fix that deepens the problem.

There’s a law underneath both stories. Goodhart’s Law: “When a measure becomes a target, it ceases to be a good measure.” The moment you reward the proxy, people game the proxy.

This is not ancient history. Engineers building AI hit the same wall and call it reward hacking: you train a system to maximize a score, and it finds a weird shortcut that boosts the score without doing the real job. That’s the Hanoi rat tail, reborn in code.

A subtler cousin: risk compensation

In 1975, economist Sam Peltzman studied 1960s car-safety rules. He argued road deaths fell less than hoped, because drivers who felt safer drove faster and took more risks - pushing some danger onto pedestrians and cyclists. The general idea is risk compensation: make something feel safer and people “spend” part of that safety on bolder behavior.

Don’t over-read this. Seatbelts clearly save lives. The honest, narrow lesson is that offsetting behavior can partly erode a safety gain - not that safety rules fail. People stretch this idea way too far to attack helmets, masks, and vaccines, and the true size of the effect is genuinely disputed.

Think at the margin

If you remember one habit from this whole article, make it this one.

To think at the margin is to focus on “one more, or one less” - the change from the next unit, not the total and not the average. Two terms do all the lifting:

• Marginal benefit - the extra satisfaction or money you get from one more unit.
• Marginal cost - the extra cost of one more unit.

The decision rule falls out automatically: do it if marginal benefit beats marginal cost; keep going until they’re equal; stop there.

You don’t ask “is pizza good?” You ask “is the fourth slice worth it to me right now?”

Should I do one more?         marginal benefit  vs  marginal cost
  -----------------------------------------------------------
  benefit > cost   ->  YES, do one more (you gain)
  benefit = cost   ->  STOP HERE (this is the sweet spot)
  benefit < cost   ->  NO, you've gone too far

Sunk costs don’t count

A sunk cost is money already spent that you can’t get back. At the margin, only forward-looking costs and benefits matter - so sunk costs are irrelevant.

The $50 movie ticket you already bought should not make you sit through a film you hate. The $50 is gone either way. The only live question is whether the next 90 minutes beat doing literally anything else.

This is the sunk cost fallacy: throwing good money after bad because you’ve “already invested so much.” Governments do it with failing megaprojects. People do it with dead-end jobs, bad relationships, and dying businesses. The past is not a reason. Only the future is.

The diamond-water paradox

Thinking at the margin quietly solves a 250-year-old puzzle. Adam Smith asked in 1776: water is essential to life yet nearly free, while diamonds are useless for survival yet cost a fortune. How can the useful thing be cheap and the useless thing dear?

The margin dissolves it. Price reflects the value of the next unit, not the value of all of it. Water’s total value is enormous, but because it’s abundant, your next glass is worth almost nothing. Diamonds give little total value, but because they’re scarce, the next one is precious.

This rests on the law of diminishing marginal utility: each extra unit gives less added satisfaction than the one before. So “how valuable is X?” is the wrong question. Ask “how valuable is one more X, given how much I already have?”

Trade isn’t a fight

Many people quietly assume trade is a battle: if I win, you lose. That’s zero-sum thinking, and for voluntary trade it’s simply wrong.

When two people freely swap, each gives up something they value less for something they value more. Both walk away richer in their own eyes or they wouldn’t have agreed. That makes voluntary trade positive-sum: it creates value rather than just moving it around.

The deepest version of this idea is comparative advantage, worked out by David Ricardo in 1817. Economist Paul Samuelson once called it the rare social-science idea that is both true and non-obvious.

The rule: specialize in whatever you give up the least to produce, then trade even if someone else is better than you at everything.

Picture a top surgeon who also happens to be the fastest typist in town. She should still hire a typist. Her hour in surgery is worth far more than her hour at the keyboard. Her absolute skill at typing is beside the point - what matters is what she gives up to do it.

One honest caveat: comparative advantage says the total pie grows, but it’s silent on who gets the slices. A country can gain overall while specific workers - say, a displaced factory town - genuinely lose. That’s why trade policy gets fought over so bitterly, from the 1800s to today’s tariff and “reshoring” debates. “Did total wealth rise?” and “Was it shared fairly?” are two separate questions. Don’t let one hide the other.

Models are wrong on purpose

Economists reason with models - deliberately simplified, partly unrealistic pictures of the world. This isn’t laziness. It’s the whole point.

A map that showed every pebble would be useless. A good map omits detail on purpose so you can actually see the route. Models do the same for ideas.

To isolate one cause at a time, economists lean on a Latin phrase, ceteris paribus - “all else held equal.” It’s a thinking device: change one factor, freeze everything else, and watch what happens. “Raise the price, ceteris paribus, and people buy less.” In the real world everything moves at once, so this is a mental lab control, not a literal claim about reality.

How do you judge a model you admit is unrealistic? Milton Friedman argued in 1953 that you judge a theory by its predictive power, not the realism of its assumptions. Physicists assume frictionless surfaces and still land probes on Mars. Statistician George Box summed it up: “All models are wrong, but some are useful.”

Be fair, though this view is contested. Samuelson called it “a monstrous perversion of science,” warning that wildly false assumptions can quietly mislead you. The debate is unsettled, and a good thinker holds both ideas in mind.

Prices are messages, not just numbers

Now combine incentives, margins, and trade into one of the most beautiful ideas in the field.

In 1945, F.A. Hayek asked a deceptively simple question: how does an economy coordinate millions of strangers when no single person knows enough to plan it?

The knowledge an economy needs is scattered - spread across millions of minds, much of it local and hard to write down. A shop owner knows her street’s demand. A miner knows his seam. No central planner could ever gather it all. So how does the system work at all?

Prices do the coordinating. A price is a compressed signal that bundles all that scattered knowledge into a single number, letting strangers cooperate without any of them seeing the whole picture.

Hayek’s example: suppose tin becomes scarcer - maybe a mine collapses, maybe a new use appears. The price of tin rises. Around the world, users economize on tin and hunt for substitutes without knowing why. They don’t need the story. The higher price is enough.

So a single price quietly does three jobs at once:

Role of a price	What it does
Signal	Carries scarcity information (“tin is scarce now”).
Incentive	Rewards acting on it (substitute, and you save money).
Rationing device	Steers the scarce good toward those who value it most.

This is Hayek’s famous case against full central planning - the “knowledge problem.” A planning office can’t replicate what the price system computes automatically, with no one in charge. Hayek won the Nobel Prize in 1974.

Common misconceptions

• “People only respond to money.” Incentives include status, time, convenience, guilt, and fear - anything that rewards or penalizes a choice. Money is just the loudest example.
• “If I won the trade, the other person lost.” In a voluntary swap, both sides expected to gain. Trade creates value; it doesn’t just shuffle it.
• “I should finish what I started.” Persistence is a virtue only when the road ahead is worth it. Honoring sunk costs is a bug, not grit.
• “Unrealistic models are useless.” A model’s job is to predict and clarify, not to mirror every detail. Wrong-but-useful beats right-but-unusable.
• “Economics assumes everyone is a cold, selfish robot.” That’s the starting model, not the conclusion. Real people satisfice, feel losses twice as hard as gains, and routinely pay a price for fairness.

How to use this

You can run any decision through five quick questions:

1. Trace the incentive’s whole slope. Before launching any reward or penalty, ask: “What’s the cheapest way for a clever, self-interested person to collect this?” If that path doesn’t serve your real goal, redesign before you launch.
2. Decide at the margin. Drop the all-or-nothing framing. Ask only whether the next unit - one more hour, one more slice, one more dollar - is worth more than it costs.
3. Ignore sunk costs out loud. When you catch yourself saying “but I’ve already put so much in,” stop. Ask instead: “Starting from right now, is continuing the best use of my time and money?”
4. Look for the positive-sum move. When a deal feels like a fight, ask whether each side could give up something they value less for something they value more. Specialize in what you sacrifice least to do, and trade for the rest.
5. Read prices as information. A rising price isn’t just bad news; it’s the world telling you something has become scarce. Treat it as a signal to economize or substitute, not just a cost to grumble about.

A final calibration. The classic model assumes a perfectly rational maximizer, but behavioral economics mapped the gap. Herbert Simon showed that people satisfice - they grab the first “good enough” option rather than optimize. Daniel Kahneman and Amos Tversky documented loss aversion, where losses hurt about twice as much as equal gains please. Richard Thaler showed we’re “predictably irrational,” and that gentle “nudges” can steer better choices. So use these tools and expect yourself and everyone else to apply them imperfectly.

Conclusion

If you take one thing from all of this, take the margin. Almost every confused decision - finishing the bad movie, buying the bulk pack you’ll never use, clinging to a sinking project - comes from staring at the total when you should be weighing the next unit. The single sharpest question in economics is not “is this good?” but “is one more worth what it costs, starting now?”

That habit also hands you a quiet warning: incentives are powerful but blind. They reward the behavior you measure, not the behavior you meant. Which raises a harder question for the next chapter - what happens when millions of these margin-by-margin choices collide in a single market? That’s where prices, supply, and demand start doing their strangest and most useful work.