Read an Excerpt

Chapter One


Thaler's Apples

The behavioral revolution in economics began in 1981 when Richard Thaler published a seven-page letter in a somewhat obscure economics journal. Richard is now a stocky sixty-three-year-old with unruly gray hair who looks more like a bartender than one of the world's leading economists. But back then, a thirty-five-year-old Richard posed a pretty simple choice about apples.

Which would you prefer:

(A) One apple in one year or

(B) Two apples in one year plus one day?

This is a strange hypothetical-why would you have to wait a year to receive an apple? But choosing is not very difficult; most people would choose to wait an extra day to double the size of their gift.

Thaler went on, however, to pose a second apple choice.

Which would you prefer:

(C) One apple today or

(D) Two apples tomorrow?

What's interesting is that many people give a different, seemingly inconsistent answer to this second question. Many of the same people who are patient when asked to consider this choice a year in advance turn around and become impatient when the choice has immediate consequences-they prefer C over D. When it comes to apples, Adam and Eve aren't alone in being impatient when presented with an immediate temptation.

The inconsistency in these answers puzzled Thaler. Richard has an incredible eye for anomalies. While many economists ignore or paper over deviations from rationality, Thaler is drawn to them. He's made a career of trying to understand them, and he even writes down algebraic formulas to capture their essence. Why would he mow his own lawn to save $15 but wouldn't be willing to cut his neighbor's lawn even for $25? Why would his decision about whether to drive through a snowstorm to see a basketball game turn on whether he paid for or was given the ticket? Thaler's obsession with the failure of traditional economics may end up earning him the Nobel Prize. By integrating psychology into economic theory, Thaler and a cadre of other behavioral economists have remade the landscape of economic thinking.

Today the defenders of the faith in economic rationality are still dominant in economic departments across the country, but they are increasingly acknowledging the power of behavioral predictions. While economists have traditionally focused on information and incentives as the core levers of human behavior, Thaler often thinks about what he calls "choice architecture." Simply and slightly reframing the context in which decisions are made can have big effects on people's behavior and happiness. Thaler is even willing to reengineer the rules of golf.

Cade Massey, a Yale business professor, played golf with Thaler soon after Thaler had taken up the game. "Dick suggested that we play with a different scoring rule," he recalls. "You'd get one point for every hole that you parred and nothing for every other hole, regardless of how badly you did. The winner was the person with the most points. This simple rule change really improved my enjoyment. The problem with traditional scoring for a golfer like me is that I could screw up a decent round with a bad score on a single hole." Richard's method allowed Cade to focus more on his successes than his failures. The story also shows how Thaler is constantly trying to use what he knows about how our minds work to increase people's happiness.

The golf purists out there will resist the idea that a scoring change could make golf more enjoyable. Or they might insist that it wouldn't be golf anymore. But Thaler has gone after much larger fish. His impulse to tinker with the rules of the game led him to champion the Save More Tomorrow program. Thaler saw that lots of people were having trouble taking advantage of their companies' 401(k) plans-even when their employers were willing to match their contributions dollar for dollar. It just hurts too much to see your current paycheck go down. But Save More Tomorrow lets people sign up for participation without their current or future paychecks ever declining. Savings contributions are made only out of a portion of future pay increases. These contributors usually reach full participation within four or five years. Over the course of a working life, this is a fairly small delay. But Richard's small change has shown big results. Employees who would never have otherwise invested in a 401(k) plan have been saving for retirement at nearly the same rate as those employees who chose to invest in their 401(k) plans from the start.

The seeds of this idea were planted back with those simple questions about whether you'd be willing to wait an extra day for an extra apple. Thaler's curiosity about the inconsistent answers to the two questions helped spark a revolution in economic thinking. Why were people more impatient when the proffered gifts were closer at hand? Both hypotheticals ask us if we're willing to wait an extra day to double the size of our gift. The only difference is the time perspective.

Now, one way out of this conundrum is to focus on trust. We're taught the old maxim that "a bird in the hand is worth two in the bush," because promised future paydays don't always materialize. The people who prefer an apple today may not trust that they'll actually be given two apples if they wait till tomorrow (whereas they don't think there's much of a difference in the likelihood of being given one or two apples if they have to wait a year).

But Thaler thought there was something else going on.

What was revolutionary about his apple example is that it illustrated the plausibility of what behavioral economists call "time- inconsistent" preferences. Richard was centrally interested in the people who chose both B and C. These people, who preferred two apples in the future but one apple today, flipped their preferences as the delivery date got closer.

Indeed, we can imagine anchoring the dates somewhere in the distant future (Would you prefer an apple on April 4 or two apples on April 5?) and then revisiting the question on a daily basis. Initially, when April is far away, people would choose to double their gift by taking the April 5 option. But because of a time-inconsistent preference, there would come a day where suddenly people would shift to the option of one apple on April 4. This flip in preference is the key to the behavioral revolution in economics.

If we change the hypothetical from apples to crack, it's easy to understand this kind of time inconsistency. A drug-deprived addict may prefer an immediate fix to a double dose tomorrow. When we hear about someone like Alex choosing the immediate small pleasure of induced sneezing to the larger future pleasure of clear lungs, we might be quick to think of him as having an addictive personality or some kind of OCD (obsessive-compulsive disorder). But Thaler's apple example shows that you don't need psychopharmacological dependence to exhibit irrational impatience. Even with regard to something as prosaic and healthful as an apple, many people in the world become impatient as their time horizon shrinks.

In at least some aspects of their lives, many people are addicted to now. We are seduced by the immediacy of gratification. The apple choice is replayed repeatedly throughout our lives. My kids put off cleaning their rooms and then waste huge amounts of time looking for a misplaced sweater or birthday invitation. A friend of our family again and again put off buying life insurance and then tragically left his family unprotected. The marginal convenience of delaying a mammogram or other mildly unpleasant task blinds us to the exponential pain just around the bend. I turned fifty almost a year ago, and I still haven't had a colonoscopy.

If asked in late January, most of us would say we would rather spend eight hours working on our taxes on April 1 than twelve hours on the day of the filing deadline, April 15. But as April Fool's day approaches, many of us (foolishly) will reverse course and choose more pain as long as it is even slightly postponed. This is no hypothetical for me. I postpone my visit with H&R Block year after year, even when I expect a fat refund.

Behavioral economists explain these reversals with the esoteric term "hyperbolic discounting." Rational choice theory predicts that the value today of some future gift should lose a fixed proportion of its value for every unit of delay. Rational actors, for example, might think that the present value of an apple decays 1 percent every extra day you have to wait for it. Rational actors would never be impatient if asked either of Thaler's questions because two apples, even with the extra one-day 1 percent discount, would always be better than one apple without this discount. But for decades behavioralists have noticed that humans tend to discount delays hyperbolically-there are big reductions in value for the first small bits of delay and relatively small reductions in value for subsequent increases in delay.

Hyperbolic discounting can explain why people are more likely to switch to the smaller but sooner reward when the reward is close at hand. In the apple example, a hyperbolic discounter might feel that delaying the immediate receipt of an apple by just a day would reduce its value by 70 percent, but delaying the receipt from a year to a year and a day would reduce the apple's present value only from an 80 percent discount to an 81% discount. The hyperbolic discounter will prefer a single apple now to two apples (discounted at 70 percent) a day from now; whereas when confronted with the analogous choice to receive two apples a year in the future, she will hold out for the two apples in part because she so strongly discounts any reward that far in the future. Hyperbolic discounters put extraordinary value in receiving rewards immediately (and in pushing off immediate burdens for even short periods), but then become relatively indifferent about when the reward (or burden) arrives in the future. For them, a reward two years from now is not that much better than a reward five or ten years from now. This is a bit of an overstatement, but a hyperbolic discounter tends to think it's "now or whatever."

Hyperbolic discounters are classic "conflict avoiders." They postpone disciplining the wayward child or addressing marital strife because when confronted with a "pay me now or pay me later" choice, they strongly prefer bearing the later but larger cost. These inconsistent preferences are more than a theoretical possibility. I want to ultimately convince you that many people manifest increasing impatience as the time before the moment of the reward shrinks.

Bird Brains

But let me start by telling you about pigeons-six male White Carneau pigeons, to be exact. Back in 1981, the same year that Thaler published his famous apple study, two Harvard researchers published a laboratory experiment testing the same idea on a very different kind of subject. Over an eleven-month period, George Ainslie and Richard Herrnstein presented the six pigeons with more than 75,000 food choices. Once

a day, each pigeon was placed in a special feeding apparatus that had a green and a white typewriter key mounted about five inches apart on a wall. For each feeding test, the two keys would light up and the pigeon had 10 seconds to peck one of the two keys. If the pigeon pecked the green (left) key, a food hopper-after a delay of "D" seconds-would open to let the pigeon eat for 2 seconds. However, if the pigeon pecked the white (right) key, the food hopper-after a delay of "D plus 4" seconds-would open for 4 seconds. For example, when D was equal to 1, the pigeon could choose to have 2 seconds' worth of food if it was willing to wait 1 second, or 4 seconds of access to food if it was willing to wait 5 seconds. Sixty seconds later, both keys would light up again and the pigeon would be confronted with the same choice. This process would repeat for about forty-five tests each day for each pigeon for eleven months (thank God for laboratory assistants).

Notice how closely this experiment parallels Thaler's apple hypothetical. In all 75,000 tests, the pigeons were given the opportunity to eat twice as much (4 seconds of access to food versus 2 seconds) if they were willing to wait a bit longer. The only difference-as with Thaler-was in the base delay, D. For example, in some sessions, the pigeons were asked to choose between waiting 12 seconds and waiting 16 seconds (i.e., D was equal to 12).

The researchers obviously couldn't explain the rules of the game to the pigeons. But by trial and error, the pigeons would learn about the nature of their environment and the relative costs and benefits of the two keys. Usually within just a few sessions, the pigeons would develop fairly stable preferences for one key or the other. The researchers maintained a constant D environment until all the pigeons' responses became stable. To keep the pigeons motivated (i.e., hungry), the researchers maintained the pigeons at 80 percent of their free- feeding body weight.

This bizarrely simple experiment provided strong evidence that you don't need to be a rocket scientist to follow the dictates of hyperbolic discounting. You don't even need to be human.

The graph in figure 2 shows the proportion of feedings across the six avian subjects where pigeons chose patience-that is, where the pigeons, by pecking the white key, were willing to wait 4 seconds to double their opportunity to eat. First, for those who think that literal birdbrains are just too limited to do more than randomly peck in search for food, notice how nonrandom the pecking patterns are. The proportion of birds choosing patience when the delay was short (say, 2 seconds versus 6 seconds) was dramatically lower than the proportion when the delay was longer (say, 8 seconds versus 12 seconds). These pecking patterns do not bounce around 50 percent, but show stark variations contingent on delay.

The experiment provides vivid support for the idea that something like hyperbolic discounting can lead to preference reversals. When presented with the chance of almost immediate gratification of 1 or 2 seconds, the pigeons are almost never able to bring themselves to wait a bit longer to double their food. The proportion choosing patience for these conditions is less than 5 percent. This result is especially striking as the pigeons learn that they will have subsequent feeding opportunities every 60 seconds. Looking at just the left-hand side of the graph, it would be easy to conclude that these silly animals are just hardwired for extreme impatience.

But when the base level of delay is increased just a bit, the pigeons behave very differently. When choosing between 8 and 12 seconds, it becomes a no-brainer to wait the extra 4 seconds to double your access to food. Just as predicted by hyperbolic discounting, as the time horizon decreases, pigeons become impatient.