Maybe It’s Not About Time

We are different people in the future. You lie down at night and silently promise to breakfast on yogurt and fruit. The next morning, fork in hand, future-you is eating eggs and bacon.

April 14, 2014

Perhaps you established a strict monthly budget to save for retirement. But wouldn’t current-you rather spring for a last-minute vacation with friends?

And yet experimental evidence that explicitly supports hyperbolic discounting is both spare and inconclusive. Many tests confound the effects of time on decision-making with other potential effects—the influence of proximity, or uncertainty, or simple practicality. Take the yogurt versus eggs and bacon example: this lapse of willpower may surface because we favor the present, and long-term health takes a backseat to gustatory pleasure. Alternatively, it may be because the smell of bacon affects our decisions. The former would support claims of hyperbolic discounting; the latter would support, one might say, the Bacon-Smell Effect.

Unconvinced of hyperbolic discounting’s “hard-wiring,” Shane Frederick from the Yale School of Management and YCCI designed an experiment that carefully isolated the effect of time on individual decision-making. To the surprise of Frederick and his coauthors—Daniel Read of Warwick Business School and Mara Airoldi of the London School of Economics and Political Science—the evidence failed to support the theory. (Their results were published in Acta Psychologica.)

The test was a straightforward series of surveys in the U.K. over four consecutive Tuesdays. Respondents made a total of 10 choices between “smaller-sooner” and “larger-later” rewards. On the Tuesday of week 1, they made all four choices (choosing between £20 “within one day” or £21 in one week, between £21 in one week or £22 in two weeks, and so on). On each successive Tuesday, they chose again between remaining options with the delay correspondingly reduced. For example, on week 2, one of their three choices was between £21 within a day or £22 in one week. The weekly follow-ups continued until the questions had been exhausted.1

The theory of hyperbolic discounting predicts that participants will typically shift their choices in favor of present rewards. If they chose £24 in four weeks, when the third week rolls around they ought to revise their choice to take the £23 that same day.

But the “data surprised us,” write the authors. Not only were most people consistent (that wasn’t especially surprising), but when they did revise their choices, they were just as likely to shift from short-term to long-term rewards. In other words, a shift toward patience was as common as a shift toward impatience.

The paper also notes that much of the anecdotal data cited on behalf of hyperbolic discounting is consistent with alternate theories—George Loewenstein’s “visceral factors” theory (1) or Yaacov Trope’s “construal theory” (2), for instance.

  1. Specific, visceral stimulation from rewards may trigger specific behavior. Again, take bacon: a subject sitting in an odorless room offered an apple now or bacon later may choose the apple. If the room smells like frying bacon, he’d likely choose the bacon. In the same way, by offering a monetary reward, researchers may trigger behavior (namely, taking that reward) that is related to something other than temporal proximity—a desire to buy something, for instance.
  2. Options that meet concrete goals tend to be favored in the short term, whereas options that fulfill more abstract goals are attractive from a distant perspective. Because it’s applicable to immediate and concrete needs—grocery shopping, bills, minigolf—the offer of a monetary reward may be immediately favorable.

Though the theory of hyperbolic discounting nicely matches real-world behavior, anecdote is not evidence. The authors note that if hyperbolic discounting truly ruled our decision-making process—by day, by week, by year—would we not perpetually impoverish our futures? Most existence is not so bleak. “Indeed,” conclude Frederick and his coauthors, rather philosophical, “at least among those still alive to participate in surveys, the dominant regret is not excessive indulgence, but failures to indulge.”

1In a second experiment, the “immediate” payment was accelerated from one day to one hour. This study also asked participants to explain their decisions.