Skip to content
Temporal ChoiceHigh Impact

Present Bias

Edited by Paweł Raja, PhD · Published

Definition

Present Bias is the tendency to give immediate rewards and costs extra weight compared with later ones, so that smaller, sooner gratification wins over larger, later benefits — the pattern economists model as hyperbolic (or quasi-hyperbolic) discounting.

Also known as:
Hyperbolic discounting, Quasi-hyperbolic discounting, Present-biased preferences, Immediacy effect
Category:
Temporal Choice
Impact:
High
Landmark study:
Laibson (1997)
Evidence here:
1 field finding, 14 cited studies

Why it happens: the psychology of present bias

Standard economics assumes people discount the future at a constant rate, so a choice between two dated rewards should not flip as time passes. Real choices do flip. Many people prefer $100 today to $110 tomorrow, yet prefer $110 in 31 days to $100 in 30 days. The extra day of waiting only feels costly when it starts now.

Laibson (1997) formalized this as quasi-hyperbolic or 'beta-delta' discounting: every future period is discounted by an extra factor β < 1 on top of the usual δ. The result is time inconsistency. Today's self makes plans — save, exercise, study — that tomorrow's self then abandons.

O'Donoghue and Rabin (1999) distinguished naive people, who do not realize their future self will also be present-biased, from sophisticated people, who do. Naive people procrastinate repeatedly and overpay for flexibility; sophisticated people seek out commitment devices.

Present bias is stronger for effort and consumption than for money. Augenblick, Niederle and Sprenger (2015) found clear present bias in how much unpleasant work people chose to do, but little when the same people allocated money over time.

Present Bias examples

Office workers choosing a snack to eat a week later mostly picked fruit, but when choosing a snack to eat right away most picked an unhealthy one (Read & van Leeuwen, 1998) — the same trade-off judged differently once it is immediate.

  • Health: signing up for a gym in January but skipping workouts all year — the distant benefit loses to today's comfort.
  • Money: 'buy now, pay later' feels painless because the cost is pushed onto a future self that seems like a different person.
  • Money: postponing retirement saving 'until next year', every year, because this month's spending always feels more pressing.
  • Work: leaving a report until the night before the deadline, even after promising yourself to start early this time.

Hyperbolic discounting vs. quasi-hyperbolic (beta-delta) discounting

The textbook benchmark is exponential discounting: each extra period of delay shrinks a reward's value by the same proportion (a factor δ per period). Under that rule a person's ranking of two dated rewards never changes as time passes, so plans made today are still optimal tomorrow.

Hyperbolic discounting replaces the constant rate with one that falls as the delay grows: waiting from today to tomorrow feels costly, while waiting from day 30 to day 31 hardly registers. A common one-parameter form values a reward delayed by t as 1/(1 + kt) of its immediate value. George Ainslie (1975) drew on delay-of-gratification experiments with animals and people to argue that discounting takes this shape, and that it explains why a larger, later reward can be preferred from a distance but abandoned as the smaller, sooner one draws near. Thaler (1981) found that the discount rates implied by people's answers fell steeply as the delay lengthened, and Loewenstein and Prelec (1992) proposed a generalized hyperbolic function to fit these and related anomalies.

Quasi-hyperbolic, or beta-delta (β-δ), discounting is the simpler two-parameter version introduced by Laibson (1997). Anything received now gets full weight; anything in the future is multiplied by an extra factor β < 1 and then discounted exponentially by δ. The single drop between 'now' and 'later' captures the key feature of hyperbolic curves — impatience concentrated in the present — while staying easy to work with in economic models. Laibson used it to show why people value illiquid 'golden eggs' such as pensions and housing that protect savings from their future selves.

An illustrative calculation (not data): with β = 0.7 and δ close to 1, $100 today beats $110 next week, because 0.7 × $110 = $77. But $110 in 53 weeks beats $100 in 52 weeks, because both amounts are scaled by the same 0.7. When week 52 arrives, the earlier reward becomes 'now' and the preference flips — the time inconsistency that O'Donoghue and Rabin (1999) analyzed in naive and sophisticated decision-makers.

In practice 'present bias', 'hyperbolic discounting' and 'quasi-hyperbolic discounting' are often used interchangeably. Strictly, present bias is the behavior and the other two are models of it. How large the bias is depends on what is being traded off: experiments that carefully separate money from consumption find modest present bias over money (Andreoni & Sprenger, 2012) but clear present bias over effort (Augenblick, Niederle & Sprenger, 2015).

How to avoid present bias

  1. Decide in advance and lock it in. Commitment devices — automatic savings transfers, pre-paid classes, deadlines with penalties — let your patient self bind your impatient one.
  2. Use automatic escalation. Committing now to save more out of future pay rises (the Save More Tomorrow design) avoids a painful cut in today's spending.
  3. Bundle temptation with the chore. Allowing a treat only while doing the dreaded task makes the immediate payoff work for you.
  4. Assume your future self will be just as tempted. Planning as if you will procrastinate (the 'sophisticated' stance) beats trusting tomorrow's willpower.

How to design for it (nudge strategy)

Implement 'Save More Tomorrow' mechanisms where commitments are made for the future, or provide small, immediate rewards for actions with long-term benefits.

Ethical use: design for choices people would endorse on reflection — a nudge, not sludge. Be transparent and keep opting out easy.

The evidence (1)

Key studies

  1. Introduced the quasi-hyperbolic (β-δ) model of present-biased, time-inconsistent preferences.

    Laibson, D. (1997). Golden eggs and hyperbolic discounting. Quarterly Journal of Economics, 112(2), 443–477. DOI →

  2. Drew on animal and human delay-of-gratification research to argue that rewards are discounted hyperbolically, which explains preference reversals and the appeal of commitment strategies.

    Ainslie, G. (1975). Specious reward: A behavioral theory of impulsiveness and impulse control. Psychological Bulletin, 82(4), 463–496. DOI →

  3. The discount rates implied by people's answers about waiting for money fell sharply as the length of the delay increased.

    Thaler, R. (1981). Some empirical evidence on dynamic inconsistency. Economics Letters, 8(3), 201–207. DOI →

  4. Catalogued anomalies in how people discount the future and proposed a generalized hyperbolic discounting function to explain them.

    Loewenstein, G., & Prelec, D. (1992). Anomalies in intertemporal choice: Evidence and an interpretation. Quarterly Journal of Economics, 107(2), 573–597. DOI →

  5. Reviewed the evidence against constant (exponential) discounting, including discount rates that decline with delay, and the models proposed to replace it.

    Frederick, S., Loewenstein, G., & O'Donoghue, T. (2002). Time discounting and time preference: A critical review. Journal of Economic Literature, 40(2), 351–401. DOI →

  6. Showed how naive versus sophisticated present-biased agents differ in procrastination and in demand for commitment.

    O'Donoghue, T., & Rabin, M. (1999). Doing it now or later. American Economic Review, 89(1), 103–124. DOI →

  7. Office workers choosing a snack for next week mostly picked fruit; choosing for right now, most picked an unhealthy snack.

    Read, D., & van Leeuwen, B. (1998). Predicting hunger: The effects of appetite and delay on choice. Organizational Behavior and Human Decision Processes, 76(2), 189–205. DOI →

  8. Members on flat monthly contracts attended so rarely that they paid more per visit than a pay-per-visit pass would have cost — consistent with overconfidence about future self-control.

    DellaVigna, S., & Malmendier, U. (2006). Paying not to go to the gym. American Economic Review, 96(3), 694–719. DOI →

  9. Many bank clients voluntarily chose a savings account that locked their money away, and those offered it saved more — evidence of demand for commitment.

    Ashraf, N., Karlan, D., & Yin, W. (2006). Tying Odysseus to the mast: Evidence from a commitment savings product in the Philippines. Quarterly Journal of Economics, 121(2), 635–672. DOI →

  10. Found substantial present bias over effort but limited present bias over money.

    Augenblick, N., Niederle, M., & Sprenger, C. (2015). Working over time: Dynamic inconsistency in real effort tasks. Quarterly Journal of Economics, 130(3), 1067–1115. DOI →

Key research

Related biases

Present Bias: frequently asked questions

What is present bias in simple terms?

Present bias is the tendency to give immediate rewards and costs extra weight compared with later ones. It is why people plan to save, diet or study 'starting tomorrow' and then put it off again when tomorrow arrives.

What is the difference between present bias and hyperbolic discounting?

Hyperbolic discounting is a model in which the discount rate falls with delay, so waiting feels most costly in the near term. Present bias is the behavior that model produces; the quasi-hyperbolic (β-δ) model is the most common way economists capture it.

What is hyperbolic discounting?

Hyperbolic discounting is the tendency for the value of a reward to drop steeply over short delays and more slowly over long ones. Because near-term waiting feels so costly, people can prefer a larger, later reward when both options are far away, then switch to the smaller, sooner one as it becomes immediate.

What is quasi-hyperbolic (beta-delta) discounting?

It is David Laibson's (1997) two-parameter approximation of hyperbolic discounting. Rewards received now get full weight; all future rewards are multiplied by an extra factor β below 1 and then discounted at the usual rate δ. A lower β means stronger present bias.

What is an example of present bias?

Paying for an annual gym membership and rarely going, leaving retirement saving until later, or choosing chocolate for right now while choosing fruit for next week.

How do you overcome present bias?

Use commitment: automate savings, set binding deadlines, pre-commit to future increases, and remove the need for willpower at the moment of temptation.

Cite this page

Raja, P. (Ed.). (2026, 8 Oct 2026). Present Bias. Behavioral Economics Lab. https://www.behavioraleconomicslab.com/biases/present-bias