Common difference effect
Adding the same delay to both options flips impatient choices into patient ones.
What it means
The common difference effect is the tendency for a preference between a smaller-sooner and a larger-later reward to reverse when the same delay is added to both: someone who takes the sooner reward when it is immediate will wait for the larger one when both are pushed into the future. This is the behavioral fingerprint of non-constant (non-stationary) discounting: any discount function that is steeper in the near term than the far term can produce it. The textbook reading attributes this to hyperbolic discounting, though subadditive discounting (Read, 2001) generates the same reversal without genuine declining impatience. Exponential discounting, with its constant per-period rate, keeps the ratio of present values fixed and so cannot produce the flip. In risky choice its formal parallel is the common-ratio effect, of which the certainty effect is a limiting case.
Why exponential discounting cannot produce it
A constant per-period discount factor makes preferences stationary: adding the same delay to both a smaller-sooner and a larger-later reward multiplies each present value by the same factor, so their ratio is unchanged and the choice cannot flip. An exponential discounter who prefers $100 now to $110 next week must also prefer $100 in a year to $110 in a year and a week. To reverse, impatience has to fall as rewards move into the future. A hyperbolic curve does exactly that: it discounts the first few weeks far more steeply than a constant rate would, then flattens. Pushing both rewards years out drops them onto that flat tail, where an extra week costs almost nothing, and the larger reward wins.
What the evidence shows
Early demonstrations were hypothetical (Thaler; Benzion, Rapoport, and Yagil), but the effect survives real stakes. Kirby and Herrnstein (1995), paying real money and goods worth $12 to $52, found that 34 of 36 subjects reversed from the larger-later to the smaller-sooner reward as the delays to both shrank toward the present. The reversal is among the most reliably reproduced anomalies of intertemporal choice and anchors Frederick, Loewenstein, and O'Donoghue's (2002) case against the discounted-utility model. Direction tracks the sign of discounting, though: Fan and colleagues (2025) show that for losses people want over with, where discounting runs negative, the same common delay flips choices the opposite way, a reverse common difference effect.
A rival to the hyperbolic story
The textbook reading treats the reversal as proof that impatience declines with how far off a reward sits. Read (2001) offers a different account: subadditive discounting. Total discounting over an interval is larger when the interval is chopped into pieces, so people look more impatient over any short stretch of time wherever it falls on the calendar. Across three experiments Read found strong subadditivity but no genuine declining impatience once interval length was held fixed. The distinction matters. It means the reversal may be a fact about how we perceive intervals, not about the present being uniquely tempting, and it cautions against reading every common-difference reversal as evidence of present bias.
Using it in practice
The design lesson follows from the asymmetry: the same person is patient about the distant future and impulsive about now, so lock in the good choice while it is still future-dated. Save-More-Tomorrow pension schemes, advance signup for a course or gym, pre-ordering the healthy lunch, and booking a screening weeks out all capture the patient preference before the sooner option becomes immediate. The reversal runs the other way for firms: auto-renewing subscriptions and hard-to-cancel defaults survive because cancellation is always a now-cost weighed against a later benefit. The practical tell is that a plan you set confidently for next month feels entirely different on the morning it actually comes due.
Examples
You take $100 today over $110 next week, but prefer $110 in a year and a week to $100 in a year — the same one-week gap, opposite choice.
Offered a small dessert tonight or a bigger one tomorrow, you take tonight's — yet push both a month out and you happily wait the extra day for the bigger one.
A team ships a rushed feature this Friday rather than a solid one the Friday after; planning next year's roadmap, the same team cheerfully budgets the extra week for quality.
A smoker vows on Sunday to quit next Monday and finds the plan painless; when Monday's craving finally arrives he lights up anyway. The resolve held only while the quit date sat a week away.
Employees who will not trim a cent from this month's pay agree to divert part of next year's raise into a pension. The Save-More-Tomorrow design works by keeping the sacrifice future-dated and therefore patient.
First described in Loewenstein & Prelec (1992); roots in Ainslie.
Key references
- Fan, Z., Zhang, X., Shen, Y., & Dai, J. (2025). Behavioral effects in time preference for losses depend on direction of delay discounting and level of data analysis. Judgment and Decision Making, 20, e28. doi.org/10.1017/jdm.2025.10004
- 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.org/10.1257/002205102320161311
- Read, D. (2001). Is time-discounting hyperbolic or subadditive? Journal of Risk and Uncertainty, 23(1), 5-32. doi.org/10.1023/A:1011198414683
- Kirby, K. N., & Herrnstein, R. J. (1995). Preference reversals due to myopic discounting of delayed reward. Psychological Science, 6(2), 83-89. doi.org/10.1111/j.1467-9280.1995.tb00311.x
- Loewenstein, G., & Prelec, D. (1992). Anomalies in intertemporal choice: Evidence and an interpretation. The Quarterly Journal of Economics, 107(2), 573-597. doi.org/10.2307/2118482