Broad bracketing
Pooling decisions and outcomes so trade-offs and risks are judged as a whole.
What it means
Broad bracketing is the practice of grouping multiple choices or outcomes together and evaluating them jointly rather than in isolation. Because aggregation lets gains and losses offset within a single bracket, broad framing tames loss aversion and yields choices closer to those a far-sighted, risk-neutral planner would make for repeated small stakes. Empirically, people who bracket broadly diversify better, take advantageous repeated risks, and show more consistent preferences across related decisions. The normative appeal is that, for many independent small gambles, the law of large numbers makes the aggregate nearly riskless, so narrow evaluation needlessly forgoes value. Encouraging broad bracketing — for instance by presenting returns over long horizons or bundling decisions — is a recognized debiasing strategy.
Why aggregation changes the choice
Under prospect theory the reference point resets with each decision a person treats as separate, so every possible loss is felt against a fresh zero and weighted more heavily than an equal gain. Narrow bracketing therefore lets loss aversion bite once per choice. Pooling the decisions changes the object being evaluated: the mind now faces a single combined distribution in which many small losses are diluted by offsetting gains and much of the probability mass sits in positive territory. The active ingredient is the aggregation itself, not merely knowing that a gamble repeats. Even when people are told a bet repeats, if outcomes are still presented one at a time the shift is muted; only laying out the combined distribution reliably moves the choice.
What the evidence shows
The core demonstrations are old and clean. Redelmeier and Tversky found that most people reject a single fifty-fifty bet to win $2,000 or lose $500, yet accept it once told they can play it five times, though nothing about any single play has changed. Gneezy and Potters showed the same lever moves real money: subjects who saw investment returns bundled over three rounds put far more into the risky asset than those who saw each round alone. Rabin and Weizsäcker sharpened the point by showing that narrow bracketers will knowingly pick combinations that are first-order stochastically dominated, strictly worse in every state. The effect is not bulletproof, though. Klos documents replication failures once the task nudges attention toward the final outcome, or once people simply misjudge how a short-run return compounds.
Where it shows up
The idea earns its keep wherever a stream of similar decisions gets chopped into isolated ones. Retirement saving is the marquee case: an investor who checks a volatile portfolio monthly sees losses often and flees to bonds, while the same portfolio judged over a decade looks almost sure to grow. Benartzi and Thaler tied this evaluation frequency to the equity premium puzzle, the stubborn gap between stock and bond returns. Insurance and extended warranties trade on narrow framing too, each policy sold against a single vivid loss. Choice architects exploit the reverse: showing long-horizon return distributions, bundling annual enrollment decisions, or defaulting savers into diversified funds all widen the bracket on the person's behalf. The design move is to present the pooled consequence, not the isolated one, before the choice is made.
Where narrow bracketing wins
Broad bracketing is a corrective, not a universal law, and Read, Loewenstein and Rabin were careful to say so. Its normative edge holds only when the pooled risks are roughly independent and small relative to wealth; combine gambles that move together and the diversification is illusory, and pool a stake large enough to ruin you and the law-of-large-numbers comfort evaporates. Narrow brackets can also do useful work. A dieter who refuses each cookie on its own terms, rather than budgeting cookies across a week, is using a narrow frame as a self-control device. And bracketing too broadly has its own failure mode: asked to pick a whole week of snacks at once, people over-diversify, choosing more variety than they actually want day to day and later regretting it. The question is which frame the decision calls for, not which is right in general.
Examples
Deciding once how to invest all monthly contributions for a year produces steadier, better-diversified choices than deciding paycheck by paycheck.
Offered a coin flip that wins $200 or loses $100, most people refuse it once but happily accept twenty of them — taken as a set, the run almost never ends down.
Extended warranties look sensible one gadget at a time; total every warranty you would buy in a decade against every gadget that actually breaks, and the bundle is plainly a bad deal.
A project manager who pads every task's deadline against its own worst case bloats the timeline; a single shared buffer at the end absorbs the tasks that overrun using those that finish early.
A firm that cancels each R&D project the moment it looks likely to fail backs only safe bets; as one portfolio, a spread of long shots wins, because rare successes cover the dead ends.
First described in Read, Loewenstein & Rabin (1999).
Key references
- Webb, E. C., & Shu, S. B. (2017). Is broad bracketing always better? How broad decision framing leads to more optimal preferences over repeated gambles. Judgment and Decision Making, 12(4), 382-395. doi.org/10.1017/S1930297500006252
- Klos, A. (2013). Myopic loss aversion: Potential causes of replication failures. Judgment and Decision Making, 8(5), 617-629. doi.org/10.1017/S1930297500003703
- Rabin, M., & Weizsäcker, G. (2009). Narrow bracketing and dominated choices. American Economic Review, 99(4), 1508-1543. doi.org/10.1257/aer.99.4.1508
- Read, D., Loewenstein, G., & Rabin, M. (1999). Choice bracketing. Journal of Risk and Uncertainty, 19(1-3), 171-197. doi.org/10.1023/A:1007879411489
- Gneezy, U., & Potters, J. (1997). An experiment on risk taking and evaluation periods. Quarterly Journal of Economics, 112(2), 631-645. doi.org/10.1162/003355397555217
- Redelmeier, D. A., & Tversky, A. (1992). On the framing of multiple prospects. Psychological Science, 3(3), 191-193. doi.org/10.1111/j.1467-9280.1992.tb00025.x