Skip to content
Menu
Finabase Learn

Investor psychology Lesson 12 of 15

Articles · Behavioral finance

Regret aversion: how the fear of being wrong shapes investment choices

What research says about regret from acting and from not acting, how it leads to waiting in cash and to half measures, and what 393 one-year periods of SPY show.

On this page
  1. The research
  2. What it looks like in an account
  3. What the record shows: regret by choice
  4. The price of the half measure
  5. What the evidence suggests helps
  6. How Finabase helps
  7. Frequently asked questions
  8. Summary
  9. Check your understanding
  10. Exercise

Regret aversion is choosing in order to avoid feeling, later, that you chose wrongly. Regret differs from disappointment. Disappointment is a bad result. Regret is a bad result that you can trace to your own decision, with a clear picture of the alternative you passed up.

Investing supplies that picture every day. Every stock you sold keeps a price, as does every stock you considered and did not buy. Markets publish the outcome of every alternative, which makes them unusually good at producing regret, and makes investors unusually likely to decide with one eye on how they will feel.

The research

Regret as part of the decision

In 1982 two papers, one by Graham Loomes and Robert Sugden and one by David Bell, proposed the same idea independently. Standard theory says people judge each option by its own possible results. Regret theory says they also compare each result with what the other option would have given, and weigh that comparison in advance. A person might turn down a good bet because losing it would hurt more than a plain loss, since they could have had the sure thing.

Acting and not acting

The same year, Daniel Kahneman and Amos Tversky described two investors. Mr. Paul owns shares in company A. He considered switching to company B, decided against it, and now learns he would have been $1,200 better off had he switched. Mr. George owned shares in company B, switched to company A, and now learns he would have been $1,200 better off had he stayed. Both lost the same amount against the alternative. Asked who feels more regret, 92% of respondents said Mr. George, the one who acted.

The reason usually given is that an action is easy to undo in the imagination. "If only I had left it alone" comes more readily than "if only I had done something".

Thomas Gilovich and Victoria Medvec then found that time reverses the pattern. Asked to compare their most regrettable action and inaction of the past week, people split almost evenly (53% regretted the action more). Asked the same about their whole lives, 84% regretted the inaction more. In a separate telephone survey, 75% of people said that, looking back, they regretted the things they had not done more than the things they had done. Regret over an action is sharp and fades. Regret over inaction grows.

For an investor the two findings pull in opposite directions at different horizons. Next month, a purchase that fell will sting more than a purchase not made. In twenty years, the decade spent in cash is what people describe.

Decisions bend toward not finding out

Marcel Zeelenberg and colleagues showed in the 1990s that people do not simply choose the safer option to avoid regret. They choose the option that protects them from learning that the other one was better. When feedback on the risky option was guaranteed either way, people became more willing to take it. Regret aversion is partly a preference for not knowing.

Evidence from brokerage accounts

Michal Strahilevitz, Terrance Odean and Brad Barber studied which stocks investors buy again after selling. Investors were reluctant to repurchase a stock they had sold at a loss, and reluctant to repurchase one that had risen since they sold it. Neither fact says anything about the stock's prospects. Buying back at a higher price means admitting the sale was a mistake, and people avoid the admission.

What it looks like in an account

  • Waiting for a better entry. Cash sits while the investor waits for a dip, because buying just before a fall would be the worst feeling. Each rise makes buying harder.
  • Holding losers. A sale makes the loss final and makes the error one's own. This is one explanation of the disposition effect.
  • Not buying back. The stock sold at $40 is now $55. Buying it again feels like paying for a mistake twice.
  • Choosing what everyone else chose. A loss shared with the crowd carries less blame than a loss taken alone, which is part of herd behavior and FOMO.
  • Half measures. Selling half, buying half, splitting an allocation 50/50 when unsure.

What the record shows: regret by choice

To see what each choice risks, take one year as the horizon and three simple options: all in stocks (SPY, with dividends), all in three-month Treasury bills, or half in each. After a year, measure regret the plain way: how many percentage points you ended behind whichever of stocks or bills did better.

Regret after one year: how far each choice ended behind the better one Worst case Average Share of years behind by over 10 points 44% 3% 13% All in stocks 28% 8% 30% Half and half 56% 13% 56% All in Treasury bills SPY total return against 3-month Treasury bills, 393 overlapping 12-month windows from March 1993.
Regret after one year for three choices: percentage points behind the better of stocks and Treasury bills. SPY total return, 393 overlapping 12-month periods from March 1993.
Choice Worst regret Average regret Any regret Behind by over 10 points
All in stocks 44 points (from March 2008) 3.2 points 21% of periods 13%
Half and half 28 points (from April 2020) 8.0 points 100% 30%
All in Treasury bills 56 points (from April 2020) 12.8 points 79% 56%

Three things stand out.

Cash was the high-regret choice by every measure. The option that feels safest left its holder behind in 79% of one-year periods, and the single worst regret in the table belongs to it: the 12 months from April 2020, when stocks returned 56% and bills almost nothing.

Half and half cut the worst case and made some regret certain. It can never be the best choice, so it finished behind every single time, by 8 points on average. It also never finished more than 28 points behind. That is the trade a split offers.

All in stocks had the rarest regret and a severe worst case. In one period in five, stocks lost to bills, and from March 2008 the gap was 44 points.

These are overlapping periods from one country over 33 years in which stocks did well, so they describe this stretch of history and do not forecast the next. A one-year horizon also flatters cash compared with longer ones. And regret is not the same as risk: the stock investor who ended 44 points behind had also lost a great deal of money.

The price of the half measure

Harry Markowitz, who won a Nobel prize for the mathematics of portfolio selection, was asked how he had invested his own retirement account as a young man. As he told the journalist Jason Zweig, he pictured his grief if stocks rose and he held none, or fell and he held only stocks, and so split his contributions evenly between stocks and bonds to minimize future regret.

The same motive drives the most common staged entry, dollar-cost averaging a lump sum. Vanguard studied investing a sum at once against spreading it over 12 months, using US returns from 1926 to 2011. The lump sum came out ahead about two-thirds of the time, by about 2.3% of the sum on average for a 60/40 portfolio. A 2023 update using global data found much the same proportion, about 68%. Lump sum versus dollar-cost averaging works through the numbers.

Put as a price: for $100,000, spreading the purchase over a year has historically cost about $2,300 on average, in exchange for removing the outcome where the whole sum is invested the day before a fall. That is an insurance premium against regret. Whether it is worth paying depends on the alternative. If the realistic alternative to staging is investing it all today, staging costs a little on average. If the realistic alternative is leaving it in cash for three years while waiting to feel sure, staging is far better.

What the evidence suggests helps

None of these has been shown in a controlled study to improve investors' returns. They follow from the mechanisms above.

  1. Decide the rule before the outcome. A schedule ("one-sixth on the first of each month for six months") turns later purchases into the execution of a plan. People feel less regret over outcomes of a policy than over a one-off choice.
  2. Record the reason. Regret is sharpest when the decision looks foolish in hindsight. A dated note of what you knew, as in investment decision journal, lets you judge the decision by its process.
  3. Count the regret of inaction. The short-run feeling favors doing nothing. The long-run evidence says inaction is what people regret. Naming both makes the comparison fairer.
  4. Split deliberately and know the cost. A half measure is reasonable when the largest possible regret would otherwise push you out of the plan.
  5. Stop tracking what you sold. The price of a stock you no longer own is information about that stock and none about your current portfolio.

How Finabase helps

The Price history tab lists daily closes for periods up to two years, enough to price a staged entry against a single one on real dates. A paper-trading account lets you run both versions side by side with no money at stake. A price alert replaces watching a stock you are waiting to buy, which removes the daily occasion for second-guessing.

Frequently asked questions

What is regret aversion in investing?

It is the tendency to choose the option that minimizes the chance of later feeling you made the wrong decision, even at a cost in expected return. It shows up as waiting in cash, holding losers, following the crowd and splitting decisions.

Is regret aversion the same as loss aversion?

No. Loss aversion is about outcomes: a loss hurts more than an equal gain pleases. Regret aversion is about responsibility: an outcome hurts more when a different choice of yours would have avoided it. You can feel regret after a gain, if the alternative gained more.

Is dollar-cost averaging a lump sum irrational?

Historically it has cost a little on average, since a lump sum came out ahead about two-thirds of the time. It also limits the worst outcome. Paying a known, modest cost to stay with a plan can be reasonable.

Do people regret action or inaction more?

Both, at different times. Soon afterward, actions that turned out badly are regretted more. Over a lifetime, people mostly regret what they did not do.

Summary

Regret aversion is deciding so as not to feel responsible for a bad result. Experiments show that actions are regretted more than inactions at first and less in the long run, and that people steer toward choices that spare them from learning what the alternative would have done. In accounts it appears as cash held for a better moment, losers kept, sold stocks never bought back, and half measures. Over 393 one-year periods the apparently safe choice, cash, carried the most regret by every measure. A split reduces the worst regret at the price of a certain small one, and the sound way to use it is on purpose, with the cost known.

Check your understanding

5 questions on this lesson. Each answer comes with the reason, and nothing is scored.

  1. Two investors each end $1,200 worse off than the alternative. One switched stocks; the other considered switching and did not. Who do most people say feels more regret?

    1. The one who switched
    2. The one who did nothing
    3. Both equally, since the money is the same
    Show the answer

    A. The one who switched In Kahneman and Tversky’s example 92% chose the investor who acted. In the short run, bad results from action are regretted more than equal results from inaction.

  2. How does that pattern change when people look back over many years?

    1. It gets stronger
    2. It reverses: most long-run regrets are about things not done
    3. Regret disappears
    Show the answer

    B. It reverses: most long-run regrets are about things not done Asked about their whole lives, 84% of Gilovich and Medvec’s subjects regretted their biggest inaction more than their biggest action. For investors, the purchase that fell stings now and the years in cash sting later.

  3. Over 393 one-year periods since 1993, which of three choices finished behind the better of stocks and Treasury bills by more than 10 points most often?

    1. All in stocks
    2. Half and half
    3. All in Treasury bills
    Show the answer

    C. All in Treasury bills Bills were more than 10 points behind in 56% of periods, against 30% for the split and 13% for stocks. The choice that feels safest carried the most regret.

  4. What does splitting money half and half between two options do to regret?

    1. It removes regret
    2. It guarantees some regret every time and limits how large it can be
    3. It maximizes regret
    Show the answer

    B. It guarantees some regret every time and limits how large it can be A split can never be the best choice, so it is always somewhat behind. It also can never be the worst. That trade has a cost and can still be reasonable.

  5. An investor sold a stock at $40. It is now $55 and they still think the company is attractive, but they will not buy it back. What is the most likely explanation?

    1. The stock has become riskier
    2. Buying back at a higher price would mean admitting the sale was a mistake
    3. Brokers do not allow repurchases
    Show the answer

    B. Buying back at a higher price would mean admitting the sale was a mistake Brokerage records show investors avoid repurchasing stocks that have risen since they sold. The earlier sale price says nothing about the stock’s prospects from here.

Exercise

One task to do in Finabase. Nothing is graded or saved.

Open SPY’s Price history tab and choose the 1Y period. Imagine you had $12,000 a year ago. Plan A: buy at the first close shown. Plan B: buy $1,000 at the first close of each of twelve months. Using the Close column, work out roughly how many shares each plan bought and what each is worth at the latest close. Before you calculate, write down which plan you would have felt better about on the day you started.

Open SPY price history
What you should find

In a year when the price mostly rose, Plan A ends ahead, because Plan B bought most of its shares at higher prices. In a year with an early fall, Plan B ends ahead. Across history the single purchase has won about two times in three. The plan that felt better at the start is usually B: that feeling is what the average shortfall pays for.

Sources and further reading

  1. Loomes and Sugden — Regret theory: an alternative theory of rational choice under uncertainty, Economic Journal 92(368), 1982
  2. Bell — Regret in decision making under uncertainty, Operations Research 30(5), 1982
  3. Kahneman and Tversky — The psychology of preferences, Scientific American 246(1), 1982
  4. Gilovich and Medvec — The temporal pattern to the experience of regret, Journal of Personality and Social Psychology 67(3), 1994 (the 53%, 84% and 75% figures)
  5. Gilovich and Medvec — The experience of regret: what, when, and why, Psychological Review 102(2), 1995
  6. Zeelenberg, Beattie, van der Pligt and de Vries — Consequences of regret aversion: effects of expected feedback on risky decision making, Organizational Behavior and Human Decision Processes 65(2), 1996
  7. Strahilevitz, Odean and Barber — Once burned, twice shy: how naive learning, counterfactuals, and regret affect the repurchase of stocks previously sold, Journal of Marketing Research 48(SPL), 2011
  8. Vanguard — Dollar-cost averaging just means taking risk later, 2012
  9. Vanguard — Cost averaging: invest now or temporarily hold your cash?, 2023
  10. Zweig — Your Money and Your Brain, Simon & Schuster, 2007 (interview with Harry Markowitz)