Skip to content
Menu
Finabase Learn

Investor psychology Lesson 9 of 15

Articles · Behavioral finance

The framing effect and narrow framing: same facts, different choices

How the wording and the time period of the same information change investment choices, what the experiments found, and how to reframe points, percents and periods.

On this page
  1. The original experiment
  2. Frame one: points or percent
  3. Frame two: the period you are shown
  4. Frame three: one holding or the whole portfolio
  5. What the evidence suggests helps
  6. How Finabase helps
  7. Frequently asked questions
  8. Summary
  9. Check your understanding
  10. Exercise

The framing effect is a change in what people choose when the same facts are described differently. Narrow framing is a particular case: judging each investment, or each short period, on its own instead of as part of the whole portfolio and the whole horizon.

A rational chooser would be unaffected by either. "Down 3%" and "down 1,300 points" describe one event. A fund's last year and its last thirty years are views of one fund. Experiments show people choose differently depending on which they are shown, and investors are shown frames all day, chosen by whoever writes the headline or designs the statement.

The original experiment

In 1981 Amos Tversky and Daniel Kahneman gave people a scenario: a disease is expected to kill 600 people, and there are two programs to choose from.

One group read the programs in terms of lives saved. Program A saves 200 people for certain. Program B gives a one-third chance of saving all 600 and a two-thirds chance of saving nobody. Of 152 people, 72% chose A, the sure thing.

A second group read the same programs in terms of deaths. Under program C, 400 people die. Under program D there is a one-third chance that nobody dies and a two-thirds chance that 600 die. Of 155 people, 78% chose D, the gamble.

A and C are the same program. So are B and D. Worded as gains, people preferred certainty. Worded as losses, they preferred to gamble. This is the pattern prospect theory predicts: caution about gains and risk-seeking about losses.

The result has held up. A large 2014 project repeated it in 36 samples across several countries and found the effect again, though smaller than in the original study.

Irwin Levin and colleagues later sorted framing effects into three kinds. The disease problem is a risky-choice frame. Attribute framing changes how one feature is described: people rated beef labeled "75% lean" more favorably than the same beef labeled "25% fat". Goal framing changes whether a message stresses what you gain by acting or lose by not acting. All three appear in finance. A fund that "beat its benchmark in 7 of the last 10 years" and one that "lagged in 3 of the last 10" are the same fund.

Frame one: points or percent

Index moves are reported in points, and points grow as the index grows. A fall that was a historic crash at one level of the index is an ordinary bad day at another.

Six bad days for the Dow Jones Industrial Average, framed two ways Points lost Percent lost Oct 28, 1929 38 12.8% Oct 19, 1987 508 22.6% Mar 9, 2020 2,014 7.8% Mar 12, 2020 2,353 10.0% Mar 16, 2020 2,997 12.9% Apr 4, 2025 2,231 5.5% Close-to-close falls. The headline frame (points) and the frame that matters to a portfolio (percent) rank the days differently.
Six falls in the Dow Jones Industrial Average, close to close, in points and in percent. The two frames rank the days in a different order.

In points, April 4, 2025 (2,231) was a larger fall than October 19, 1987 (508) by a factor of more than four. In percent, 1987 was more than four times larger: 22.6% against 5.5%. The 38-point fall of October 28, 1929 would not make the news today, and it took 12.8% off the index.

Percent is the frame that matters to a portfolio, because your holdings move in proportion. On a $50,000 holding that tracks the index, the 2025 day cost $2,750 and the 1987 day would have cost $11,300. "Largest point drop ever" headlines become more frequent as an index rises, without markets becoming more dangerous.

The same applies to share prices. A $5 move in a $500 stock is 1%. A $5 move in a $20 stock is 25%. It applies to costs as well: a 1% annual fee sounds small as a percentage and is $1,000 a year on $100,000, every year, whatever the return.

Frame two: the period you are shown

Shlomo Benartzi and Richard Thaler asked university employees how they would divide retirement savings between a stock fund and a bond fund. One group saw a chart of the distribution of one-year returns for each. Another saw the same history presented as simulated 30-year returns. The first group put a median of 40% in stocks. The second put 90%. The data were the same in both charts and the reporting period differed.

Two laboratory experiments from 1997 reached the same conclusion with real money. Uri Gneezy and Jan Potters had people bet repeatedly on a gamble with a positive expected return. Those who saw results and could change their bet every round wagered 50.5% of their stake on average. Those who saw results only every three rounds wagered 67.4%. Thaler, Tversky, Kahneman and Alan Schwartz had people allocate between a stock-like and a bond-like fund: those shown results monthly put 41% in the stock fund, those shown yearly put 70%, and those shown five-yearly results put 66%.

Seeing results more often led to less risk-taking and, in these experiments, to lower earnings. The mechanism is loss aversion meeting short periods: the shorter the period, the more often it shows a loss.

The same fund, four reporting periods: how often SPY showed a loss Each day (last two years) 46% of periods Each month 35% of periods Each 12 months 18% of periods Each 10 years 8% of periods Daily: closing prices over two years. Longer periods: total return, all windows from March 1993.
Share of periods in which SPY showed a loss, by length of period. Daily figures use closing prices over two years; longer periods use total return over all periods since February 1993.

Over the two years to late October 2026, SPY's price rose about 24%, and it still closed lower on 46% of days. An investor who looked every day saw a loss nearly every other look. One who looked monthly since 1993 saw a loss about one time in three, yearly about one time in six, and over ten-year periods about one time in twelve. The fund is the same in each row. The figures describe this fund in this period and are no promise about future decades.

Loss aversion covers this interaction, known as myopic loss aversion, in more detail.

Frame three: one holding or the whole portfolio

Kahneman and Dan Lovallo described narrow framing in 1993 as treating each risky choice as if it were the only one. Nicholas Barberis, Ming Huang and Thaler later used it to explain why many people decline small favorable bets and hold few or no stocks: each risk is judged alone, and alone, any risky asset looks alarming.

A worked example shows the difference. A portfolio holds two funds, $60,000 in stocks and $40,000 in bonds. In a year the stock fund falls 15% and the bond fund rises 5%.

Frame What you see
Stock fund alone −$9,000 (−15%)
Bond fund alone +$2,000 (+5%)
Whole portfolio −$7,000 (−7%)

Seen line by line, one holding is "failing" and invites action: sell it, or stop adding to it. Seen as a whole, the portfolio lost 7% in a year when stocks lost 15%, which is what a 60/40 mix is built to do. The bond fund was held for that year. Judging each line alone leads investors to remove whichever part is currently doing its job least visibly. Mental accounting is the related habit of treating pots of money as if they were separate.

What the evidence suggests helps

The experiments above change behavior by changing the frame, which is evidence that reframing works on choices made in a laboratory. Whether a habit of reframing improves an investor's long-run results has not been tested directly.

  1. Translate points to percent, and percent to dollars. Do it for market moves, for single stocks and for fees.
  2. Match the reporting period to the purpose of the money. For money needed in twenty years, yearly or longer views carry the relevant information. Daily views are mostly noise.
  3. Look at the portfolio before the positions. Total value and total return first, lines second.
  4. Restate every claim in the opposite frame. "Beat the market in 7 of 10 years" becomes "lagged in 3 of 10". "90% of clients are satisfied" becomes "1 in 10 is not".
  5. Ask who chose the frame. A headline is written to be read and a fund advertisement to sell. Neither is written to help you size a position.

Reframing should not be used to hide risk. A long-term chart does not make a 40% decline painless when it arrives, and an investor who can only hold stocks by never looking has a plan that depends on not finding out.

How Finabase helps

The Indices page lists each index with its change in both points and percent, so the translation is already done. A symbol's chart switches between periods from one day to many years, which puts a bad day beside the period that matters for long-term money. A portfolio shows total value and total gain above the individual holdings, the broad frame first.

Frequently asked questions

What is the framing effect?

It is a change in people's choices caused by how options are described and not by any change in the options themselves. The same outcome worded as a gain or as a loss, or the same return shown over one year or thirty, leads to different decisions.

What is narrow framing in investing?

It is evaluating each investment or each short period by itself instead of looking at the whole portfolio over the whole horizon. It makes individual risks look worse than they are in combination.

Why are stock market falls reported in points?

Convention, and because large numbers make stronger headlines. Points are not comparable over time: a 500-point fall was 22.6% of the Dow in 1987 and about 1% in recent years. Percent changes are comparable.

How often should I check my portfolio?

There is no single right answer. Experiments show that people who see results more often take less risk, because short periods show losses more often. A common approach is to match the review schedule to the money's purpose, for example quarterly or yearly for long-term savings.

Summary

The same facts lead to different choices when they are worded as gains or losses, quoted in points or percent, shown over a day or a decade, or presented one holding at a time. In experiments the reporting period alone moved the median stock allocation from 40% to 90%. The Dow's record point falls are far smaller in percent than 1987 or 1929, and SPY has shown a loss on nearly half of days and in about a twelfth of ten-year periods. Since some frame is always in use, the practical habit is to restate any figure a second way (in percent, over a longer period, for the whole portfolio) before deciding what it means.

Check your understanding

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

  1. A headline reads “Dow suffers one of its largest point drops ever: down 1,200”. The index stood near 48,000. How large was the fall in the frame that matters to a portfolio?

    1. About 2.5%
    2. About 12%
    3. About 25%
    Show the answer

    A. About 2.5% 1,200 ÷ 48,000 = 2.5%. Point records become more frequent as an index rises. The 508-point fall of 1987 was 22.6%.

  2. Savers shown a chart of one-year stock returns put a median 40% in stocks; those shown 30-year returns from the same history put 90%. What differed between the groups?

    1. The underlying data
    2. Only the period over which the same data were presented
    3. Their age
    Show the answer

    B. Only the period over which the same data were presented Benartzi and Thaler changed nothing but the reporting period. Short periods display losses often, and losses weigh heavily.

  3. SPY rose about 24% over two years. On roughly what share of days did it close lower?

    1. 10%
    2. 25%
    3. 46%
    Show the answer

    C. 46% Nearly half of days were down in a strongly rising period. An investor who checks daily sees a loss almost every other look.

  4. A $100,000 portfolio is 60% stocks and 40% bonds. Stocks fall 15% and bonds rise 5%. What does the broad frame show?

    1. A 15% loss
    2. A 7% loss on the whole portfolio
    3. A 5% gain
    Show the answer

    B. A 7% loss on the whole portfolio −$9,000 + $2,000 = −$7,000, or −7%. Judged line by line, the stock fund looks like a failure. Judged together, the mix did what it was built to do.

  5. A fund advertisement says it “beat its benchmark in 7 of the last 10 years”. What is the same fact in the opposite frame?

    1. It lagged its benchmark in 3 of the last 10 years
    2. It beat its benchmark by 70%
    3. It will beat its benchmark 70% of the time in future
    Show the answer

    A. It lagged its benchmark in 3 of the last 10 years Restating a claim in the opposite frame adds no information and often changes how it feels. Neither version says by how much it won or lost, which matters more.

Exercise

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

Open SPY’s Price history tab and choose the 3M period. Count the days with a negative Change %, and divide by the number of days shown. Then compare the first and last Close in the list to get the change over the whole three months. Write both down: the share of down days, and the three-month change.

Open SPY price history
What you should find

In most three-month periods between 40% and 50% of days are down, whether the period as a whole rose or fell. The daily frame and the three-month frame describe the same prices. If the share of down days surprised you, that gap is what checking daily adds to how a holding feels without adding information about where it ended.

Sources and further reading

  1. Tversky and Kahneman — The framing of decisions and the psychology of choice, Science 211(4481), 1981
  2. Klein et al. — Investigating variation in replicability: a “many labs” replication project, Social Psychology 45(3), 2014
  3. Levin, Schneider and Gaeth — All frames are not created equal: a typology and critical analysis of framing effects, Organizational Behavior and Human Decision Processes 76(2), 1998
  4. Levin and Gaeth — How consumers are affected by the framing of attribute information before and after consuming the product, Journal of Consumer Research 15(3), 1988
  5. Benartzi and Thaler — Risk aversion or myopia? Choices in repeated gambles and retirement investments, Management Science 45(3), 1999
  6. Gneezy and Potters — An experiment on risk taking and evaluation periods, Quarterly Journal of Economics 112(2), 1997
  7. Thaler, Tversky, Kahneman and Schwartz — The effect of myopia and loss aversion on risk taking: an experimental test, Quarterly Journal of Economics 112(2), 1997
  8. Kahneman and Lovallo — Timid choices and bold forecasts: a cognitive perspective on risk taking, Management Science 39(1), 1993
  9. Barberis, Huang and Thaler — Individual preferences, monetary gambles, and stock market participation: a case for narrow framing, American Economic Review 96(4), 2006
  10. S&P Dow Jones Indices — Dow Jones Industrial Average, historical closing values