Studies

Research summaries for US savers, adapted from the Swedish RikaTillsammans study notes. Study names, authors and findings follow the source. Items marked US review need a US fact check before publishing.

Active versus passive

Can an AI invest better than an index fund?

A classic buy and hold strategy beat the AI agents on risk adjusted return in every stock selection tested, using simulated historical data from 2004 to 2024. Buy and hold scored a Sharpe ratio of 0.70, while the best AI agent's score was negative. The AI produced no statistically significant excess return, and more advanced models did not do better.

The simple beat the clever, once again. For a saver, a chatbot or "AI portfolio" is no reason to leave a low cost index fund.

Source: Li, W.W., Kim, H., Cucuringu, M. and Ma, T. (2026) "Can LLM-based Financial Investing Strategies Outperform the Market in Long Run?", Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '26). New York: Association for Computing Machinery. https://arxiv.org/abs/2505.07078. doi: 10.1145/3770854.3785702.

Good returns, or just a good feeling?

A British report reviewed more than 550 portfolios at 110 managers and set two numbers against each other: 96% of savers felt confident that things had gone well, while 84% of the managers failed to beat even a cheap, similar index basket in 2025. The pattern held three years running (92% in 2023, 88% in 2024), and underperformance averaged 4.3 percentage points in the latest year. Portfolios mostly rose in value. The point is the return a simple index would have delivered instead.

The report comes from a commercial wealth firm and uses its own unreviewed benchmark, but the same pattern appears in independent SPIVA data.

The takeaway: measure your savings against a cheap broad index fund, not against zero and not against your advisor's words.

Source: Y TREE report on portfolios at 110 managers (2025 results), as summarized in the original article.

Stop choosing the funds that just did best

Morningstar star ratings mainly capture luck, and luck does not repeat. In a study of 3,662 US stock funds over 30 years, five star funds underperformed one star funds by 0.075 percentage points per month over the following twelve months, a small but statistically significant difference. Funds that had good luck in the last year also tend to attract more money.

The conclusion is not to buy one star funds. It is that stars and past returns do not work as a buy signal. Choose low fees, broad exposure and a strategy you can keep.

Source: Gao, J. and Yao, J. (2025) Luck and Skill in the World of Diseconomies of Scale. Working Paper. Sydney: The University of Sydney Business School. https://doi.org/10.2139/ssrn.5211928.

Active funds rarely beat the index over time

Only 21% of US active funds both survived and beat a cheap passive counterpart after fees over ten years. The cheapest active funds succeeded almost twice as often as the most expensive, 27% against 15%. Over one year, 33% beat their passive peers, down 14 points from 47% the year before.

The measurement through June 2025 covers 9,204 funds with about 24 trillion dollars in assets, roughly 68% of the US fund market. The message is not to search harder for the winner but to avoid the lottery: choose broad, cheap, and global.

Source: Armour, B. et al., 2025. "Morningstar's US Active/Passive Barometer Midyear 2025." Morningstar.

Index funds and the 500 billion dollar fee saving

US index funds had an average fee of 0.09% against 0.56% for active funds as of December 31, 2024, a difference of 0.47 percentage points. The report says index funds have saved investors around 500 billion dollars in fees since 2000, but that is a hypothetical calculation, not an observed saving.

The report answers six common criticisms, including whether index funds can grow so large they harm the market, using US fund market data from 1990 to 2024. Remember who wrote it: Vanguard is the world's largest index manager and is defending its own business. The principle holds regardless: choose broad, cheap and global.

Source: Rowley, J.J. et al., 2025. "Setting the record straight: The truths about index fund investing." Vanguard Research.

Economists agree: an index fund beats a few stocks

In two expert panels at Chicago Booth, one European in March 2017 and one US in January 2019, 80 of 80 responding economists agreed that a broad, cheap, passive index fund will usually beat a handful of individual stocks for someone without inside information. Of 92 asked, 80 answered: 51 chose "Strongly Agree" and 29 "Agree." Nobody was uncertain and nobody disagreed. Average confidence was about 8 out of 10, and Nobel laureate Richard Thaler summed the advice up as "Investing 101."

The survey measures expert consensus, not returns, so it is best paired with the empirical evidence from SPIVA and from Bessembinder.

Source: Kent A. Clark Center for Global Markets at Chicago Booth (formerly the IGM Forum), expert panels of March 2017 (Europe) and January 2019 (US).

Twenty years of data contradict the myth of consistent winners

After 20 years of half yearly measurements across more than 100 fund categories and 9 geographies, S&P's conclusion is that the chance of beating the index falls the longer the measurement period, and past outperformance does not tend to persist. This does not mean no active fund can win. It means picking the winner in advance is hard.

For most US savers, a broad, low cost index fund is therefore a more reasonable base than the hunt for a star manager. The summary uses data through June 2022 and contains no new tables.

Source: S&P Dow Jones Indices, "TalkingPoints: Assessing the Impact of 20 Years of SPIVA", S&P Global, 2023.

The incredible shrinking alpha: why the index wins more often

Fund managers have never been more skilled, yet it gets harder to beat the index, because competition grows faster than skill. When old alpha becomes cheap beta and the less experienced counterparties leave the market, professionals meet mostly other professionals. The skill exists before fees, but the net after costs turned negative.

The authors note that the share of managers beating the index after fees fell from about 20% in the early 1990s to 1.6% in 2011. For most savers the conclusion points to cheap index funds and discipline.

Source: Swedroe, L. and Berkin, A., "The Incredible Shrinking Alpha".

Small investors behave badly when choosing funds

Research on 3,341 US funds found that Morningstar star ratings are the strongest explanation for which funds receive inflows, stronger than raw returns or risk adjusted alpha. Five star funds drew in 656 billion dollars during 1997 to 2011, twice as much as funds topping the raw return rankings. When Morningstar changed its rating method in 2002, 54% of funds changed rating without a change in saver behavior.

The study shows the behavior, not the cost in dollars. The practical fix is fewer fund decisions and more automation.

Source: Ben-David, I., Li, J., Rossi, A. and Song, Y. (2022) "What Do Mutual Fund Investors Really Care About?", Review of Financial Studies.

Eight in ten top rated funds get worse

Five star funds may already have had their best period. In a study of 273 US stock funds, 77 to 95% performed worse in the three years after their first five star Morningstar rating, depending on the measure. The drop was not explained by higher fees, which actually fell. It points more toward mean reversion, higher risk and possible problems with large inflows.

The sample was diversified US stock funds that first earned five stars between 1993 and 1998.

Source: Morey, M.R. (2003) "Kiss of Death: A 5-Star Morningstar Mutual Fund Rating?"

Focus on low fees, not fund ratings

Ratings are a weak buy signal: in Vanguard's study, 39% of five star funds beat their benchmark over the following 36 months, against 46% of one star funds. This does not mean low ratings are best, only that the stars did not predict the future.

Stars look backward, built on historical risk adjusted returns within a category. The problem arises when we read the rating as a forecast. The fee is a better starting point, because it is known in advance and charged every year.

Source: Philips, C.B. and Kinniry Jr., F.M., "Mutual Fund Ratings and Future Performance", Vanguard.

Fund managers create marginal or no value

Mark Carhart followed 1,892 US stock funds from 1962 to 1993, a total of 16,109 fund years, including the roughly one third of funds that closed during the period, so the study is free of survivorship bias. Almost all of the difference between winner and loser funds was explained by momentum, fees and trading costs, not skill. Each extra percentage point of expense lowered risk adjusted return by about 1.5 percentage points a year, and the only pattern that persisted was that the worst funds kept being the worst.

The lesson is to choose cheap, broad and passive instead of chasing last year's winner.

Source: Carhart, M. (1997) "On Persistence in Mutual Fund Performance", The Journal of Finance.

Invest it all at once or spread it out over time?

Spreading out purchases of a lump sum feels safe but is in practice a bet against the market. The money waiting misses the risk premium, and stocks beat cash 76% of the time from 1976 to 2022. In Vanguard's review, investing right away produced a median of 1.8% more at the end of one year than spreading over three months in a 60/40 portfolio, and won in 68% of cases.

If you must split the purchases to dare to invest, keep the period short, three months at most. The study draws on global market data from 1976 to 2022 and 10,000 simulated scenarios, and comes from a fund company whose business benefits from invested money.

Source: Vanguard, "Cost averaging: Invest now or temporarily hold your cash?" (2023), Finlay and Zorn.

When everyone gets better, luck matters more

The gap between the best and the average fund manager has shrunk for five decades: the spread of excess returns among US large cap funds fell from roughly 13% in 1967 to roughly 7% in 2012, according to Mauboussin's report. When professionals become more even, luck decides more of who beats the index. Cornell (2009) attributes about 92% of the variation in annual fund performance to chance, and SPIVA 2024 shows 65% of active US large cap funds failed to beat their index.

The mechanism is called the paradox of skill. For a saver it favors a broad, cheap, global index fund, where you do not have to find the winner in advance.

Source: Mauboussin, M., "Alpha and the Paradox of Skill", Credit Suisse, 2013.

Six in ten stocks do not beat the risk free rate

Bessembinder and colleagues followed 64,738 stocks in 43 markets from 1990 to 2020, covering about 88% of world market value. Only 43.2% beat the risk free rate over their lifetime. Meanwhile 2.39% of companies created the entire global net wealth of 75.66 trillion dollars.

That is not an argument against stocks but against owning too few of them. A broad index fund captures the winners without knowing in advance which they are.

Source: Bessembinder, H., Chen, T.-F., Choi, G. and Wei, K.C.J., "Long-Term Shareholder Returns: Evidence from 64,000 Global Stocks", Financial Analysts Journal, 2023.

Bessembinder: most stocks do not beat Treasury bills

Bessembinder went through all 25,967 US listed stocks from 1926 to 2016, over 3.5 million monthly returns, and compared each stock's lifetime return with one month Treasury bills. Only 42.6% beat Treasury bills over their lifetime, and 4.3% of companies created all the net wealth.

Individual stocks are far more lottery like than the market average suggests. The conclusion is not that stocks are bad, but that concentration is risky. A broad index fund means you do not miss the few winners.

Source: Bessembinder, H. (2018) "Do Stocks Outperform Treasury Bills?", Journal of Financial Economics (forthcoming), SSRN abstract 2900447, doi: 10.2139/ssrn.2900447.

Active management costs 0.67 percentage points a year

In 2008 Kenneth French put a price tag on active management: on average it cost 0.67 percentage points a year on the US stock market from 1980 to 2006. It is not about ignorant managers but arithmetic. Active investors together own the market before costs, so the average manager ends up below the index after costs.

French looked at what all investors together paid to try to beat the market, adding fund fees, institutional management, hedge funds and trading costs against what it would have cost to simply hold the market. The mechanics speak strongly for broad, cheap index funds as the base.

Source: French, K.R., 2008. Presidential Address: "The cost of active investing", The Journal of Finance.

Can small investors beat the market? 35 of 36 professors answer no

In January 2019 the Chicago Booth IGM finance panel, made up of named finance professors at top US universities, took a position on two statements. Of the 36 who responded, 35 agreed on both, and none disagreed: a small investor cannot consistently predict individual stock moves, nor expect better results from a few stocks than from a broad, cheap, passive index fund. Average confidence among those agreeing was 8.6 and 8.5 out of 10.

This is expert consensus, not new return data. The practical advice: make the index fund your base and keep individual stocks in a small sandbox.

Source: Chicago Booth IGM Forum, Finance panel, January 2019.

Bank fund managers avoid their own funds

In interviews with 70 active managers and analysts in London, Chicago and New York between 2019 and 2021, several admitted they would rather recommend index funds than the funds they manage themselves. One said outright: "All my own money is in index funds." Fewer than 5 of the 70 had done anything to adapt to the rise of index funds. Instead the industry defends itself with four recurring arguments, which the researchers call epistemic opportunism.

The study does not measure how large a share avoid their own funds, so read the quotes as telling voices, not statistics. It does not measure returns either.

Source: Millo, Y., Spence, C. and Valentine, J. (2023), Economy and Society.

Only 2% of winning funds are still top half five years later

Only 2.4% of US large cap funds that were in the top half in 2020 stayed in the top half every year through 2024, against 6.25% if chance had decided. None of the top quartile funds stayed on top. Active funds can certainly win in individual years, but very few keep winning.

Picking funds on past returns did worse than chance. Focus instead on what you can know in advance: low fees, broad diversification and a strategy you can hold.

Source: Ganti, A.R. and Edwards, T. (2025, March 4). "SPIVA U.S. Year-End 2024." S&P Dow Jones Indices (U.S. Persistence Scorecard).

Is investing luck or skill?

Fama and French followed 3,156 US stock funds from January 1984 to September 2006, 273 months. After fees, the active fund collective underperformed the market by about its entire fee, equal to minus 0.81% to minus 1.00% a year. The roughly 30% of funds that did beat the index were rarely distinguishable from pure luck, and the spread of true skill was only about 1.25% a year before fees.

That makes a broad, cheap index fund a reasonable default.

Source: Fama, E.F. and French, K.R. (2010) "Luck versus Skill in the Cross-Section of Mutual Fund Returns", The Journal of Finance.

The math: active management must lose in the long run

William Sharpe showed in 1991 that active and passive investors as a group earn exactly the same market return before costs. Therefore the average active dollar must fall behind once fees are deducted, in any market and any period. Some active managers beat the index even after costs, but they are a minority that nobody can pick out in advance.

The essay is three pages long and contains no tables or fund data, yet it is one of the strongest arguments for cheap index funds. Only the four basic operations of arithmetic are needed.

Source: Sharpe, W.F. (1991) "The arithmetic of active management", Financial Analysts Journal.

99% of professionals are worse than an index fund

Skill exists, but fees eat it. Before fees, 9.6% of fund managers were skilled stock pickers; after fees only 0.6%, a share not even statistically different from zero. The study followed 2,076 US actively managed stock funds from 1975 to 2006, so 0.6% is 12 funds out of 2,076.

The point is not that managers are bad at their jobs. The point is that fees consume the skill before it reaches the saver. For most savers, begin with a broad, cheap and global index solution and deviate only when the reasons are strong enough.

Source: Barras, L., Scaillet, O. and Wermers, R. (2010), Journal of Finance; working paper "False Discoveries in Mutual Fund Performance: Measuring Luck in Estimated Alphas" (April 20, 2009).

The myth of the five star fund

The Wall Street Journal reviewed about three million fund ratings over 14 years: only 12% of funds with Morningstar's top rating were still on top five years later, while 10% fell to one star. Money follows the stars anyway: five star funds attracted new money in 69% of the months they held the rating, against 29% for one star funds.

A top rating is history, not future. Morningstar itself is clear that the star rating looks backward, built on historical risk adjusted return within a category.

Source: Grind, K., McGinty, T. and Krouse, S., "The Morningstar Mirage", The Wall Street Journal, October 25, 2017. https://www.wsj.com/articles/the-morningstar-mirage-1508944334.

Kenneth French: five things I know about saving

The finance professor behind the Fama French model saves exactly as simply as the research recommends: broad, cheap and patient. His sharpest points are that the average actively managed dollar after costs loses to the index (Sharpe's arithmetic), that even a manager with 5% alpha would need 64 years of data to prove skill, and that risk is really about whether the money lasts for the life you want to live.

His argument for a home market overweight is American, which makes it directly relevant to a US saver but worth thinking through against global diversification US review.

Source: French, K.R., 2022. "Five Things I Know about Investing." Dimensional, 17 Mar. 2022.

The article that made John Bogle start the first index fund

Paul Samuelson's essay in the first issue of The Journal of Portfolio Management in 1974 noted that superior manager performance is unproven and urged a large foundation to set up a portfolio that simply tracked the S&P 500. Two years later John Bogle took the call literally and launched the First Index Investment Trust, the first index fund for ordinary savers.

Skilled managers may exist, but according to Samuelson they cannot be identified in advance. His practical advice still holds: trade less, diversify more broadly and keep costs down.

Source: Samuelson, P.A. (1974) "Challenge to Judgment", The Journal of Portfolio Management.

Ten years of history says little about skill

"This fund has beaten the index for ten years" is one of the most common sales arguments, and it sounds reasonable. Morningstar's own researchers say it is not enough. Funds that beat their index over 15 years still had continuous droughts averaging 9 to 11 years, and a manager who won with a short drought has only about 45% probability of being genuinely skilled. A truly skilled manager can trail the index for almost ten years in a row, and a poor fund can shine as long.

Past returns are weak evidence of skill, even when the period feels long. A broad, cheap index fund sidesteps the whole burden of proof by delivering the market return at low cost.

Source: Kaplan, P. and Kowara, M., "Are Relative Performance Measures Useless?", The Journal of Investing, 2019.

The Loser's Game: win by not losing

In 1975 Charles Ellis showed that an active manager must beat the market gross by well over 40 percent to give the saver 20 percent more net: costs and constant trading eat the rest. When the professionals have become the market, the outcome is decided by who makes the fewest mistakes, just as in amateur tennis, where about 80 percent of points are lost on unforced errors.

Ellis offered no statistical proof. The essay is cost logic built on a simple equation, performance data and a tennis image, which is why it has aged well. For a saver it points to a broad, cheap index fund, low turnover and fewer decisions that can go wrong.

Source: Ellis, C.D., 1975. "The Loser's Game", Financial Analysts Journal.

Strategies that beat the index tend to stop working

Strategies that beat the market historically usually look better in the rear view mirror than they work in real life. A University of Chicago study shows that patterns in old price moves, such as buying stocks that did well recently, gave much lower returns to an investor standing there in the moment than the historical tables suggest. The patterns were real and visible as early as the 1960s, but they weakened sharply around when they became widely known, and after the 2008 financial crisis they were largely gone.

The conclusion is not that the patterns were made up, but that a strategy rarely stays free money once everyone knows it. The broad, cheap index fund holds up better.

Source: Marrow, B. and Nagel, S., 2024. "Real-time discovery and tracking of return-based anomalies."

Morningstar ratings are poor at finding winners

In 0 of 24 tested comparisons did five star funds beat three and four star funds on future return, when Blake and Morey examined 135 to 184 US growth funds per year from 1992 to 1997. Low ratings were more informative: one and two star funds were significantly worse in 12 of 24 tests.

So use Morningstar ratings as a warning filter, not a buy signal. Five stars do not mean the fund has better prospects than three or four stars. Keep in mind that the study covers the old system before 2002, and the authors' own 2006 follow up found the rebuilt system predicts better US review.

Source: Blake, C.R. and Morey, M.R., "Morningstar Ratings and Mutual Fund Performance" (March 15, 1999). SSRN: https://ssrn.com/abstract=168668.

Investor behavior

The return gap: what your own timing costs

Investors in US spot bitcoin ETFs lost 5.8% a year through June 2026, while the funds they owned gained 8.5%. Through the end of 2025, the average dollar earned about 22% a year, nearly 14 percentage points less than the funds' own return. The behavior gap costs the most where swings are biggest: US stock funds showed a gap of 0.4 percentage points, but the most volatile fifth of funds showed a gap of 2.1 points.

The report also carries a more disputed number, an average behavior gap of 1.2 percentage points. Peer reviewed critiques put it closer to 0.10, so treat the 1.2 with caution.

For a US saver the lesson is practical: the more volatile the investment, the more your buying and selling decisions can cost you. A broad index fund bought on a schedule and left alone narrows the gap.

Source: Ptak, J. et al., 2026. "Mind the Gap 2026: US stock fund investors made history, crypto mavens stumbled." Morningstar.

Only three reasons to invest: what are yours?

Finance professor Meir Statman divides the value of investments into three benefits: the wallet (utilitarian), self image (expressive) and feeling (emotional). The last two are not wrong, but they get expensive when they dress up as rational return chasing.

His book maps sixteen such drivers across sixteen chapters, from the wish for profit beyond risk, through herd behavior and mental accounting, to status, values and the dream of leaving a legacy. Name the motive at each purchase, and let the emotional investments live in a small sandbox so they do not eat into your serious savings.

Source: Statman, M., 2011. "What investors really want: know what drives investor behavior and make smarter financial decisions." SCU Leavey School of Business Research Paper, no. 11-03. Available via ssrn.com/abstract=1743173.

The more savers traded, the less they earned

Over the ten years to December 31, 2024, US funds returned 8.2% a year, while the average invested dollar earned 7.0%. The gap of 1.2 percentage points a year arises when money moves into and out of funds at the wrong times. Funds with stable cash flows lost 0.8 points a year against the fund's return, while the most traded lost 1.8.

The 1.2 point total is disputed, and one replication lands at 0.10% a year. Morningstar itself warns against reading the gap as proof of "dumb money," since even healthy habits like monthly saving open a gap mathematically, because more money arrives late in a rising period. The study covers 25,848 US funds and ETFs and is an aggregate analysis, not an experiment.

The pattern holds up: automation, steady contributions and all in one solutions reduce the behavior cost.

Source: Ptak, J. et al., 2025. "Mind the gap 2025: the more investors traded, the less they made." Morningstar.

Mental accounting and saving

The same sum is saved differently depending on the form it arrives in. In the study, participants planned to spend 1,200 dollars of a sum spread out as monthly income, 785 dollars of a lump sum, and 0 dollars of money five years away. The explanation is mental accounting: the brain sorts pay, bonus and retirement money into different mental accounts with different willingness to spend.

The practical takeaway is to move money to your savings or investment account on payday, before it has time to feel spendable, so the system does the work instead of willpower. Automatic 401(k) deferrals and automatic transfers do exactly this.

Source: Shefrin, H. and Thaler, R. (1988) "The Behavioral Life-Cycle Hypothesis", Economic Inquiry.

Small investors do even worse than the pros, who do worse than the market

Researchers followed all 42,211 individual investors on the Colombian stock exchange from 2006 to 2016, over 5.3 million trades. As a group, households underperformed the market by 4 to 4.4% a year, risk adjusted and before costs, and the most active did worst even before commissions.

The figures are Colombian, not American, and the study is observational, so it shows association and not proven cause. But the pattern matches earlier research in the US and Taiwan: retail investors as a group do not beat the market, and those who trade most do worst. More trading mostly means more behavioral risk, not more skill.

Source: Garay, U., "The performance of retail investors, trading intensity and time in the market: evidence from an emerging stock market", Heliyon, 2021 (Garay and Pulga).

Two systems in the brain: why we make poor saving decisions

System 1 and System 2 explain why money often feels harder than the math says. Kahneman's research shows a loss feels about 2 to 2.5 times as heavy as an equal gain feels good, and that the default often decides: 97.4% were enrolled in organ donation where it was the default, against 18% where it was not.

System 1 is fast, automatic and emotional, and is practically always on. System 2 is slow, effortful and thoughtful, and in theory should control the fast one. The practical point is to build a savings plan where the right decision is the default: automatic, broad, cheap and hard to tinker with. A 401(k) with automatic enrollment and automatic escalation does this.

Source: Kahneman, D. (2003). "Maps of bounded rationality: Psychology for behavioral economics." American Economic Review.

Institutions are as bad as small investors

Imagine access to analysts, consultants and managers with the finest degrees. Would you manage to move your money into the right funds at the right time? According to the study, no. US pension funds, foundations and endowments systematically moved money into products that then did worse than the products they left. The gap averaged about 1 percentage point a year across about 80,000 observations from 1984 to 2007. Most of the underperformance came from manager selection, not from the split between asset classes, and the researchers priced the decisions at about 170 billion dollars in lost value.

The conclusion matches ours: do not chase yesterday's winners, stay in a broad, cheap and global strategy.

Source: Stewart, S. et al., 2009. "Absence of value: An analysis of investment allocation decisions by institutional plan sponsors."

Those who trade most lose most

In a study of 66,465 US households with brokerage accounts over six years, 1991 to 1997, and more than 3 million trades, the fifth that traded the most earned 11.4% a year net, while the fifth that traded the least earned 18.5%. Gross returns were nearly identical, so the 7.1 percentage point gap was due to the trading itself, not to worse stock picks.

The lesson is not that stocks are wrong, but that frequent switching often costs more than it adds.

Source: Barber, B.M. and Odean, T., "Trading is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors" (2000).

Despite warnings, most people pick funds on past returns

In three experiments with 1,600 participants, people chose essentially the same funds whether or not the warning text was shown: 86.4% picked the cheap fund without the text, 83.5% with the standard warning. Nearly half (46%) followed the rule of thumb of picking the fund that had done best last time. A text pointing to low fees as the yardstick lifted the share of correct choices to 89 to 91%.

Participants chose between a cheap and an expensive fund 60 times in a row, with feedback after each choice. The funds held the same assets and had the same expected return before fees, so the cheap fund was always the rational choice. The standard line "past performance is no guarantee of future results" changed almost nothing.

Source: Weiss-Cohen, L., Newall, P. and Ayton, P., "Persistence is futile: Chasing of past performance in repeated investment choices."

Markets and risk

The market is more concentrated than in 27 years, and it does not matter

Many savers worry that a global index fund is about 70% US stocks, and that a large share of that sits in a handful of mega cap tech companies. The researchers measured concentration in the S&P 500 with an "effective number of stocks" and found it at its most concentrated in roughly 30 years, but not unusual over 100 years of history.

They then tested whether cutting stock exposure when concentration rises would help, using data from 1926 to 2025. Both portfolios averaged about 70% in stocks. The dynamic portfolio returned 4.7% a year, the buy and hold portfolio 5.6%. The dynamic one also carried more risk: a standard deviation of 12.1% against 10.7%. Lower return and higher risk, the worst of both worlds.

In a further comparison, the quintile of the very largest companies, averaging 8 stocks, had about the same risk as the quintile with the smallest companies, averaging 328. The authors give three reasons, including that large companies are diversified businesses, not single bets.

For a US saver, concentration is not a reason to abandon a total market fund.

Source: Kritzman, M. and Turkington, D. (2025) The Fallacy of Concentration. Cambridge, MA: Windham Capital Management and State Street Associates.

Gold: a deep dive

Gold works mainly as insurance, not as the engine of a portfolio. Over a very long time it has roughly preserved purchasing power, but its long term real return is close to 0%. The stronger argument is crisis protection, not inflation: gold protected in 8 of 11 major stock market declines between 1975 and 2025, but it swings too much to be a reliable inflation hedge over normal saving horizons.

The paper is a working paper on SSRN, not yet peer reviewed, though the authors have studied this question for over ten years. For a US saver, a small gold allocation can be seen as a hedge, not a growth asset. Check how gold funds are taxed in your account type US review.

Source: Erb, C.B. and Harvey, C.R. (2025) "Understanding Gold". Available at SSRN: https://papers.ssrn.com/ (working paper).

Why index funds often beat actively managed funds

Risk in a single stock is largely company specific, and for that risk the market gives you no extra expected return. William Sharpe showed this in 1964 in the theory that became CAPM and earned him the 1990 Nobel Prize in economics, shared with Harry Markowitz and Merton Miller.

You can get rid of the unpaid risk almost for free by owning the whole market broadly and cheaply instead of a few companies. That is the root reason a broad index fund often beats active management over time. The paper is pure theory, a mathematical derivation with no empirical test, and later research has found where it falls short.

Source: Sharpe, W.F. (1964) "Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk".

A realistic long term return: why 3.7% may be more realistic

Average return and sustainable withdrawal are two different numbers. Using 71,156 stocks in 43 countries, the study finds a global stock portfolio historically supported about 3.7% a year in real, sustainable withdrawals over 1990 to 2022, the amount that could be taken out without eating into capital. The number describes history, before taxes and fees, and is no promise about the future.

The order of returns also matters a great deal: weak market years early in the withdrawal phase can lower the sustainable withdrawal sharply. The study, by Bessembinder and three colleagues, compares seven ways to measure long term return and covers more than 9.3 million monthly returns from January 1990 to December 2022. It rhymes with the 4% rule debate US review.

Source: Bessembinder, H. et al. (2025) "How Should Investors' Long-Term Returns be Measured?", Financial Analysts Journal.

The myth about index funds distorting the market

If everyone saves passively, who sets prices? Owen Lamont argues there is no point below 100% passive ownership where the market stops working, as long as enough knowledgeable active investors remain. An index fund owns but barely trades, so prices are set by the active investors buying and selling against each other.

Research is split. One study Lamont himself cites estimates passive flows lifted the 50 largest US companies 30 percentage points more than the market from 1996 to 2020, and the essay comes from an active management firm. For a saver the advice stands: a broad, cheap global index fund is still a very good default.

Source: Lamont, O. (2024) "Don't Blame Indexing for Your Problems". Acadian Asset Management.

Stock price moves cannot be told apart from chance

No mechanical trading rule beat a simple buy and hold strategy after commissions when Eugene Fama tested all 30 Dow Jones stocks in 1965. The link between one day's move and the next was negligible, while extreme days came far more often than a normal distribution predicts. The study used about 40,000 daily price quotes.

For a saver, a broad, cheap index fund left alone is a better choice than chasing patterns and switching funds often.

Source: Fama, E.F., 1965. "The behavior of stock-market prices." Journal of Business, 38(1), pp. 34 to 105. DOI: 10.1086/294743.

An 85% drop in a stock is common, and only half recover

The median US stock fell as much as 85.4% from peak to trough between 1985 and 2024, and about 54% of stocks never returned to their old peak. The S&P 500 fell 58% at worst in the same period and always recovered.

The report, based on more than 6,500 stocks, concludes that diversification is not theory but a way to survive the road to returns. The hard part of owning single stocks is not finding a good idea but holding on when it falls 50, 70 or 85%. A broad index fund spares you from carrying the full company risk yourself.

Source: Mauboussin, M. and Callahan, D., Morgan Stanley Counterpoint Global (report title not given in the source) US review.

Good results over three to five years are just noise

You do not need to wait 26 years for returns, but that is how long it took the US stock premium (1927 to 2010) to become statistically distinguishable from zero. A single fund's three or five year record is therefore nearly pure noise when the question is skill and not luck.

Evaluate cost, diversification and whether you can stick with the strategy rather than the latest results list. If even the whole market's risk premium needs decades to prove it exists, a short fund history says almost nothing about skill.

Source: Steiman, B., "Statistical Significance: An Essential Concept in Investment Analysis", Dimensional Fund Advisors, 2011.

Portfolio and risk

100% stocks for life? The research says maybe

A simulation using 39 countries' returns from 1890 to 2023 found the best lifetime portfolio was almost entirely in stocks: roughly one third domestic and two thirds international, with essentially no bonds. A target date style fund needed 63% more saving for the same retirement outcome (16.27% against 10.00% of income), and the risk of running out of money was lower with the stock strategy, 6.7% against 19.7%.

But the stock strategy had an average maximum drawdown of 55% during the working years, and it works only if you sit still through the falls and the rises. The point is global diversification and a long horizon, not maximum risk. The researchers simulated 1,000,000 life paths on actual returns from 39 developed countries.

In the US, this challenges the two most common pieces of retirement advice, to lower stocks with age and to always mix in bonds, which are built into target date funds in many 401(k) plans US review.

Source: Anarkulova, A., Cederburg, S. and O'Doherty, M., "Beyond the Status Quo: A Critical Assessment of Lifecycle Investment Advice".

Lifetime financial planning: the risk sits in the whole picture

The biggest asset most people own is not their savings but the present value of their future pay, which behaves more like a bond than a stock. In the authors' worked example, 35% stocks on total net worth equals about 96% stocks on financial savings at age 25, sliding to about 41% at 45 and about 24% at 65.

The book is a 246 page synthesis of how financial planning can hang together over a whole life, with an American example. The numbers are model results, not a general recommendation. The takeaway is to set risk on the whole balance sheet, including your job, your home and Social Security, not just the brokerage account US review.

Source: Idzorek, T.M. and Kaplan, P.D. (2024) "Lifetime financial advice: A personalized optimal multilevel approach." CFA Institute Research Foundation.

A portfolio of individual stocks gives worse returns

The market's return is driven by a few extreme winners, and with few stocks in the portfolio you are likely to miss exactly those. In Vanguard's simulation of 10,000 portfolios per size, the chance of beating the index rose from 11.1% with one stock to 48.4% with 500 stocks, and a random 30 stock portfolio averaged 0.4 percentage points a year below the index.

Ten hand picked favorites feel like backing your best ideas. The simulation drew equal weighted portfolios from 1 to 500 stocks out of the US Russell 3000 index for 1987 to 2017. Individual stocks can be fun, but they belong in a small sandbox. The base should be a broad, cheap, global index fund.

Source: Tidmore, C., Kinniry Jr., F.M., Renzi-Ricci, G. and Cilla, E. (2019) "How to increase the odds of owning the few stocks that drive returns." Vanguard Research, February 2019.

Trick yourself into a better portfolio

A study co authored by Nobel laureate Harry Markowitz shows something reassuring: splitting your savings into buckets with their own goals, instead of treating your whole finances as one lump, costs almost nothing in return. In the study's example it is about 0.12 percentage points a year. Guessing wrong about how much risk you can tolerate hurts considerably more, between 0.05 and 0.40 percentage points.

That supports thinking in horizons, as the Money Path does. Note that the study is worked out in theory and does not measure what real savers actually do. Most of us cannot answer "how risk tolerant are you, 1 to 10?" for our whole finances, but we can answer simpler questions about each bucket.

Source: Das, S., Markowitz, H., Scheid, J. and Statman, M., 2010. "Portfolio optimization with mental accounts." Journal of Financial and Quantitative Analysis, 45(2), pp. 311 to 334. https://doi.org/10.1017/S0022109010000141.

Retirement and withdrawals

Retirees spend their lifetime income, not their savings

At 65, retirees withdraw only 2.1% a year from their savings, about half of what the 4% rule suggests. At the same time, around 80% of lifetime income is spent, meaning Social Security, pensions and annuity style payments. Mental accounting is the explanation: money that arrives as a monthly amount feels spendable, while money sitting in an account feels untouchable.

Turning more savings into lifetime income could raise spending by up to 80% during the active retirement years, according to the study.

For a US saver this is a reason to plan a spending rule early, and to consider where guaranteed income such as Social Security timing or an annuity fits. We also find dividend paying funds easier to spend from, since dividends feel like "found money," though that is a behavioral trick, not a better return US review.

Source: Blanchett, D. and Finke, M. (2025) "Retirees Spend Lifetime Income, Not Savings", Financial Planning Review.

How much can you withdraw from your portfolio in retirement?

For a new retiree with a 30 year horizon and a 90% chance the money lasts, Morningstar's base case gives a safe starting withdrawal rate of 3.9% a year. Flexible withdrawal methods can lift the starting rate to as much as 5.7%, but then your income moves with the market.

The figures are US, before taxes and fees, and come from Monte Carlo simulations with 1,000 outcomes per portfolio and Morningstar's own assumptions about returns and inflation. It is a forecast, not a fact. The rate rose from 3.7% in the previous report, partly because the calculation method changed.

For a US saver this is a benchmark, not a guarantee. Treat 4% as a reasonable starting point and be ready to adjust.

Source: Arnott, A.C., Benz, C., Kephart, J. and Guo, T. (2025) The State of Retirement Income: 2025. Morningstar. https://www.morningstar.com/business/insights/research/the-state-of-retirement-income.

Withdrawal strategies for early retirement (FIRE)

Early retirement puts pressure on the 4% rule, since the capital may have to last 40 to 50 years instead of the 30 the rule was built for. A flexible withdrawal was the strongest strategy in Vanguard's simulation, lifting the chance that the portfolio lasted from 56.3% to 90.3%. But flexibility has a price: in the worst outcome the annual withdrawal fell to 24,709 dollars instead of the planned 40,000.

This is a model study from a fund company, not an empirical study of real savers, and it rests on Vanguard's own forecasts. It used Monte Carlo simulations with 10,000 return paths per scenario to test the effect of horizon, fees, diversification and withdrawal rule.

Source: Vanguard, "Fuel for the F.I.R.E.: Updating the 4% rule for early retirees" (2021).

Better financial decisions can add thousands to retirement income

Morningstar's Gamma study suggests retirement planning can be worth more than fund choice: five decisions about withdrawals, the whole picture and taxes gave 22.6% higher safe retirement income in their model, where dynamic withdrawal was the largest factor (9.88%). They convert the total to a return equivalent of 1.59 percentage points a year.

The authors simulate an American couple, both 65, and compare against a base case with a rigid 4% withdrawal and 20% stocks. The numbers come from a Monte Carlo simulation with 50,000 return paths, not real retirees, and Morningstar sells tools built on this thinking. The practical point: review your withdrawal each year and count Social Security, workplace plans and savings as one whole US review.

Source: Blanchett, D. and Kaplan, P., Morningstar, "Alpha, Beta, and Now... Gamma" (2013).

The 4% rule: how much you can withdraw without it shrinking

Anyone who withdrew 4 percent of the portfolio in the first year, and then raised the amount with inflation, had the money last at least 33 years in every US scenario from 1926 to 1976 with 50 to 75 percent stocks. Turned around, you need about 25 times your annual spending in savings.

William Bengen tested every possible retirement start year against what actually happened in the markets, not against averages. The rule is calibrated to about a 30 year horizon, before taxes and fees. For a longer horizon, such as early retirement, 3 to 3.5 percent is a more cautious starting point.

Source: Bengen, W.P., 1994. "Determining withdrawal rates using historical data", Journal of Financial Planning, October 1994.

The famous Trinity study and the 4% rule

The Trinity study tested withdrawals of 3 to 12 percent of starting capital against US market history from 1926 to 1995. An inflation adjusted 4 percent withdrawal lasted at least 30 years in 95 percent of periods for a stock heavy portfolio, while a pure bond portfolio managed only 20 percent.

You take 4 percent in the first year and then raise the same dollar amount with inflation each year. You do not take 4 percent of the current value every year. With $1 million, that is $40,000 the first year, then $40,000 plus inflation. The figure is before taxes and fees and is a historical outcome, not a guarantee.

Source: Cooley, P., Hubbard, C. and Walz, D. (1998) "Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable".

Withdrawal strategies: from portfolio to money to live on

Dynamic withdrawal, where the year's withdrawal moves with the market within a ceiling and a floor, succeeded in 86% of cases in Vanguard's simulation over 35 years, against 55% for a fixed inflation adjusted withdrawal. The flexibility also made room to start at a 5.0% withdrawal rate instead of 4.3%.

The simulation used 10,000 runs per scenario, a 35 year horizon, a portfolio of 50% stocks and 50% bonds, and an initial withdrawal of 5%. These are probabilities calculated from a model, not facts. The takeaway: being willing to adjust spending in bad years is worth a lot.

Source: Vanguard, "From assets to income: A goals-based approach to retirement spending" (2020).

Advice and fees

Women get worse advice at the bank

Financial advice is not always neutral. With the same finances and risk profile, women more often received pricier investment advice, fewer fee discounts and recommendations of the bank's own funds, which raises costs over time. The explanation lies in how advisors judge a client's knowledge, and in the bank's incentive to sell more profitable products.

The study draws on more than 27,000 real advice meetings at a large German bank. The researchers also ran an experiment with people who had essentially identical profiles, differing only in gender. The result was clear: women got fee discounts less often, were pointed to in house funds more often, and within the same fund category more often got the most expensive option.

For you, small differences in advice can cost thousands over time. Ask how your advisor is paid, whether they are a fiduciary, and what every fund costs in total. The data is German, so the US picture needs checking US review.

Source: Bucher-Koenen, T., Hackethal, A., Koenen, J. and Laudenbach, C. (2025) "Gender Differences in Financial Advice", American Economic Review, 115(12), pp. 4218 to 4252. doi: 10.1257/aer.20211024.

Financial advice reduces stress and saves time

In Vanguard's survey of 12,443 investors, 86% of advised investors reported more peace of mind and 76% said advice saved them time, a median of about 2 hours a week. High financial stress was 14% among advised against 27% among self directed investors.

It is an industry survey with no control group, and advised investors are wealthier and older, so it shows association and not cause. A simple system with a broad, cheap global index fund and automatic monthly saving often captures much of the same calm without costly advice. If you do hire an advisor, ask how they are paid.

Source: Vanguard, "The emotional and time value of advice", 2025.

The value of a financial advisor is up to about 3% a year

The largest single item in Vanguard's review of advice is behavior: avoiding costly panic switches is worth about 1.5 percentage points, more than low fees and rebalancing combined. The total of about 3% arrives unevenly over time, and several components rest on US tax rules.

Treat the figure with a warning flag. Vanguard is a fund company that sells funds and services to the very advisors the framework addresses, so the number is a marketing white paper, not independent research. The practical point is low fees, a written plan and the discipline to keep it.

Source: Vanguard (2019). "Putting a value on your value: Quantifying Vanguard Advisor's Alpha." Vanguard Research.

The value of advice is outsourcing your financial worry

When 2,228 Australians ranked what they value in an advisor, understanding their own situation (48.5%) and trust (44.8%) came ahead of competitive returns (22.2%). Only 24.9% had ever sought a financial advisor for their wellbeing, while 77.3% had gone to a doctor.

The report is industry funded and based on self reported answers at a single point in time, so it shows experiences, not proven effects. The data is Australian, so US attitudes may differ US review. The sense of control can often be built more cheaply with an emergency fund, the Money Path and a simple global index portfolio.

Source: Fidelity, "The Value of Advice", November 2019 (2,228 Australians).

Meta analysis: the value of financial advice

Merrill Lynch's report is built on twelve external studies and does not subtract the cost of advice, which is often 1 to 2.5 percentage points a year, so the net can end up small, zero or negative. The largest item is behavioral coaching, 150 to 400 basis points, which measures how much savers hurt themselves more than what the advisor adds.

"Advice gives 2 to 3 percent extra a year" is one of the most widely repeated sales figures in the industry, and it is before fees. The report is a compilation, produced by a firm that sells personal advice, in response to the rise of cheap robo advisors. A large part of the gain can be captured free with a broad index fund and a written plan, so always ask for the net figure after fees.

Source: Merrill Lynch, "The Value of Personal Financial Advice", 2017.

Advice gives 2% higher annual return

Morningstar has calculated that a skilled advisor can raise returns by about 2% a year for an average saver. It is a theoretical calculation, not extra return measured on real accounts, and the figure applies before the advisor's own fee is deducted. Most of the value is not about finding better funds but about avoiding your own mistakes: staying in the market instead of moving in and out at the wrong times, and having a risk level that fits.

For someone who already saves cheaply and broadly in index funds and leaves them alone, the value shrinks to about 0.4% a year, also before fees. The study splits portfolio building into seven decisions.

Source: Blanchett, D. and Kaplan, P., Morningstar, "The Value of a Gamma-Efficient Portfolio", 2017.

Saving and planning

Savings rate beats fund choice in Vanguard's life cycle model

In Vanguard's model, raising the savings rate from low to medium increased the probability of meeting retirement goals by more than 8 percentage points, while fine tuning fund choice and asset allocation had a clearly smaller effect.

This is a US model outcome from a company authored white paper, not independent peer reviewed research. But the direction is clear and matches what we have said for years: start with savings level, risk and time before you optimize the portfolio. The model builds a personal glide path for stock and bond shares through life, based on goals, savings rate, retirement age and risk preference.

Source: Aliaga-Diaz, R.A., Rayfield, Z., Daga, A. and Dinerstein, M. (2025) "Vanguard's Life-Cycle Investing Model (VLCM): A general portfolio framework for goals-based investing". Vanguard Research. Valley Forge, PA: The Vanguard Group.

Rebalancing once a year is enough

In Vanguard's simulation of portfolios with 35 to 90 percent stocks, annual rebalancing was the most effective and daily rebalancing the worst. The added value over monthly or quarterly strategies is small, 6 to 28 basis points (0.06 to 0.28 percentage points), so the point is that you lose nothing by being calm. The advantage is greatest when markets are at their shakiest.

If you invest through a robo advisor or an all in one fund, rebalancing is already automatic. In a taxable account, remember that selling to rebalance can trigger tax, so use new contributions first US review.

Source: Zhang, Y., Ahluwalia, H., Ying, A., Rabinovich, M. and Geysen, A., 2022. "Rational rebalancing: an analytical approach to multiasset portfolio rebalancing decisions and insights." Vanguard Research, October. The Vanguard Group, Valley Forge, PA.

An emergency fund makes you feel better

An emergency fund is not just money in an account, it is a mental cushion. In Vanguard's survey of 12,443 savers in summer 2024, those with at least about 2,000 dollars in savings rated their financial wellbeing 21% higher, spent nearly half as much time on their finances (3.7 against 7.3 hours a week) and reported rising stress three times less often. The links held after controlling for income, debt and wealth.

It is a survey, so it shows an association, not cause and effect. Still, it fits what we see. Start with a sufficient cushion rather than waiting for a perfect one.

Source: Vanguard, "The relationship between emergency savings, financial well-being, and financial stress" (2025).

Money and happiness

Millionaires: richest in the room, most worried day to day

In a survey of 200 British millionaires, respondents rated their daily worry at 4.10 out of 10, against about 3.23 for the average Briton, even though they saw themselves as less neurotic than most. The top 22% by income were both the most satisfied with life and the most worried. Relationships beat money on the group's own satisfaction list (7.40 against 6.87).

The report was commissioned by a wealth company that sells financial planning to this very group, has no peer review, and rests on 200 self reported answers. Read it as food for thought, not proof.

The question it raises is a fair one for any saver climbing the ladder: what is the money for? The sample is British with at least 1 million pounds of investable assets, so it may not map neatly to US households US review.

Source: Y TREE (2026), "The Deliberate Edge", research led by Paul Dolan, London School of Economics.

If money does not make you happy, you may be spending it wrong

The link between money and happiness is weaker than many think, but that does not mean money does not matter. How you spend often matters more than how much you have. In a US survey of over 1,000 people, 57% said they were happiest with their experience purchase, against 34% with their material purchase. Extra dollars tend to do the most good on experiences, small recurring pleasures, anticipation and meaningful generosity.

An important caveat: some findings, especially spending on others, depend on context more than the 2011 article implied.

Source: Dunn, E.W., Gilbert, D.T. and Wilson, T.D. (2011) "If money doesn't make you happy, then you probably aren't spending it right," Journal of Consumer Psychology, 21(2), pp. 115 to 125.

More money, more happiness

Yes, more money goes with more happiness. For most people wellbeing keeps rising with income, long past the famous 75,000 dollar mark, which was never an actual breaking point. When the researchers behind the earlier conflicting results analyzed the data together, "happiness keeps rising" got stronger support than "happiness levels off."

Two caveats. The link is real but fairly weak (a correlation of 0.09). And for the 15 to 20 percent who feel worst, the increase levels off at around 100,000 dollars of annual income. Our practical reading: financial security first. After that, money seems mainly to amplify the life you already have, not to create happiness on its own.

Source: Killingsworth, M., Kahneman, D. and Mellers, B. (2023) "Income and emotional well-being: A conflict resolved", PNAS.

Die With Zero by Bill Perkins

The book rests on one concrete finding. A study the author cites shows that retirees with at least 500,000 dollars in median savings had used only 11.8% of their capital 20 years after retiring. So the book is not about wasting everything, but about using the money while health, time and relationships remain.

Perkins himself calls exactly zero an impossible target. The idea of a "memory dividend" captures the point: experiences keep giving value for decades, while money in an account can become unused life. What if the biggest risk is not running out of money, but running out of life?

Source: Perkins, B., "Die with Zero".

Skipped items

Studies that rest on Swedish data, Swedish rules or the Swedish site itself, with no US equivalent in the source. Listed by Swedish source id.

• 10899: Swedish Financial Supervisory Authority report on the Swedish fund market: Swedish data and regulator, no US equivalent study in the source.
• 5079: Swedish premium pension (AP7 Saefa default vs own fund picks): Swedish public pension system data, no US equivalent.
• 5119: Retirement planning for the self employed in Sweden: Swedish pension thresholds and rules and a Swedish agency survey, no US equivalent study.
• 4759: Swedish equity funds 1999-2009 (Flam and Vestman): Swedish funds and Swedish fee data only; the US evidence is covered by the other active versus passive studies.
• 5030: How much Swedes hold in funds and shares (Swedish regulator survey): Swedish savings statistics, no US equivalent.
• 5048: Swedish Pensions Agency rules of thumb (23% automatic pension saving, Swedish replacement rates): built on the Swedish pension system.
• 5127: Swedish own savings versus public and occupational pension (Swedish Pensions Agency survey): Swedish pension system data.
• 5123: Whether pricier funds pay off on the Swedish premium pension fund platform: Swedish public pension platform data.
• 5125: Low fee benchmark for global index funds (Swedish Pensions Agency, Swedish market): Swedish fund fees, no US equivalent in the source.
• 5374: Swedish central bank memo on amortizing versus investing: built on Swedish mortgage rules, Swedish taxes and 1997 to 2012 Swedish data.
• 6330: Book by the Swedish site's own hosts on child saving: names real people from the Swedish site and rests on 2016 Swedish tax and brokerage rules.
• 8801: Swedish IPOs 1981 to 2006 (Lund master's thesis): Swedish exchange data only.