Investing Basics: What the Research Says About Returns, Costs and Behaviour

This article summarises published research on investing. It is general information, not financial advice: it doesn't know your situation or your country's rules, and it doesn't recommend any product or fund. Past returns don't predict future ones.

VettedVids rates books and videos by what people report happened after using them. Research is different: we don't score it. We summarise what each study found, the market and period it covers and its limits, and where a finding or method is disputed we say so rather than settle it. Much of this research uses US data, and we say where. Every claim below links to its source, and the same findings are in the Research tab of our Start investing page.

How investing works: risk, return and diversification

• Shares have paid more than bonds and cash over very long periods, but not in every decade. In the US from 1900 to 2025, equities returned 6.6% a year after inflation, against 1.6% for bonds and 0.5% for Treasury bills (Dimson, Marsh & Staunton 2026, UBS Global Investment Returns Yearbook). That is US data only, and the US has been one of history's strongest markets; the yearbook covers 35 markets, and other countries' results vary.
• Most individual shares do worse than cash; a few drive the market's gains. Most US common stocks listed since 1926 had lifetime buy-and-hold returns below one-month Treasury bills, and the best-performing 4% of companies explain the entire net gain of the US stock market over that time (Bessembinder 2018, Journal of Financial Economics). The study covers US stocks from 1926 to 2016 and describes individual stocks, not diversified portfolios: the point is why diversification matters.
• The basics aren't widely known. In a 2004 US survey of people aged 50 and over, only about half answered simple questions on compound interest and inflation correctly, and only about a third got all three basic questions right. Similar gaps show up in surveys around the world (Lusardi & Mitchell 2014, Journal of Economic Literature). Three-question measures are rough proxies for financial knowledge.

Getting ready: making saving automatic

When a US company switched its 401(k) plan to automatic enrolment, participation rose significantly. Many new employees stuck with the default contribution rate and fund, even though few had chosen those before (Madrian & Shea 2001, Quarterly Journal of Economics). This was one large employer, and the switch took effect in April 1998. Defaults raise participation, but a low default rate can also anchor people to saving too little.

We haven't summarised research on emergency funds or paying off debt. The rated books and official guides on the Start investing page cover those steps.

Making your first investments

• A lump sum or spreading it out. Across historical and simulated data, investing a lump sum immediately did better than spreading it over several months about two-thirds of the time, and spreading it out did better than holding the money in cash about 69% of the time (Finlay & Zorn 2023, Vanguard Research). It was published by a fund manager and uses US, UK, Australian, Canadian and EU market data, mostly from 1976 to 2022. It's about a windfall, not regular investing from each paycheck.
• Home bias. Investors held more than 98% (Japan), 94% (US) and 82% (UK) of their equity portfolios in domestic stocks, a concentration that implies they expected their home market to outperform others by several percentage points a year (French & Poterba 1991, NBER Working Paper 3609). The holdings estimates date from the early 1990s, when the paper came out; home bias has fallen since then but hasn't gone away.

What research found on index funds compared with actively managed funds is in the costs section below.

Staying the course

• Trading more, earning less. Among 66,465 US brokerage households from 1991 to 1996, the most active traders earned 11.4% a year while the market returned 17.9%. The average household earned 16.4% and turned over 75% of its portfolio each year (Barber & Odean 2000, Journal of Finance). The data come from one US discount broker during a strong bull market and cover direct stock holdings only.
• Overconfidence. Using data on over 35,000 households at the same broker (February 1991 to January 1997), men traded 45% more than women. Trading cut men's net returns by 2.65 percentage points a year, against 1.72 points for women (Barber & Odean 2001, Quarterly Journal of Economics). Gender stands in for overconfidence here; it doesn't mean individuals are bound by type.
• Investors earn less than the investments they hold. Because money tends to flow in after rises and out after falls, investors' actual (dollar-weighted) returns have been systematically lower than buy-and-hold returns in nearly all major stock markets studied (Dichev 2007, American Economic Review). The study uses market-level data for the US and major international markets, so it measures aggregate flows, not individual investors. Hayley argues that for mainstream US stocks much of this gap reflects a hindsight effect in the dollar-weighted method rather than poor timing (Hayley 2014, Journal of Financial and Quantitative Analysis).
• The gap in funds today. Over the 10 years to end-2025, the average dollar in US funds earned 8.7% a year, while the funds themselves returned 9.9%: a gap of about 1.2 percentage points a year. Morningstar links gaps mainly to the timing of purchases and sales, but notes a gap doesn't necessarily reflect bad behaviour: steady regular investing can create one too. Gaps were narrowest in broad US stock and allocation funds (Morningstar 2026, Mind the Gap). This covers US funds only and comes from a commercial research firm. Methods differ: Morningstar uses fund-level dollar-weighted returns, while the widely quoted Dalbar study has been criticised for overstating the gap.

Costs: why fees matter

• The arithmetic. Before costs, the average actively managed dollar must earn the same as the average passively managed dollar, because together they make up the market. After costs, the average active dollar must therefore earn less (Sharpe 1991, Financial Analysts Journal). This is a logical identity about averages, not a claim that no active fund can outperform.
• What active investing cost. From 1980 to 2006, investors spent about 0.67% of the total US stock market's value each year on fees, expenses and trading costs in pursuit of above-market returns. The typical investor would have gained about 0.67 percentage points a year by holding a passive market portfolio instead (French 2008, Journal of Finance). Fees have fallen substantially since.
• Most active funds trail their benchmark over long periods. To mid-2026, 94% of equity funds sold in Europe (all categories) underperformed their benchmarks over 10 years (SPIVA Europe Mid-Year 2026), as did 89% of Australian general equity funds over 15 years (SPIVA Australia Mid-Year 2026) and 74% of Indian large-cap funds over 10 years (SPIVA India Mid-Year 2026). In the US, 79% of active large-cap funds trailed the S&P 500 in 2025 alone (SPIVA U.S. Year-End 2025). All four scorecards are from S&P Dow Jones Indices. Benchmarks and categories differ by region, and results vary from year to year.
• Past winners rarely keep winning. Among US equity mutual funds, in a sample free of survivor bias, common market factors and fund expenses almost fully explain why some funds appear to keep outperforming. The only strong persistence was among the worst funds, which kept underperforming (Carhart 1997, Journal of Finance). The paper concludes its results don't support the existence of skilled fund managers, a claim others still debate. More recently, only 4.5% of US large-cap funds that were above median in 2021 stayed above median every year through 2025, fewer than the 6.25% expected by chance (S&P Dow Jones Indices, U.S. Persistence Scorecard Year-End 2025).
• Fees predict fund success. Funds in the cheapest fee quintile were about three times as likely as the priciest quintile to survive and outperform their category peers. For US equity funds, the success rate was 62% for the cheapest and 20% for the priciest (Kinnel 2016, Morningstar). This uses US fund categories with data to end-2015, from a commercial research firm.

Keeping it going

• Committing future pay rises to saving. In 'Save More Tomorrow', employees committed in advance to put part of future pay rises into retirement saving. In the first implementation, 78% of those offered joined, and their average saving rate rose from 3.5% to 13.6% over 40 months (Thaler & Benartzi 2004, Journal of Political Economy). That was one US firm, the first of several implementations, and the authors designed the programme.
• How much a retirement portfolio can pay out. Using US data from 1926, a first-year withdrawal of 4% (then adjusted for inflation) never exhausted a stock/bond portfolio in under 33 years, the origin of the '4% rule' (Bengen 1994, Journal of Financial Planning (reprinted 2004)). Repeating the test across 17 developed countries, a 4% rate would only have been 'safe' over 30-year retirements in 4 of them (Pfau 2010, Journal of Financial Planning). The rule is contested and context-dependent: results depend on asset mix, fees, timeframe and country, and both are historical simulations, not forecasts.

How we put this together

We collected research that answers questions people face when they start investing: what returns have looked like, why costs matter and what tends to go wrong. The sources are peer-reviewed journals, index-provider scorecards and fund-industry research, and we note which is which. Each finding is listed with the market and period it covers, the type of evidence and its limits in the Research tab of the Start investing page, alongside rated books, rated videos, official guides from regulators in several countries, free calculators and adviser registers. Last checked 18 October 2026.