Diversification: how many stocks do you actually need?

Ten stocks was the textbook answer in 1968. Later research pushed it to 30, then 50 — and the Motley Fool still recommends 50. What counts is how different the holdings are, and how much money sits in each one.

AI illustration for this news story: Diversification: how many stocks do you actually need?
AI-generated illustration, not a photo.

The old answer: ten

In 1968, two researchers asked a simple question: how many stocks does it take to build a diversified portfolio? John Evans and Stephen Archer, at the University of Washington, built portfolios of randomly chosen stocks, added names one at a time, and watched the volatility fall. Most of the benefit arrived in the first handful of stocks, and they put the minimum portfolio size at 10. The finding entered textbooks as a rule of thumb: ten stocks is enough. It was a neat answer, and it stayed in place for almost twenty years.

The problem was what it measured. Evans and Archer tracked only how much a portfolio's returns bounced around — its standard deviation. They did not ask what a ten-stock portfolio could still lose, or what it could miss. Morningstar data from June 2021 shows one side of that: about 39% of all U.S. stocks had ever suffered three-month losses of 50% or more, while fewer than 1% of diversified stock funds had incurred losses that severe.

The curve flattens fast — then barely moves

Another study from around the same time put numbers on the curve. The standard deviation of a portfolio of just one stock was 49.2%. With 10 stocks it fell to 23.9% — less than half. With 50 stocks it was 20.2%: the first 10 stocks removed most of the risk, and the next 40 removed only a little more. The shape is the lesson: the first stocks do nearly all the work, and each one after that adds less.

The same arithmetic shows up in a simpler rule. With a portfolio of 20 equally weighted stocks, even if one of them loses its entire value, the portfolio only drops by 5%. That is why many advisers suggest keeping no more than 3% to 5% of a portfolio in any single stock: a single failure should be a bad day, not a ruined year.

Standard deviation by portfolio sizeSource: Kiplinger
  • 1 stock49.2%
  • 10 stocks23.9%
  • 50 stocks20.2%

The number kept rising

Meir Statman was not satisfied with the ten-stock answer. In a much-cited 1987 paper, he used a different method — weighing the benefit of each added stock against the cost of holding it — and concluded that investors need 'no less than 30 stocks.' Another group of economists, led by Harvard's John Campbell, determined that you need 50. The Motley Fool's recommendation is also 50 stocks, though its own writers note that 50 is a lot of work — and that owning 50 technology stocks still leaves a portfolio exposed to a single sector.

Investopedia's verdict is blunter: there is no consensus answer, only a reasonable range. The research says 10 at the low end and 50 at the high end, and the spread is the point: the count that fits one market and one decade does not fit the next.

Dozens of small glass jars of different sizes and colours lined up on a long wooden shelf
AI-generated illustration, not a photo.

Count is not diversification

Even the right number can mislead, because the studies assume stocks picked at random, in equal weights, across the whole market. Real portfolios are not random. Kiplinger's test is simple: if all 30 of your stocks were in energy, your annual average return over the past 10 years would have been 5.6%, against an annualized 13.9% for the S&P 500 as a whole. Thirty stocks, one sector, one bet.

Weights matter as much as count. You are not diversified if you own 30 stocks with 29 of them each representing 1% of total assets and one of them representing 71% — the fix is to rebalance at the end of every year or six months. Enron is the extreme case: it was the seventh-largest company in the Fortune 500 when it collapsed in 2001, yet its market value was less than 1% of the S&P 500 — so $10,000 in an index fund lost just $100 when the stock went to zero.

A single broad fund sidesteps the counting question entirely: the Fidelity 500 Index holds about 500 companies, and the Vanguard Total Stock Market fund owns 4,119 separate stocks. Investopedia's shortcut is the same idea: $10,000 spread across a few ETFs can buy exposure to thousands of stocks. The question worth asking is not how many names are on the statement, but how many genuinely different bets they are, and how much money rides on each. Educational content, not investment advice.

Sources

Educational content, not investment advice. Facts as of the publish time above; figures come from the linked sources and markets change quickly. Genexus Alpha is not a registered investment adviser or broker.