[[Start here]] → [[What works in stocks?|what works]] → [[diversify for resilience|diversity]] → five stocks --- Back in 1968 a couple of academics, Evans and Archer[^1], tried to solve the question of "*how many stocks are enough to diversify a portfolio*" and essentially came up with a range... **five to ten**. They laid out one of the first classic texts on this subject, showing that most of the gains to be had from diversification come _from adding just the very first few stocks to a portfolio_. According to their results adding just 4 more stocks to a 1 stock portfolio gives you 55% of the benefits of diversification (in terms of volatility reduction) compared to owning the whole market. The benefits tail off as you add more and more stocks. In other words, there are diminishing returns from increasing diversification. The chart below from their paper illustrates this point. It plots the average portfolio volatility against the number of stocks in the portfolio on the horizontal axis for six-month holding periods. ![[evans-archer-1.png]] An eye opening quote from the conclusion of the paper: > The results raise doubts concerning the economic justification of increasing portfolio sizes beyond 10 or so securities. It’s easy to jump to conclusions that you should only own 5-10 stocks from this paper, and many papers after it’s publication came to similar conclusions, but there are some real issues. Firstly, the paper only looked at the averages, and averages are extremely deceptive. Investors don’t own the “average portfolio of five stocks”. They own a unique portfolio of five stocks. *Unique portfolios have a huge dispersion of results.* Some will be far more volatile, some will be far less volatile. Investors at the extreme won’t be happy. Secondly, Evans and Archer’s research focused solely on how many stocks are needed to reduce _portfolio volatility_, but *completely ignored the risk that the resulting portfolio fails to show adequate performance*. Again, the huge variability of return outcomes in five stock portfolios won’t make a great proportion of investors happy. I recommend reading my next piece which explains how [[owning 20 stocks reduces the chance of significant underperformance]]. Nonetheless, the key insight from the paper remains. > [!info] Rule of Thumb > The average volatility of a portfolio is significantly reduced by owning the first five to ten stocks. [^1]: [[Evans & Archer - diversification and the reduction of dispersion]]