Diversification is about better returns not lower risk
– David Wanis, November 2019
All investors have heard this one before; Holding a portfolio reduces single stock volatility and that the volatility diversification benefit tends to plateau after about 30 stocks. By adding more names to a portfolio, you are simply diversifying away from your best ideas and reducing portfolio performance.
“Concentrated portfolios containing a relatively small number of stocks are a better idea and more likely to outperform for one simple reason – they are more likely to contain a manager’s best ideas. Adding more stocks to the mix wouldl only water down exposure these ideas, so investors should opt for concentrated portfolios, trusting their money to managers who have the courage of their convictions. That’s the theory anyway. Does the evidence support it?
– Proinsias O’Mahony from the Irish Times expresses the sentiment and language most investors feel about the logic of concentration.
Diversification blunting performance plays to a common narrative and how focused effort works in most endeavours. It feels like this is how it should work in markets, but does the evidence support it?
Assume you are asked to play a game where a dice is rolled and bets are placed on the outcome of each roll. Did you just imagine a six-sided dice? Did you assume it was unloaded – meaning every side has equal odds of landing? Was it, like the dice you used when you played Monopoly, numbered in a linear fashion?
This may be how dice games are played, but not how the market works. We need to be careful trying to fit games to a reality where they don’t apply.
If we consider individual stock returns, the stock market is more like a 20-sided dice, loaded so that each side does not have the same equal probability of appearing, and the numbers on each side are non-linear. Do your assumptions based upon the ‘normal’ dice still hold? They probably shouldn’t.
We mentioned last month the positive skew in the market which results in median individual stock returns being less than average (index) returns. This skew is like our loaded 20-sided dice and creates unusual outcomes. Outcomes that make assumptions about how returns, skill and concentration work misleading.
Recent studies have looked at the implications of positive skew in equity markets. They discovered, through empirical observation and simulation, that all other things being equal a more concentrated portfolio has a lower chance of outperforming the market than one with a greater number of holdings. The simple reason; index performance is heavily influenced by a very small number of stocks that deliver very high returns. Missing these few very big winners is very hard to make up for elsewhere in a portfolio. Simply having a larger portfolio increases the odds of holding one of these superstars.
Positive skew means that although volatility may indeed be low enough at 30 holdings, returns are more likely to benefit from a much larger portfolio – 100 holdings or more. This begs the question – what are you trying to achieve in your equity investment program? Are you trying to minimise stock specific volatility or are you trying to maximise your long term compound returns? We have yet to see an investment performance report which shows volatility vs benchmark over 1, 3 and 5 years with performance buried in the footnotes on page 2.
Evidence of performance benefits in small cap diversification
We often focus on why small caps are different, and the difference in skew is another reason why a small cap investment strategy may differ to large caps. As we outlined last month, the positive skew in small caps is more pronounced than large caps, which would imply a more diversified approach should do better over time.
Morningstar in the U.S investigated this and their conclusion last year was as follows;
“The data is clear: the more concentrated a portfolio is, the greater the risk of missing out on the market’s biggest winners and underperforming. This risk is greater among mid- and small-cap stocks than it is among large ones. It is hard to identify the market’s big winners ahead of time, and more difficult still for a concentrated manager to ride those stocks all the way up because doing so would eliminate any semblance of diversification. It is also difficult to hold on to winners because their valuations likely become stretched along the way, which may tempt managers to sell to lock in the gains and invest in a more attractively valued alternative. Concentrated active managers who do catch some of the big winners can crush the market, but the odds are against them.”
– Morningstar Research, November 2018.
The most compelling statement for most people will be “concentrated active managers can crush the market”. The bit about “the odds are against them” doesn’t trigger the same emotional response.
The performance data Morningstar refers to showed that from their manager database, the most concentrated quintile of large cap blend managers underperformed the most diversified quintile by 76bps p.a between 2008 and 2017.
For mid blend and small blend, the concentration drag on the same basis was 226bps p.a and 149bps p.a respectively. Taking home an extra 1.5% to 2% per annum in small caps is worth paying attention to, even if “diversification and positive skew” is not as sexy as “crush the market”!
Skill and diversification are not mutually exclusive. It is just as feasible to apply skill in security selection, position sizing and portfolio construction to a diversified portfolio as it is to a concentrated portfolio.