Ron Bewley says the heart of a portfolio determines volatility.
About a year ago, I had an article published in this magazine about how many stocks an investor needs in a portfolio to reduce volatility to a manageable level. If the stocks in the portfolio have equal weights, I concluded that eight to 15 ASX 200 stocks are sufficient in normal volatility times to diversify away much of the risk. And this number might be increased to 13 to 15 stocks in more volatile times.
Investors who are new to portfolio construction might still find it hard to find sufficient (hopefully) good stocks to get going. With the increase of interest in exchange-traded funds (ETFs), some investors are looking towards investing in the index, rather than individual stocks. But what about a compromise? Say, a few stocks and an ETF?
In this article I revisit the number of stocks question, assuming one of those stocks happens to be the index – and not necessarily equally weighted with the others. I used data from 2010-11 to construct simulations of a number of equally weighted portfolios of top 200 stocks, plus a core holding in the index of between zero and 80 per cent – in 20 per cent increments. That is, an 80 per cent core and five stocks has 80 per cent invested in the index and the additional five stocks get 4 per cent each – making 100 per cent in total.
‘I concluded that eight to 15 ASX 200 stocks are sufficient in normal volatility times to diversify away much of the risk’
Rather than just estimating the average volatility of each hypothetical portfolio, I computed the median volatil- ity and the 99th percentile – which is the maximum volatility from the one million simulations after the worst 1 per cent had been eliminated (see Chart 1). The solid purple line (0 per cent core) represents the 99th percentile of the portfolios with no index included. The dotted purple line represents the median volatilities from the same zero-per-cent-core portfolios. Both purple lines are flattening out by the time 10 stocks have been included in the portfolio.
The yellow line represents the 99th percentile of portfolios that include 20 per cent invested in the index. So, in the one-stock portfolio, there is 20 per cent invested in the index and 80 per cent in the single stock. In the two-stock portfolio, there is 20 per cent invested in the index and 40 per cent in each of the two stocks, et cetera.
There is no dramatic improvement by including a 20 per cent core. However, as the size of the core increases, important gains are made. For a 40 per cent core, the 99th percentile (green line) gets close to the median line for portfolios with no core by the time five or six stocks accompany the 40 per cent core. When the core is 60 per cent, the red line for the 99th percentile is lower than the median zero per cent-core line for two or more stocks. An 80 per cent core brings the 99th percentile below the median zero per cent core for all numbers of stocks. In other words, an almost worst-case scenario for an 80 per cent core is better than a median zero per cent-core portfolio.
The conclusions are straightforward. An investor who can tolerate the typical volatility of an equity market need only find two or three stocks to combine with a core of 60 per cent or 80 per cent in the index. While this combination reduces volatility over zero per cent-core strategies, it does make it harder to get much outperformance or alpha. Next month, I will combine the results of my last three articles to suggest a strategy for a self-managed super fund.
Ron Bewley is executive director of Woodhall Investment Research.
