Just how much can you diversify risk?

It’s no great secret that diversifying can reduce risk, but to just what extent? Portfolio theory developed by Harry Markowitz in 1952 suggests that holding a wide range of investments can help reduce the systematic risk. For this theory to work the rates of return, risks/standard deviation and relationships between the investments are considered. Various complex calculations and modelling are done to provide what is known as the “efficient frontier” of different assets that are allocated to provide optimum returns. However, in reality it’s reported that not even the experts are following through with their own advice!

“I should have computed co-variance of the asset classes and drawn an efficient frontier – Instead I split my contributions 50/50 between bonds and equities.” (Professor Markowitz)

So computer modelling aside what are the experts suggesting? It’s thought the vast majority of UK-based academics maintain a 60/40 split between equities and bonds, with just a couple of funds of each. Though as you can see below diversification isn’t just confined to the domestic market, diversifying internationally is also thought to provide benefits to your portfolio.





The y-axis here shows the risk as a percentage of the risk on a single share. Interestingly, holding international stocks can reduce that level of risk from about 33% down to about 15%, the problem I would have with this is that currency fluctuations can be volatile and so this does add to the risk. Also, it’s worth noting that any increase on 20 stocks doesn’t mitigate the risk further. In fact, it was said by a US analyst that anything over 10 reduces the ability to add value; however, I think that depends entirely on your risk appetite.

When deciding on a portfolio, one thing analysts can agree on is that the covariance between the companies should be low, for example if you invest in five airlines and there’s a major incident or oil prices spike then it’s likely the companies will all react in a similar way. Therefore, it is better to diversify across many different industries so that bad news for one could be offset by good news in another.

 

As you can see in the graph above, two perfectly negative correlating stocks can eliminate volatility and provide a healthy return. In practice it wouldn’t be as easy as this but do you really need to be trying to decrease volatility? It has long being thought that volatility is the investor’s worst enemy and investing on those stocks which just slowly but surely increase at a slow pace is the way to go. Nonetheless this changed in the aftermath of the financial crisis, these stocks that been labelled as placid and dormant were suddenly revealed to be extremely toxic and assets with low standard deviations lost all of their value. It’s these toxic securities that either do quite well or when the market turns will totally destruct, which then of course it is too late. It’s all too easy using the standard deviation method to describe these stocks as non-problematic and a “safe bet”.

An important part of a diversified portfolio is rebalancing, although you may start your portfolio with a certain percentage of domestic stocks, international stocks, bonds and cash, the mix will change over time.

This chart's hypothetical illustration uses historical monthly performance from April 1994 through April 2014 from Morningstar/Ibbotson Associates; stocks are represented by the S&P 500 and MSCI EAFE Indexes, bonds are represented by Barclays U.S. Intermediate Government Treasury Bond Index, and short-term investments are represented by U.S. 30-day T-bills. Chart is for illustrative purposes only and is not indicative of any investment.

Looking at the charts above, you can see that the breakdown of each investment has changed dramatically in those twenty years, with now over 80% being invested in shares. This means that stocks have had larger swings than bonds/cash and therefore going on historical data it is now subject to more risk because of this. Let's consider if the investments had been rebalanced semi-annually to their original levels and whether that lowers the risk at all.

 

Chart is for illustrative purposes only. The chart depicts observed historical risk based on the performance of broad diversified indexes named in the chart “How an investment mix can change over time.” A portfolio that is not diversified within asset classes may experience different levels of risk.

As you can see it levels out the risk and so if you were wanting a more cautious investment then it is definitely worth rebalancing to the original structure.

Personally I believe that market volatility can help investments, giving investors access to "cheap" stocks and the potential for large gains. However, it is entirely dependant on the individuals circumstances, if you are retired and hoping to live off the income from an investment then naturally your appetite for risk will be fairly low. On the other hand as a young professional who has other sources of income, taking a risk on a volatile stock might be worth it for a larger pay off. Dividends vs. capital gains is something I will be discussing in my next blog and assessing just how important each of these are.

I don't think it's necessary to work through complicated formulaes to try and achieve the perfect portfolio but it is certain that diversifiation can help reduce fluctuations. For a successful portfolio it would seem around 10/15 stocks across different industries is suffice, preferably with some negative correlation and an effort to rebalance at least annually.