Without having read every article on portfolio construction, I’m relatively confident that many mention Harry Markowitz’s portfolio theory – and possibly even deride it for its weaknesses in today’s world.
But last week I found myself reminiscing of the days in the lecture hall at university where this theory was being explained, and began wondering whether the criticisms I’ve heard since were known then. I’m sure they were, but perhaps they were just not relevant at the time.
It comes to light again due to the articles at the front of this week’s issue: the idea that a portfolio could contain no allocation to fixed income is interesting.
At a basic level, one would assume a modern portfolio’s starting point should be 60:40 allocation, with adjustments made either side to account for today’s vast array of alternative asset classes and sentiment at the time.
Any bond manager you speak to in the current climate talks of the greater downside risk than upside potential in their asset classSo Old Mutual Wealth’s decision to remove fixed income almost entirely from its range of risk-adjusted model portfolios is a significant move. Let’s not underestimate this. It can of course cause issues for advisers in terms of compliance and costs, and it becomes difficult to explain to clients who may have preconceptions on what a portfolio should look like.
But, in today’s market, and using the basic points of Markowitz’s theory, it is perfectly understandable. Any bond manager you speak to in the current climate talks of the greater downside risk than upside potential in their asset class. “Yields surely can’t go any lower”, is a term I’ve heard more often than not. So, with that perceived certainty, why would anyone allocate?
The answer is of course, always, diversification. But does that even hold true? Year to date, the Barclays Global Aggregate index, a fixed income benchmark that includes corporate debt, has a 0.74 correlation with the MSCI World. Similarly, the Barclays Gilts index has a 0.34 correlation to the FTSE All Share – lower but still positive. Between September 2008 and the end of 2009, a time of great equity volatility for documented reasons, the same indices had correlations of 0.27 and -0.04, demonstrating at least some downside protection.
Harry Markowitz’s theory was not designed for the bizarre set of circumstances we find ourselves in. But at the basic level it told us not to include correlated assets meant to diversify if they no longer do so. If two asset classes are as correlated as demonstrated above, then surely they cannot be included in the same portfolio. You have take one of them out, decided by which one has the greatest downside risk in the short term.
So wealth managers and advisers reducing allocations to fixed income proves the theory holds some credibility.
Many are quick to slam the 1952 theory for its greater weaknesses in 2016. But no one predicted negative bond yields, negative interest rates and fixed income valuations at record highs while the FTSE 100 shifts above 7,000 again. The simplistic points Harry Markowitz made 64 years ago might cause a headache for advisers in conversation with clients, but it might just protect a greater downside.
Taha Lokhandwala is news editor of Investment Adviser
