The quest for decorrelation between different assets has been disrupted by the economic environment. Despite the recent rise in rates, alternatives to bonds and stocks exist.
Portfolio theory developed by Harry Markowitz in the early 1950s seems more relevant today than ever. The notion of diversification provides, in fact, particular insight to investors, whose portfolios are mainly made up of stocks and bonds, in a world where the correlation between these two asset classes seems to have (re)become positive .
Historically, it was difficult to do better than the benchmark portfolio, called 60/40 (60% in stocks, 40% in bonds), and those of us who have experienced anything other than an environment extremely favorable to these two asset classes – in fact, favorable to all major traditional asset classes – are, at best, close to retirement. Indeed, the last 40 years have seen an almost uninterrupted fall in rates (see graph 1). This trend has obviously been profitable for bonds, whose prices are very directly linked to the evolution of rates, but it has also been beneficial for all traditional asset classes, like share prices, which were therefore carried ever higher by ever lower rates.
Evolution of rates
The last 40 years have been marked by an almost uninterrupted decline in rates, which has benefited traditional asset classes.
Source: BCV
And that’s not all. For more than two decades, the correlation between stocks and bonds has been mostly negative, and often very significantly so (see chart 2). A portfolio made up of stocks and bonds therefore benefited from the best of both worlds: high returns and significant diversification potential. Because even though stocks represent almost all of the risk in the 60/40 portfolio, this negative correlation has generally allowed bonds to play a welcome diversifying role during downturns.
Correlation between stocks and bonds
Since the early 2000s, the correlation between stocks and bonds has been mostly negative, and often very significantly so.
Source: BCV
Limited diversifying power
But now, despite their recent rise, rates now find themselves at historically low levels. Worse, their correlation with stocks, which we have been accustomed to being negative for more than twenty years, is currently very largely positive, effectively limiting the diversifying power of this asset class. And this is anything but trivial.
The role played by this correlation between bonds and stocks in the process of constructing a well-diversified portfolio is, in fact, much more fundamental than it seems. In fact, financial theory teaches us that this is almost the only truly relevant factor.
So, what would Markowitz tell us if we presented him with the following simplified example? Let us assume that the expected risks and returns of stocks and bonds remain, over the next decade, the same as those observed in previous decades, and that only the correlation between these two asset classes is now different.
Concretely, the expected returns – in excess of the risk-free rate – are set at 7% for stocks and 2% for bonds and their respective volatility at 15% and 5%, i.e. levels representative of their historical values. To construct the two efficient frontiers (see graph 3), only the correlation was therefore modified, to first reflect the level observed over the last two decades (around –0.40), then secondly that observed more recently, around +0.40.
Markowitz’s efficient frontiers
The consequences of increasing the correlation are even more problematic for conservative profiles.
Source: BCV
Risk or return
These two efficient frontiers reveal the problem currently facing an investor whose portfolio is 60/40: maintaining the same level of risk requires accepting an expected return that is 14% lower (4.3% versus 5.0%). . On the other hand, maintaining the same level of expected return involves taking a 19% higher risk (10.0% versus 8.4%). This alteration in the risk-return ratio is significant and results only from the increase in the correlation between stocks and bonds, their respective returns and risks having been kept constant.
The result of the analysis (see graph 3) also suggests that this now positive correlation could prove even more problematic for more conservative investors, i.e. those located lower on the efficient frontier. The gap between the two curves indeed seems more significant, and the visual impression is correct. As an illustration, a pension fund which holds a more conservative portfolio, say 40/60, for example due to the high average age of its policyholders, would see its return reduced by 32% (2.7% compared to 4 .0%) for the same level of risk, or its risk would increase by 39% (7.7% versus 5.5%) if it kept its 40/60 allocation unchanged with the aim of targeting an identical return.
Correlation
Taking more risk to maintain returns at their historical levels is a phenomenon that has been with us for several years now, with many investors having reallocated part of their bond pocket towards a little credit, or private assets in particular. However, this approach in fact only amounts to adding even more “traditional” risks – and therefore positively correlated to stocks, because they are all sensitive to the same vagaries of the economic cycle – to portfolios which are already full of them. This approach actually only addresses a tiny part of the problem: remuneration, considered insufficient, for government obligations. On the other hand, it completely obscures the substantial gains in terms of diversification – and therefore return/risk – which could be provided by assets or investment strategies with little correlation to traditional assets.
Alternative solutions
And that’s precisely what some alternative investment strategies offer. Only some, because not all have the same diversifying potential. So-called Equity Long-Short strategies, for example, generally maintain significant exposure to equity markets, and are, therefore, highly correlated to them. In turn, the indices representative of the hedge fund industry – around half of which are made up of this type of fund – are correlated with the stock markets, which seems to make hedge funds ineffective in terms of diversifying traditional portfolios.
Diversity of hedge funds
However, it is not the case. The hedge fund industry is very heterogeneous and certain strategies are able to provide the desired decorrelation, without lowering return expectations. A puzzling example of this are Trend Following strategies, which aim to follow trends within all asset classes. Although they tend to generate lower returns than stocks during periods of strong stock rises, they have always significantly outperformed the latter during the main crises in history. Isn’t that almost the very definition of the term diversification?
