Is traditional asset allocation no longer an option?

Sometimes, I wonder if asset allocators realise how lucky they used to be.

We used to have the luxury of combining bonds with equities to form a diversified portfolio. Sovereign bonds – from many countries – used to be of high quality. But I am afraid that those times are now over. Why? Because quantitative easing has destroyed the very properties of fixed income that made the asset class an essential part of a balanced allocation. Factor investing may be the best solution we have.

For most European investors, domestic sovereign bonds form the basis of their asset allocation. These bonds used to provide an attractive yield as well as strong diversification benefits. But what do we actually mean by diversification? For me, it means they could be relied upon to protect an investor’s assets in times when it really mattered: good sovereign bonds normally rose in value when the equity markets were falling. What’s more, the coupons they provided would in many cases account for most of an investor’s expected returns, and sometimes their projected liabilities, on their own. This now seems like something of a dream.

What other asset classes can investors look to for diversification? Real estate is an obvious example, but its poor liquidity is a major drawback. Institutional investors tend to have mixed feelings about commodities, as they are attracted by their (sometimes) low correlations but put off by their high levels of volatility. Hedge funds, for their part, seemed to be the miracle cure around 2000 but – and this only applies to the better ones – they are at best interesting satellite solutions. Strangely, the fees they charge have been more of a concern to investors than their correlation to the equity markets.

The problem is quite simple. Investors want to be able to replicate the past behaviour of sovereign bonds, but in today’s world that’s just impossible. Sovereign bonds used to be of the highest quality, providing excellent diversification benefits as well as an attractive coupon.

Now, they are no longer safe (as we saw in the Greek debt crisis) and, without a yield, they may have lost their ability to provide diversification (unless we imagine a world of prolonged significant negative interest rates). Worse still, we can envisage scenarios in which rates move higher while the stock markets fall. And what will be the long-term rationale for investing in bonds that do not provide a yield? Counterparty diversification cannot justify the high probability of investors losing their money after the effects of inflation.

The natural thing for investors to do in recent years has been to seek diversification by allocating to asset classes that resemble sovereign bonds: these include convertible bonds, high yield, emerging market fixed income and private debt to name a few. But a major problem with these strategies is that they behave more like equities than sovereign bonds.

Through a combination of the falling quality of sovereign bonds and the forced diversification of their fixed income bucket, investors have made their fixed income allocation more equity-like. This is far from reassuring! It’s probably the exact opposite of true diversification.

Alternatives, meanwhile, are a fantastic concept, but quantitative easing has killed off volatility and, in the process, most long-volatility strategies. This has led investors to favour equity-related strategies instead. What’s more, the hedge fund space has become extremely blurred between areas such as liquid and illiquid absolute return, total return, unconstrained fixed income and more.

At the same time, equities have become more complex. The ultra-low-yield environment and the search for return has resulted in some parts of the equity universe becoming a substitute for fixed income, and more likely to react negatively if rates increase sharply. In short, equities are becoming more like fixed income, and fixed income like equities.

So to summarise, we still use an old asset allocation system that has worked well overall for a number of decades – even though the returns of equities were hard to predict, fixed income returns were relatively easy to forecast over longer timeframes. But now, the ingredients of a traditional balanced portfolio have changed massively in nature. Against this new backdrop, what are our chances of future success?

First, we have to understand that for the moment we can no longer count on fixed income to create diversification in our investments.

If one of the two main motors of portfolio performance is off for now, the only one still working is the equity beta motor. We can all conceive the effects of a world in which sovereign bonds don’t provide the security that they used to. In such a scenario, we would have to sharply increase the speed at which we add and remove risk to and from portfolios as the safety jackets we used to have on board no longer work. We would also need to make more tactical allocation moves.

Second, we would have to expand our investment universe to improve diversification and seek additional sources of return.

So what do we as asset allocators have at our disposal in our armoury today? Not much, other than factors. Factors are quite similar to asset classes in the sense that they both require a risk premium, but factors are more straightforward to deal with – they are a pure mathematical measure.

Factors don’t care if the data series relates to convertible bonds or equities; they will spot the moving force behind the series.

The bad news is that client constraints, regulation and investment processes will have to evolve for us to be able to use them in place of the old system combining bonds and stocks. I hope this will happen quickly enough that we don’t have to wait until major issues arise in traditional asset allocation, forcing us to recognise that there is a problem.

Factors are not a panacea. Their relationships are not stable over time and their definitions are still to be standardised, but they provide us with a useful level of diversification.

This is the case for the standard factors such as capitalisation size, quality, dividends and country, as well as the more alternative ones such as momentum and carry.

Factor investing is not a cheap option, but it represents a fascinating new dimension in how to measure and respond to risks.