By Paul Sweeting, European head of strategy at J.P. Morgan Asset Management
So far, the 21st century has not been much fun for institutional investors. Equities have underperformed not only expectations, but fixed income as well, calling into question the 60/40 split between equities and bonds that has become the baseline for strategic asset allocation.
In this harsh environment, the risk parity approach to strategic investing has flourished, not only because its leveraged fixed income exposure has produced high returns, but also because it advances a novel investment principle: that the key to asset allocation is to allocate equal shares of portfolio risk to each asset class. Clearly, leveraged exposure to a top-performing asset class is likely to produce satisfied investors. The important question for investors is whether risk parity will work as well in the future. This is more a matter of the investment principles it involves than whether bonds will continue to outperform stocks, although leveraged fixed income does seem like a less attractive future strategy given the current low yields. The difficulty of building portfolios based on return assumptions – the central tenet of modern portfolio theory, which dominated asset allocation for more than half a century – was one of the drivers behind the development of the risk parity theory. The larger the return assumed for an asset class, the more it is overweighted, and also, the greater the chance it will underperform expectations. risk parity rejects making return forecasts and focuses on equal risk shares instead. However, using equal risk shares is also problematic as a guide to investment choices; a more flexible approach could be beneficial. To the extent that you are sure that the average return on stocks is going to be equal to your 8% forecast, it would not seem wise to ignore it. But if you are uncertain, you should probably tone down your reliance on the forecast, and if you are very uncertain, you could be better off ignoring the forecast altogether, which is what risk parity does. Forecasts are by definition uncertain, and investors would do well to factor this uncertainty into asset allocation decisions. One solution is to use the “forecast hedge” asset allocation rule. This strikes a balance between the conventional approach to asset allocation (ignore the uncertainty) and risk parity (ignore the forecasts), by measuring the amount of uncertainty and weighting the two accordingly. In tests across a wide range of return environments, the forecast hedge outperforms the two extremes. Here, as elsewhere, there is no free lunch. To use the forecast hedge, you need to supply a measure of the confidence or uncertainty you attach to your return forecasts. This is extra work and it may be difficult to do this precisely, but these are not reasons to avoid trying. After all, if you are uncertain about your forecast uncertainty, how can you be certain about your forecast? Moreover, our analysis suggests that precision is not crucial and that the benefits of taking a position on uncertainty can be substantial. Download the paper Improving on Risk Parity: Hedging forecast uncertainty at http://am.jpmorgan.co.uk/ institutional.
Comments