Taking account of regime shifts
Understanding what economic regime we are in and for how long before we transition to a different one is vital for any strategic asset allocation.
Predicting the returns for different asset classes is the Holy Grail of asset allocation. The problem is that risk premiums and returns are instable over time. According to our analysis, over the long term (our data stretches back 115 years) there is a 90 percent probability of achieving an annual average return of 8 percent with a 60/40 portfolio. But that probability declines sharply as the time span shortens. Once the standard deviation reaches a certain point, one could argue that to speak of an “average” is meaningless and that investment success over a reasonable time horizon becomes a matter of luck. By the same token, one could argue that any bid to measure a “standard” risk premium for the purposes of long-term investing is just as meaningless.
But starting from the fact that the investment environment alters over time, we believe that reasoning in terms of macroeconomic “regime” can help determine an appropriate strategic asset allocation. This regime approach posits that a strategic asset allocation requires deep analysis of the macroeconomic issues driving market returns that goes beyond one centered on traditional risk premium factors. The probability of achieving higher returns is reduced if one adopts a stable view of risk premiums: these higher returns, we believe, are more likely if one accepts that risk premiums vary across regimes.
Accepting that changes in the macroeconomic environment are the true drivers of variations in risk premium has been part of the resurgence in a risk factor-based approach to asset pricing and is now an intrinsic part of strategic asset allocation. Understanding what economic regime we are in and for how long before we transition to a different one is vital for any strategic asset allocator. We believe that the essential macro risk factors are real economic growth and inflation. The 10 years between 2007 and 2016, in the US, can be characterized as a “low growth and low inflation” regime, for example, with both measures struggling to exceed 2 percent. Our present strategic asset allocation is including the possibility that we are currently moving into a new regime.
An illustration of varying risk premiums as a function of macroeconomic regimes can be identified through asset return analysis. For example, since 1950, the S&P500 returned on average 5.6% annually in a low growth and low inflation regime, but 12.1% in a high growth and high inflation regime, with these regimes occurring 4% and 10% of the time. In the regime that prevailed the most often—that is to say, the moderate growth and moderate inflation regime, which occurred 39% of the time—the S&P500 returned 15.6% on average annually.
Inflation and growth vary over time, so ability to spot shifts in trends is fundamental in strategic asset allocation. The next step is to take account of a large range of factors in order to develop a view of a potential change in macroeconomic regime.
Using growth and inflation data stretching back to the end of World War II, we at Pictet Wealth Management have developed a methodology that identifies nine main economic regimes resulting from the interaction of three different phases of inflation and three types of growth. Regime shifts occur when the interaction between inflation and growth changes—with varying degrees of probability. We keep in mind that abrupt changes from one regime to another (say, from a regime of sluggish growth and disinflation to an innovation shock that produces high inflation) are highly improbable. But we believe our methodology is pertinent over typical strategic asset allocation horizons.
We do admit that more in-depth research is needed in this area. Using the Markov-switching model, for example, we can identify nine overarching changes in the economic environment in the US since 1950—but the importance the Markov model assigns to some highly improbable scenarios limits its usefulness in mapping out regime “shifts” according to our criteria. Nonetheless, our research indicates that the rebalancing of equity weightings within a regime-based portfolio to take account of shifts in the inflation and growth regime could help boost returns over a typical long-term investment horizon.