Updated: May 18, 2022
I recently came across a paper written in 2008 by Robeco’s Pim van Vliet and David Blitz entitled Dynamic Strategic Asset Allocation: Risk and Return across Economic Regimes. Those of you who have been following us for the last couple of years (or longer!) will know that our approach to tactical asset allocation is centred around analysis of the business cycle. While van Vliet and Blitz make a distinction between tactical asset allocation and what they call “dynamic strategic asset allocation”, they can in this instance be considered the same thing (they are both seeking to optimise in relation to a static strategic asset allocation though in slightly different ways).
"Equity returns are best during ‘recession’ and ‘recovery’ phases"
The paper’s analysis focuses solely on the US but considers a reasonably wide range of asset classes: equities (large, small, value & growth), Treasuries, credit, commodities, and cash over 60 years. The authors use four indicators to define the business cycle: credit spread (difference between the Baa and Aaa spreads), earnings yield (E/P ratio of the S&P500), the ISM manufacturers’ survey production index, and the seasonally adjusted U.S. unemployment rate.
They then assign each month of the indicator to one of four phases of the business cycle (expansion, peak, recession or recovery) depending on whether it is high and rising (expansion), high and falling (peak), low and falling (recession). Finally, they consider the performance of each of the aforementioned asset classes in each of the four phases and build recommended portfolio weightings for each phase for each asset class in relation to a fixed strategic asset allocation (the objective of the Dynamic Strategic Asset Allocation is to optimise both return and volatility in each of the four phases). The results are summarised in the table below.
As for returns of each asset class in each of the four phases (as well as over all phases), they are summarised below. Note that returns are stated as excess returns in relation to cash.
There are numerous interesting features of the above two tables as well as of other parts of the study: It appears that recommended weights do not differ that much from SAA weights. This is because the study imposes tracking error limits. Increasing tracking error limits would increase the size of positions in relation to SAA.
Although equity returns are below average during expansion phases (3.7% during expansion versus 5.6% for all phases), the model still recommends overweighting equities (44% total versus SAA total equities of 40%).
This is because returns from other asset classes during expansion phase are also below average (e.g. bonds negative 0.4% versus +0.6% for bonds during all phases).
Although equity returns are best during the phases ‘recession’ and ‘recovery’, the model still recommends underweighting them in relation to SAA.
In fact, the model also underweights equities during the ‘peak’ phase as well, so equities are underweighted in three of the four phases.
This is because the model is seeking to optimise in relation to volatility as well as return (models that optimise for return only will overweight equities during the phases ‘peak’ and ‘recovery’).
It should also be noted that the model makes material shifts between ‘large’, ‘small’, ‘value’ and ‘growth’ during the four phases, even though overall equity recommended weights do not move far from SAA weight.
An allocation to credit is recommended in one phase only: ‘recovery’.
The strong performance of credit in recent months and years on both sides of the Atlantic suggests that we may still be in ‘recovery’ phase, though inflation data suggests the US is closer to ‘expansion’ than Europe.
It may well be time given the various conclusions of the paper to be moving out of credit, at least in the US (note the strong performance in 2016 of CCC credits mentioned earlier in this letter). The question is, what should one move into? Emerging market debt might be one option, particularly since there tends to be a positive correlation between emerging markets and commodities, which tend to do well during the ‘expansion’ phase that we may well be moving into in the next year or so.
As the paper notes, “most assets exhibit above-average returns during recessions and recoveries and below-average returns during expansions and peaks”. Although in some ways this is counter intuitive, it is also generally understood that risky assets tend to anticipate good times (expansions and peaks) well in advance i.e. they perform well during recessions and recoveries (bad times).
Many other studies use NBER data to define different phases of the business cycle which is problematic because such data is only available ex post. This study on the other hand uses four indicators that are available ex ante, and is thus one that has more practical application for asset allocators.
Published in Investment Letter, January 2017
The views expressed in this communication are those of Peter Elston at the time of writing and are subject to change without notice. They do not constitute investment advice and whilst all reasonable efforts have been used to ensure the accuracy of the information contained in this communication, the reliability, completeness or accuracy of the content cannot be guaranteed. This communication provides information for professional use only and should not be relied upon by retail investors as the sole basis for investment.