In reviewing the behaviour of markets in the first quarter we can use the prism of our Model Portfolios as well as the general benchmarks to try and capture the important elements of what happened, partly for review, but also hopefully for some forward looking insights and perspective.
The quite extra-ordinary events of 2020, and from a market perspective the dramatic rallies that began almost exactly a year ago, have tended to mean that the calendar numbers have obscured some of the shifts that occurred within the 12 month period. In particular we saw the emergence of some important directional trends during q4 2020 that have continued into q1 2021, most noticeably the shift in perceptions on growth and inflation as markets looked through the near term uncertainties around the US election and the arrival of Covid vaccines. As we put it at the time, the markets chose to shorten duration, which meant initially in q4 a rotation towards cyclicals and commodities and subsequently in Q1 a shortening of duration in terms of both bond yields and ‘equity duration’, whereby profits were taken in the winners of 2020 (principally tech) and switched into high(er) yielding stocks in more economically sensitive areas.
Bonds – an existential threat from MMT
Our Global Bond model portfolio had been heavily weighted to Long Bonds (via the TLT US ETF) in q1 of 2020, but had been rotating away since the late summer and by q4 2020 we were zero weight in this area as the risk numbers in our models rose sharply. We remained out of the area throughout q1 2021 which undoubtedly helped our relative performance, as while the Portfolio was down 1.8% in q1, the overall Global Aggregate Bond benchmark was down 4.5%. Also helping was the (relative) bias towards High Yield Bonds, again consistent with the risk numbers.
Generally though, with 3 out of the 5 Bond subsectors (long Bonds, Investment Grade and Global Aggregates) at our maximum risk level of 5 and with TIPS at 4 and only High Yield broadly neutral at 3, it is difficult to see much case for any fresh allocations to bonds at this point.
The strength of long bonds in q1 2020 helped us to a 12.8% return for the calendar year 2020 versus the benchmark 9.2%, the importance of q1 demonstrated by the fact that the 1 year returns (i.e. April to March) were 5.6% and 4.4% respectively. The idea of our Dynamic Asset Allocation is to take more ‘risk’ – as conventionally defined by volatility – when our own risk models suggest that it is appropriate to do so and thus we would normally expect the volatility of our Model Portfolio to be higher than the benchmark. Over 3 years this is true – at 7.6% volatility versus 4.8%for the benchmark and similarly 6.4% versus 4.5% over a 5 year period. But this increase in so called risk also pays off hugely in returns, 22.5% v 8.6% over 3 years and 34.3% versus 14.1% over 5 years. Interesting then that in q1 we saw annualised volatility for the Model Portfolio actually lower, at 1.96% versus 3.17% for the benchmark.
The nerves in the Bond markets reflected the recognition of the dramatic increase in Government spending that is coming down the pipe. After announcing $3.2trn of Covid relief spending in 2020, the US have announced a further $1.9trn for 2021 with an additional request for a further $2trn on infrastructure. With real yields already low to negative, it is only those forced to own bonds for so called regulatory requirements that are doing so.
Core Global Equity – cyclical rotation towards cash flow and value
The rotation towards economically sensitive equities was flagged up during q4 by the drop in the risk scores for both Size and Value in the basket of Global Factors, principally at the expense of Momentum. While this could perhaps initially have been explained away as position squaring, this continued into q1, suggesting it is more tactical, even perhaps strategic in nature. The continuation of this trend obviously helped our returns for the quarter at 5.5% versus 4.9% for the broader benchmark. The dramatic market moves of the last year or so have illustrated a key advantage of our Dynamic Asset Allocation process however, which is that during q1 2020 we were able to go heavily underweight equities generally (100% cash in February 2020) before putting money back in from April onwards. This meant that, while we missed some of the dramatic rally from the very lows, we also missed having to receive those lows. Thus while the 1 year number (April 20 to March 21 ) shows our model at ‘only’ 38%, while the benchmark is at 52%, the calendar number is arguably more representative at 21.2% versus 15.9%. Similarly the 3 year numbers of 57.3% v 45.6% or the 5 years at 106% versus 87.1% show not only higher returns but lower volatility. It’s an investment truism that people always criticise active managers for not outperforming a benchmark that they themselves did not want to invest in at the time. The advantage of our DAA is that it allocates on a disciplined risk return basis and helps avoid the tendency to sell at the bottom (or at least sit on the sidelines for too long).
Thematic Equities – a pause due to profit taking
The thematic equity basket is biased towards ‘long duration’ equity and was always going to have a weaker q1 than the core equity basket in the light of the rotations discussed above. True to form it fell 3.4% versus a positive return of 4.5% for the benchmark in the first quarter, reflecting a weakness across most tech and growth stocks. Traditional active manager would most likely be coming under a lot of ‘pressure’ for quarterly ‘alpha’ of -8%, but we need to put this in the context of the 35% positive alpha for 2020 as a whole. Interesting, that while none of the Equity Thematic baskets has hit the highest risk level on our models (they are currently all at 3), the one that is in there as a ‘balance’, Gold, and which helped a year ago, has been sat at level 5 (highest) for a number of months now.
The growth bias of course meant that the 2020 numbers were much higher than the Core Equity, at 61.2% for the April to March 1 year ‘rally from the low’ number and 51.9% for 2020 as a whole. Both considerably better than the benchmark as well obviously. On a 3 year view, this is now giving annualised returns of 17.6% versus 13.4% for the benchmark, but interestingly with volatility at 10.9% versus 15.8%. As with the Core Global Equity, a significant part of this was down to the fact that the Models were at 100% cash in February and March last year.
Within the basket of themes, the strongest individual performer in 2020 was the Clean Energy ETF, which we acknowledge has a degree of risks such as those associated with smaller stocks and concentration risk (smaller universe chased by growing number of funds), but equally we acknowledge those as a source of potential returns.
Other Thematics: 5Cs plus Japan
During q4, we introduced the idea of the 5Cs as a further way of playing the emerging economic and market environment in addition to our Core and Thematic Dynamic Asset Allocation funds. The 5 Cs, were China, Commodities, Convertibles, Cash Flow and Crypto Currencies. Of the 5, the biggest risk reward has been Crypto Currencies as proxied through the Grayscale BitCoin ETF, which has more than doubled since November. As noted earlier, it seems that for now at least Bitcoin is acting as the ‘balance’ to portfolios previously provided by Gold. Convertibles meanwhile have only given a 4% return on the quarter, albeit after a 53% return in 2020. Partly this was down to the relatively high weighting in Tesla convertible bonds but the fundamental idea of yield plus growth remains attractive. Commodities, which we proxied through the ETFs of PICK (industrial metal miners) and WOOD, (timber related companies), are up 17,2% and 11.2% respectively year to date. The WOOD ETF is up 21% since the start of q4 and the rotation into commodities we have been discussing, while PICK is up over 60% over the period. Cash Flow is of course quite generic, but we proxied it previously with the Global Equity Factor Quality Index, which is part of our Core Equity basket and is up 7.1% year to date and 19% since the start of q4. The outlier therefore is China, where the proxy we used (CNYA LN) is down 2.9% year to date and up ‘only’ 12.4% since the start of q4. However, the Chinese markets themselves were up around 50% last year, so here again, there is a need for some consolidation.
A final idea that we had was to combine Asia, Commodities, Value and cash flow as ideas into a play on the Japanese Trading companies, a group of stocks that Warren Bufffet (rather than leveraged hedge funds) started to buy into at the end of last year. Having cleaned up their balance sheets and with their exposure to both commodities and global trade – as well as being ‘Asia but not China’ these tick a lot of strategic as well as thematic boxes. Since the start of q4 an ETF that tracks a basket of these stocks is up 26%, (17% year to date). Note however that this is in Yen terms – it is lower for a $ investor as the $ rallied from 103 to almost 110 over that period. However, that rally has stalled and if one believes the bond markets and their view on relative inflation in the US, then we can throw a weaker $ in as a another reason for international investors to diversify into this area.
