Can a Tactical ETF Strategy be used as a Liquid Alternative?
This blog post doesn’t have anything to do with country-focused or factor-based investing, so hopefully you’ll stay with me. I recently attended the Investment Management Consultant’s Association annual conference in Boston. It was two days of (mostly) strong content, and gave me several ideas for new lines of research. As part of the conference, I had the opportunity to moderate a panel entitled “Liquid Alternatives and Tactical Allocation – Two Sides To The Same Coin?” In the panel discussion, we talked about the rise of tactical asset allocators (also known as ETF strategists, ETF asset managers, ETF model providers, etc.) and how the use of dynamic beta exposure management had increased substantially since 2008. We discussed the different levels of tactical management, including a spectrum of managers who range from Core/Satellite (not a lot of tactical swings) to managers who will go from 100% invested to 100% cash. We also reviewed the current availability of alternative strategies in 40 Act wrappers, discussing the pros and cons of each type of strategy and whether some were more fit than others for transparent, liquid vehicles. As we talked, it occurred to me that at the moment we were missing some good old-fashioned quantitative data on the question.
The idea of this post is to answer a simple question: Can investors use a tactical allocator instead of an “alternative” strategy to benefit the balanced portfolio of a traditional investor? What follows here is NOT an exhaustive review of the subject, and we are well aware of the difficulties in defining and analyzing alternative strategies versus traditional asset classes. We are also aware of the short-term nature of many of the tactical track records (although some are over 10 years at this point) as well as the small universe of tactical managers with a sufficient track record for the analysis.
We looked at data going back to the end of 2006, chose a comparison universe comprised of 3 tactical managers (one or two of which everyone would know), 2 balanced funds (also well-known), 3 traditional asset class benchmarks, and 7 different alternative benchmarks. The names of the tactical managers and funds have been hidden, as the specifics of which managers we are looking at are less relevant. So here is the list.
All performance in our analysis is sourced from Morningstar Direct, and covers the time period from 12/31/2006 to 3/31/2014.
Our first look was at the absolute and risk-adjusted performance of the strategies.
Table 1
In Table 1, both the tactical and balanced portfolio groups compared favorably to the alternative indexed from a return perspective, but definitely had significantly higher standard deviation than all of the alternatives except for managed futures. Each had positive alpha to the 45/45/10 benchmark, with betas higher than 1. The R2 of the multi-strategy alternatives ranged between 0.57 and 0.79, which was somewhat similar to those of the tactical managers ranging from 0.32 to 0.86. The balanced funds R2 were much higher, on the order of 0.88 to 0.92. In addition, the sharpe ratios of the multi-strat alternative indexes were in the 0.5-0.68 range, which was similar to the range of 0.38 to 0.75 for tactical and balanced managers. At least from this first cut, it looked like the tactical managers were in the same neighborhood as the alternative benchmarks.
Next we looked at the information and capture ratios for the strategies.
Table 2
In Table 2, all of the tactical and balanced strategies had strong information ratios vs. the benchmark, while the alternatives indexes struggled in that same area. The worst performers were the HFRI FoF: Diversified index, as well as the liquid alternatives fund 1. The Up/Down capture ratios, however, told a different story. In all cases, the tactical and balanced managers had higher up and down capture ratios than the benchmark, while the alternatives indexes all showed up/down capture ratios of 45%-75%. This was not a great surprise given the realized volatility differentials that we saw in Table 1. We decided to continue this line of research by looking at some of the downside statistics in Table 3.
Table 3
The theme of higher downside risk and drawdowns continued in our analysis of Table 3. The worst month for the tactical and balanced managers was similar to that of the benchmark, but higher than any of the alternative indexes. The differential was less when looking at the worst quarter, but still was slightly higher for tactical and balanced managers. The max drawdown was similar in magnitude for all of the indexes and managers, and the max drawdown valley date for all of the investments was within 3 months of each other(with the exception of managed futures that has a 9/30/13 date – but that is a subject for another blog post).
Up to this point, we had seen that tactical managers are generally more volatile than both the benchmark as well as the alternative universe. This might imply that they in fact, are not good substitutes for more traditional alternatives investments. However, the astute reader will rightly recognize that no analysis of the contribution of an investment to the overall portfolio is complete without a correlation analysis. Tables 4-6 include a correlation analysis of 3 time periods: Table 4 analyzes the time period from 12/06 – 3/09, Table 5 the time period from 3/09 to 3/14, and Table 6 the entire time period from 12/06-3/14. The reason for breaking up the time periods is to see how the strategies performed in both up and down trending markets.
Table 4 – Correlations from 12/06–3/09
Table 5 – Correlations from 4/09–314
Table 6 – Correlations from 12/06–3/14
Since there are so many numbers on these charts, it is important to note that the alternative universe includes investments #7 through #13. As we look at the correlations on Table 4, it is not a great surprise to see that most of the assets are correlated to the benchmark, and to each other. This makes sense in the context of the high inter-asset correlations that were commonplace in 2007 and 2008 during the crisis. The notable exceptions are tactical manager 2, managed futures, and bonds. Table 5, surprisingly, tells a similar story. So we moved on to Table 6 to see if we could draw some conclusions. It is clear from the tables that the investments with the lowest correlation to the 45/45/10 benchmark, and the rest of the universe are tactical manager 2, managed futures, and bonds. The alternatives universe (ex-managed futures) had correlations to the benchmark of between 0.75 and 0.89. Frankly, those aren’t great numbers in a portfolio construction sense. The balanced funds had correlations of 0.94 and 0.96, which essentially are the same as the benchmark. The tactical managers ranged between 0.57 and 0.93 correlations.
So what is the bottom line on all of this? How do tactical managers stack up to alternatives? Are they a viable element in portfolio construction? When compared to alternative benchmarks, we noted that tactical managers have the following characteristics:
- Higher average returns with higher volatility
- Similar downside characteristics
- Similar correlation characteristics
Can tactical managers perform the same function as some alternative strategies? I believe they can. As always, the trick is to figure out how to best use them in a client portfolio. Allocating a portion of the client’s “alternative” investments to tactical managers may be a good place to start.
© Accuvest Global Advisors
