For Profit Announcements

08Feb13

A couple of for profit announcements…

num.01 For all the reasons I’ve discussed previously, I’ve officially cut ties with tactical asset allocation (and/or its many variations) as my core wealth management solution.

In my previous post I talked about why the concept has lost its luster for me: the one-two-three punch of middling returns, long hold times, and most importantly, the limited upside remaining in Treasuries.

Like most folks who do what I do, I enjoy showing off the very good things we’ve been doing for the last seven years, but unlike most folks who do what I do, I don’t believe in hiding our uglies (even ones that we freely shared like the TAA model), so I’ll continue to carry the real-time actual results of the TAA model indefinitely at MarketSci.com.

num.02 As subscribers to the Volatility ETF Strategy already know, this month we’ve added a partial beta hedge to an already very effective strategy.

The goal is to reduce exposure to the portion of VXX/XIV returns driven by changes in the stock market, which is inherently difficult to predict, while maintaining exposure to the portion of returns driven by the VIX futures term-structure (i.e. contango or backwardation) that are much easier to predict.

This requires margin to precisely mirror the trades we’re taking in our own portfolios, but because of the negative correlation between VXX/XIV and the hedge ETF, it’s designed to actually reduce portfolio volatility. For users who cannot or do not wish to employ margin, we also provide an unmargined signal.

Click to learn more about the Volatility ETF Strategy at MarketSci.com.

Happy Trading,
ms

. . . . .

To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed or Email Feed.



11 Responses to “For Profit Announcements”

  1. 1 JRC

    Wondering – what type of strategy are you using for your core wealth now that you’ve dumped TAA? You alluded to something new…

    • 2 MarketSci

      Hello JRC – building up a bit of real-time results now in a separate tracking account (as of the beginning of January). Will start talking about it in more depth once I have a couple months of verifiable returns under the belt. michael

  2. 3 Jon

    I take it adding something like TBF to the TAA mix didn’t do it for you?

    • 4 MarketSci

      Hello Jon – good question – few issues here:

      1. Saying that there is a mathematical certainty of limited upside in Treasuries is very different than betting on Treasuries to go down. Returns could be middling for a long time to come.

      2. I haven’t seen anyone produce a Treasury timing model good enough to overcome point #1 above.

      3. And even if such a model existed, you would lose the inverse correlation to equities (and most other major asset classes) that is the basis of the low portfolio volatility that all of these models exhibit historically.

      That’s all a long way of saying no, TBF doesn’t do it for me.

      michael

  3. Hey Michael,

    You’ve put a stake in the ground on your TAA call, so I don’t expect you to recant. However, there is some pretty convincing evidence that TAA / trend related strategies do just fine in rising interest rate environments. A couple of simple examples:

    In AQR’s paper, “A Century of Trend Following”, they provide returns by decade, and during the decade from 1973 – 1982 – a period of surging rates (especially toward the back half of the decade) – trend following delivered over 40% per year. In fact, it was the best decade of the century.

    Faber’s simple but elegant paper on TAA goes back to 1970, and the 70s rising interest rate environment didn’t look much different than the other three decades in the sample.

    Admittedly, these examples are not exhaustive, and you might assert that they are imperfect analogs, but they do represent examples of how simple trend based asset allocation strategies have proved reslient to rising interest rate environments in the past.

    It’s been a weak few years for trend following and TAA, but all strategies go through sideways periods. I’d say there is more than enough evidence that trend following works across regimes for us to persevere, and a year or so of contrary evidence is hardly statistically meaningful.

    Again, this is your backyard, so I don’t want to be argumentative, but it seems to me that you may just need to be a little bit more patient.

    Love your stuff!
    Adam

    • 6 MarketSci

      Hello Adam – I’ve covered this previously. (I quite enjoy you folks’ work, so don’t take this as snark, only a difference in conclusion)…

      Mathematically speaking, there is a big difference between rates rising from an elevated level (like the 1970’s) and rates rising from a depressed level (like today) in terms of the impact on bond funds, which tend to maintain a somewhat constant maturity. In the second scenario, there is less coupon today to compensate for the change in bond prices. I’ve heard this argument re: the 1970’s multiple times before, but (again, mathematically speaking) the two are in no way analogous.

      It’s a pretty simple exercise to estimate the impact of changes in rates to constant maturity investments of various lengths. I would suggest running the numbers and looking at how such an investment (ex. 10Y UST) would perform when rates go from say 5 to 10% versus 0 to 5% over realistic timeframes. Very, very different results.

      michael

      • Hey Michael,

        Thanks for the quick response.

        I just consulted my long term constant maturity Treasury chart, and it looks to me like rates bottomed near 2% sometime in the mid 40s at near 2%, rising to about 5% by 1960, and 8% by 1970 (all approximate). If your thesis is correct, we should see poor performance in the 50s and 60s as rates came up off of their lows.

        43 – 52: 19.4%
        53 – 62: 24.8%
        63 – 72: 26.9%
        73 – 82: 40.3%

        The worst periods over the past century for trend following appear to be 1933 – 1942 and 2003 – 2012 (with most of the poor returns to the latter period coming over the years 2009 – 2012). These are actually periods where rates fell from low levels to very low levels, and both are periods of extremely high levels of government intervention.

        Trend following relies on the development of organic trends in markets that are slow to be recognized by market agents. The purpose of government intervention is to short circuit organic trends when those trends are politically unpalatable.

        So long as markets are dominated by government intervention, trend following strategies are likely to struggle at times, but as you know, active risk management and diversification are likely to limit negative outcomes to longer periods of flat returns rather than major drawdowns. And eventually the times will change.

        Thanks again,
        Adam

      • 8 MarketSci

        Hello Adam – based on your response, I’m not sure if you’re arguing that UST will or will not underperform historical norms as rates rise from these depressed levels. I’ll assume you’re agreeing that it will underperform, b/c the math seems pretty simple on this one. So setting that aside…

        if I could summarize your response, it is (correct me if I’m wrong):

        Yes, mathematically UST (and by extension, similar investments) will underperform historical norms, but trend-following/diversification/momentum/etc is going to be enough to overcome that deficiency.

        That may very well be the case – that is the key debatable point for which there is no clear answer. My response would be…

        Consider the major asset classes in today’s market. Specifically, consider how highly correlated nearly all asset classes are in today’s market to equities, and how few major asset classes provide any real level of diversification with equities.

        Now imagine you take UST and similar investments out of the mix as a significant return source.

        Now consider all these backtests burning up the industry showing smooth sailing over the last few decades and ask, in a world where UST et al. is no longer a significant source of return, and almost everything is too well correlated to equities, are we really going to be able to achieve anything remotely close to that in the near future?

        Again, this is a debatable point, but I would say no. Actually, more specifically, I would say I am far less confident relative to other competing (and more compelling) concepts.

        michael

        P.S. I’m on Asia time at the moment and turning off the computer, so apologies for what will be a delayed response.

  4. On Asian time? I’m so jealous! My wife and I spent two incredible years in SE Asia, and we would go back in a second.

    Yes, the mathematics of rising rates on a constant maturity bond index are conclusive: Treasuries will not provide the tailwinds over the next few years that they have provided over the past 30. TAA enthusiasts should necessarily lower expectations.

    However, given the evidence from longer term studies across multiple regimes I would still strongly assert that TAA represents a much better strategy going forward than most other approaches. There will always be room for satellite approaches related to fleeting anomalies or esoteric indices, but TAA is likely to endure.

    Samuelson asserted that markets are micro efficient and macro inefficient, with macro inefficiency largely due to structural biases and decision making structures at large institutions. Should institutions begin to shift substantially to a focus on tactical alpha, then we might have reason to worry. However, at present AQR suggests that tactical alpha amounts to approximately 0.2% of the size of the underlying equity markets, 3% of the underlying bond markets, 5% of the underlying commodity markets, and 0.2% of the underlying currency markets – I think we have a ways to go before we need to worry.

    That said, it sounds like you are exploring an interesting new core approach. I look forward to hearing more about it when you’re ready to share ;)

  5. 10 steve

    of course you could have eliminated treasuries for HY (jnk) and solved that problem. understand that jnk is correlated to stocks but they also trend better than treasuries.

    • 11 MarketSci

      Hello Steve – one of the core concepts behind why TAA (and its alternatives) has done so well historically is diversification. Diversification with equities in today’s market is limited to just a handful of asset classes. HY is not one of those, particularly during market crises. michael


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s