Thanks for reading the fifth installment of equity/fixed income momentum strategies. This article discusses the historic return profile of momentum strategies between fixed income and equity classes, and gives an overview of the reasons behind last month's relative performance and a look forward to relative returns in May. With the yields on long Treasuries hitting new year-to-date lows in April while the S&P 500 (SPY) hit new all-time highs, it was a fascinating month for these strategies, and this article will discuss implications for relative returns in the near future.
The purpose of this series of articles is to demonstrate the success of these strategies, and give Seeking Alpha readers with differing risk tolerances tips on how to employ these strategies themselves to improve the performance of their balanced portfolio. These are useful strategies for Seeking Alpha readers, especially those who allocate dollars to their investment plan on a subscription basis, like 401k investors making automatic payroll deductions. These switching strategies can be used to adjust periodic allocations to capture the momentum effect and improve portfolio returns, especially in tax-deferred accounts.
The most basic momentum strategy involving the equity and fixed income markets is between the benchmark Treasury index (GOVT) and the S&P 500 . This monthly strategy switches between the two asset classes, owning the asset class that performed the best in the trailing one-month forward for the next one month.
In difficult market environments, Treasury bonds typically rise in value as a flight-to-quality instrument while risky assets sell off. In improving economic environments, the opposite is usually true as equities rally and bonds sell-off as investors reap the returns of equity ownership and Treasury bonds are negatively impacted by rising inflationary impacts.
Below is a graph of the historical performance of the S&P 500, the Barclays Capital Treasury Index, and a momentum strategy that buys the asset class that had outperformed in the trailing one month. In these strategies, the one month holding period is a calendar month beginning with the first day of the month.
Over trailing 1, 3, 5, 10, and 40 year time horizons, this momentum strategy has outperformed the S&P 500 with lower volatility. The results above should prove interesting to any Seeking Alpha reader trying to balance their allocation between stocks and bonds. Since 1973, when the Barclays Treasury index was first published, the momentum strategy has generated 86ps of annual outperformance versus the S&P 500 while only exhibiting roughly 2/3 of the volatility. Below is a graph of the risk/return profile of buy-and-hold portfolios with varying mixes of stocks and bonds versus this momentum portfolio historically with data through year-end 2012.
This switching strategy would have allocated to stocks just over 55% (266/480 months) of the time from 1973-2012. Instead of allocating funds in a traditional 60% stocks/40% bonds balanced portfolio, allocating to the asset class that had outperformed in the trailing one month would have seen less dollars flow to equities, but cumulatively generated roughly 130bps of average annual excess returns. Using the momentum approach to allocate funds, investors would have actually put less money into stocks over that forty-year period than investors using a 60%/40% stocks/bonds split. However, the momentum approach would have produced meaningful incremental returns with lower realized volatility.
(For a mathematical look at why Treasuries were chosen as the fixed income asset class in these equity/fixed income momentum strategies, see my inaugural article on these strategies.)
Readers wishing to implement this S&P 500/Treasury switching strategy in their own portfolio would have owned the S&P 500 in April 2013, given the outperformance versus Treasuries (S&P 500 3.75% vs. Treasuries 0.1%) in March. Subscribers to this momentum theory (and articles) would have been meaningfully rewarded in April as the S&P 500 outperformed Treasuries by 104bps. This outperformance suggests that investors should want to own the S&P 500 relative to Treasuries in May as well. The momentum trade between Treasuries and the S&P 500 would have kept investors in equities in all five months so far in 2013, generating outperformance in each period thus far.
Treasury Bonds/Small-Cap Domestic Stocks
The efficacy of these momentum strategies is driven by the low correlation between Treasuries and stocks as the two asset classes perform well in differing market environments. These strategies suggest that the market is slow to adjust to new information, allowing for outsized returns by following the trade that has been outperforming in short trailing periods (and maybe more importantly avoiding stocks, or risky assets generally, when markets are falling). It stands to reason that an asset class with an even lower correlation with the Treasury market historically, and higher long-run expected returns should produce a momentum strategy profile that fares even better than the aforementioned S&P 500/Treasury switching strategy. This is exactly what we find when small cap domestic stocks are incorporated.
From 1979-2012, investors who employed a monthly switching strategy between small-cap stocks (Russell 2000 ETF: IWM) and Treasury bonds would have cumulatively ended the period with nearly three times as much money as those who bought and held the S&P 500. Over that time period, the switching strategy between small caps would have also had slightly less variability of returns than owning the broad equity market index outright. The Treasury index and Russell 2000 had slightly negative correlation between their return profiles over this time period (r= -0.01). Readers should take note that the Treasury/small cap switching strategy is a souped-up version of the Treasury/large cap switching strategy with 450bps of incremental average annual excess returns, but 400bps of incremental annualized volatility. In my February 2013 article, "The Small Cap Stock Index For You," I demonstrated that the S&P Smallcap 600 (IJR) is a preferable index to the Russell 2000, but use the latter index here for its longer historical track record.
Readers wishing to implement this switching strategy in their own portfolio would have owned the Russell 2000 in April 2013 given the massive 452bp outperformance versus Treasuries in March 2013 (4.62% vs. 0.1%). An investor implementing this strategy would have modestly underperformed in April as small cap stocks posted negative total returns and Treasuries rallied. This relative outperformance in April signals that small-cap stocks' relative underperformance is likely to extend into May as well. The aforementioned momentum strategies are therefore giving contradictory signals about large and small cap domestic stocks.
Treasury Bonds/Emerging Market Stocks
If substituting small caps for the S&P 500 further enhanced the risk/return profile of the momentum strategy historically, then moving into emerging markets should further increase both risk and expected return. With data from the MSCI Emerging Market Index (replicated through EEM) from 1989-2012, an enhanced momentum strategy is exactly what substituting emerging markets has historically produced.
Emerging market stocks have been the most negatively correlated with Treasuries of the aforementioned equity asset classes (r=-0.17). EM stocks' tremendous performance in the mid-2000s, which featured annual returns of at least 22% from 2003-2007, handed off nicely to the Treasury outperformance during the credit crisis. In the last ten years, this switching strategy has beaten the S&P 500 by over 4.1% per annum with similar volatility. Over the totality of the dataset, EM stocks have been a high beta function of the developed world, and this momentum strategy has generated tremendous returns rotating towards the hot market during rallies and towards the safe haven of Treasuries during bear markets.
Of course, the trailing twenty-five years have been historically strong for both emerging market stocks and U.S. Treasuries, which I believe we will look back on in future periods as a historic anomaly. While investors should not expect a 17.1% annualized return over the next twenty-five years, I believe that this switching strategy will continue to produce alpha. Even in the trailing five years as emerging market stocks produced a negative cumulative return, this momentum strategy managed a 7.5% annualized return, which would have outperformed the S&P 500 on both an absolute and risk-adjusted basis.
Readers wishing to implement this switching strategy in their own portfolio would have owned Treasuries in April 2013, given the underperformance of EM stocks versus Treasuries in March (-1.71% vs. 0.1%), an indicator that proved to be prescient as Treasuries slightly bested emerging market stocks for the month. Emerging market stocks continue to lag domestic stocks given the emerging world's higher weight towards lagging materials and commodities stocks, heightened geopolitical risk on the Korean peninsula and in the Middle East, and increasingly accommodative monetary policies of developed nations which are weakening their respective currencies and shrinking export-driven emerging markets competitive pricing advantage. The most oft cited emerging market index has now trailed the S&P 500 by 15% over the last twelve months as exporters have been hampered by a weak European economy and concerns over a hard landing in China depressed values. April's underperformance by emerging market stocks would suggest that investors continue to own Treasuries over emerging market stocks in May. The performance gap between Treasuries and EM stocks has shrank. If the global risk rally is to continue, emerging market stocks, and the commodities so important to some of these economies, must eventually join.
For Seeking Alpha readers interested in gleaning long-term alpha through momentum, I will be updating the performance of these three trades (Treasuries/S&P500, Treasuries/Russell 2000, Treasuries/EM Stocks) at the end of each month. By providing performance over the trailing one month, momentum investors can follow along with these trades in their own portfolios. Please check out my aforementioned article on fixed income momentum strategies, and look for an update of my equity momentum strategies shortly.
I enjoyed the commentary that followed last month's version of this article, and wanted to post along a few notes. A paired trade strategy with iShares Barclays 20+ Year Treasury Bond Fund (TLT) was mentioned. Long Treasuries would have even less positive or more negative correlation with these equity asset classes, and likely produced solid momentum strategies historically. I am structurally negative on long Treasuries, whose volatility is now equivalent to that of the equity market, but offer long-run nominal average returns of just 3%. Interestingly, since the long Treasury Index returned 4% in April, it would be the favored leg in a momentum strategy with all of these equity classes. I am hesitant to detail trades using the long Treasury index, but will add these trades if there is sufficient reader interest.
In the past, it has been suggested that I should use levered Treasury funds in these strategies. I have not found a long term index that tracks these funds using the same daily compounding methodologies underlying the ETFs, which have relatively short lifespans, and have decide to eschew using these funds for this article.
I use broad indices instead of certain mutual funds in all three of the articles to get the best historical dataset and because of these indices broad use as benchmarks. Returns are always geometrically linked total returns. Occasionally some readers have not factored in the coupon return on my fixed income strategies or dividends on my equity strategies when analyzing the returns of these security classes. As important components of the total return, these cash flows are always included.
A discussion over different types of switching signals has also been proffered, but a simple turn of the calendar has generated enough alpha to be meaningful, and is a simpler trade to discuss with the broad community.
I find considerable value in the comments left by readers, and hope to add a postscript every few months to these articles to elaborate on key points that arise.