Substituting Liquid Alternative Funds
March 1, 2016
Analysis of Dreyfus Dynamic Total Return Fund
A recent cover story in Barron’s features liquid alternative funds from AQR. According to the article
The liquid-alt pitch is that individuals can access the same types of investments as university endowments and other big institutions, to diversify equity-heavy portfolios, typically with a 10% to 20% allocation to liquid alts… The advantage of the [AQR Managed Futures] strategy […] is that it is uncorrelated with other asset classes, and “has the most consistently strong performance in equity bear markets.” That is when diversification matters most, as was the case in the third quarter of last year and the early part of this year.
Ideally, returns of a liquid-alt fund should not only be uncorrelated with those of both stocks and bonds but also significantly positive over a long evaluation period. Let’s take a look at the performance of three AQR funds with a sufficiently long history.
The following chart shows rolling return correlation of the AQR Managed Futures Strategy Fund (AQMIX) with the Vanguard Total Stock Market ETF (VTI) and the Vanguard Total Bond Market ETF (BND):
Please note that AQMIX had the first full month of returns in February 2010. Consequently, the first rolling 36-month return became available at the end of January 2013. As could be expected, the fund had lower correlation to stocks than to fixed income, although both coefficients were quite low (generally, correlation below 0.6 provides diversification benefits).
Here is a similar chart with related statistics for the AQR Multi-Strategy Alternative Fund (ASAIX):
Compared to AQMIX, this strategy had a higher correlation to bonds.
Here is a similar chart with statistics for the AQR Diversified Arbitrage Fund (ADAIX):
In contrast to AQMIX and ASAIX, this strategy had a higher correlation to equities than bonds; however, both coefficients were still pretty low.
The problem with any of these strategies is the lack of accessibility for most individual investors:
AQR’s approach can be hard to understand. Because of this—and to deter hot money—the firm sells its liquid-alt funds almost entirely through financial advisors. Retail buyers can access the funds directly through fund supermarkets like Fidelity, but direct investments involve a minimum of $1 million. Investments through advisors and 401(k) plans have no minimum.
Is there a way to substitute these liquid-alt funds with readily available ETFs? Let’s explore this possibility using Alpholio™’s patent-based analysis service for mutual funds. One variant of this methodology constructs a reference portfolio of ETFs with fixed both membership and weights. Here is the resulting cumulative RealAlpha™ chart for the AQR Managed Futures Strategy Fund (to learn more about this and other performance measures, please visit our FAQ):
As the statistics section below the chart shows, since its inception the fund had a smaller return and a much higher volatility (measured by standard deviation) than those of the reference portfolio. The following chart illustrates the constant composition of the reference ETF portfolio in this analysis:
The major positions in the reference portfolio were the PowerShares DB US Dollar Index Bullish Fund (UUP; fixed weight of 38.1%), iShares 20+ Year Treasury Bond ETF (TLT; 22.9%), iShares MSCI Netherlands ETF (EWN; 9.3%), Guggenheim CurrencyShares® Swiss Franc Trust (FXF; 6.0%), Consumer Staples Select Sector SPDR® Fund (XLP; 5.5%), and Utilities Select Sector SPDR® Fund (XLU; 4.7%). The Other component in the chart collectively represents addition five ETFs with smaller fixed weights.
The return correlation of the reference ETF portfolio over the entire evaluation period was 0.16 with VTI and 0.58 with BND. Given that these figures for AQMIX were approximately -0.07 and 0.21, respectively, the reference portfolio was not as good a diversifier for stocks and bonds as the fund was. However, the reference portfolio only had long positions in non-leveraged ETFs. It also returned about 8% more than the fund on a cumulative basis and with a 59% lower volatility. Similar analyses can be conducted for ASAIX and ADAIX. In the end, it is up to the investor to weigh the pros and cons of using reference ETF portfolios as substitutes for these funds in the context of the overall portfolio.
We hope that our Investment Toolkit™ will provide useful services for investors who want to construct well-diversified portfolios. If you would like to use it, please register on our website.
June 15, 2015
Merger Arbitrage Funds as Portfolio Diversifiers
A recent piece in Barron’s profiles the Dreyfus Dynamic Total Return Fund (AVGAX, Class A shares). This $1.1 billion fund has a maximum 5.75% front sales charge, 1.50% net expense ratio and 124% turnover rate. According to the article
The $1 billion fund has beaten 99% of its peers in Morningstar’s Moderate Target Risk category over the past five years.
The primary benchmark for the fund is the MSCI World Index. One of the practical implementations of this index is the iShares MSCI World ETF (URTH). Alpholio™’s calculations indicate that since the ETF’s inception in January 2012, the fund returned more than the ETF in only 25% of all rolling 12-month periods and 6% of 24-month periods. However, an all-equity index is not the best choice for a benchmark of a fund that
…normally invests in instruments that provide investment exposure to global equity, bond, currency and commodity markets, and in fixed-income securities.
Let’s take a closer look at the performance of the Dreyfus Dynamic Total Return fund using Alpholio™’s patented methodology, which truly adjusts for a fund’s holdings and risk. In the simplest variant of the methodology, the reference ETF portfolio for the fund has a static membership and fixed position weights. The current lead manager began running the fund in May 2010. Since then, the fund produced a negative 0.6% of annualized discounted cumulative RealAlpha™ (to learn more about RealAlpha™, please visit our FAQ). The fund had top equivalent positions in the iShares Morningstar Large-Cap Growth ETF (JKE; constant weight of 19.5%),
Guggenheim CurrencyShares® Japanese Yen Trust (FXY; 14.7%), Vanguard Long-Term Corporate Bond ETF (VCLT; 12.0%), and WisdomTree Europe Hedged Equity Fund (HEDJ; 10.7%). At 8.8%, the reference portfolio’s standard deviation (a measure of risk) was lower by about 0.4% than that of the fund, whose RealBeta™ was 0.65.
In a more elaborate approach, Alpholio™’s methodology keeps the membership of the reference ETF portfolio fixed but allows the ETF weights to fluctuate to better match the composition of the fund over time. Here is the resulting chart of the cumulative RealAlpha™ for the Dreyfus Dynamic Total Return fund:
From 2010 to early 2014, the cumulative RealAlpha™ for the fund declined. Despite a subsequent rebound, the annualized discounted cumulative RealAlpha™ was only slightly positive for the regular measure (+0.3%) and negative for the lag measure (-0.2%). The lag RealAlpha™ curve was generally below the regular one, which indicates that not all new investment ideas worked out as well as expected. At 9%, the fund’s standard deviation was about 0.9% higher than that of the reference portfolio. The fund’s RealBeta™ was 0.6.
As the following chart shows, investor’s could have avoided a period of the fund’s relative underperformance by taking advantage of the buy-sell signal automatically generated from the smoothed cumulative RealAlpha™:
The final chart illustrates changes of ETF weights in the reference portfolio for the fund over the same analysis period:
The fund had equivalent positions in the iShares iBoxx $ Investment Grade Corporate Bond ETF (LQD; average weight of 18.4%), iShares Global 100 ETF (IOO; 15.1%), iShares Morningstar Large-Cap Value ETF (JKF; 14.8%), iShares 1-3 Year Treasury Bond ETF (SHY; 14.7%), iShares Morningstar Large-Cap Growth ETF (JKE; 9.1%), and iShares MSCI Germany ETF (EWG; 7.2%). The Other component in the chart collected represents six additional ETFs with smaller average weights.
After a true adjustment for risk and factor exposure, the Dreyfus Dynamic Total Return fund did not exhibit a remarkable performance. The median 36-month correlation between the fund’s and a broad-market ETF’s (VTI) returns was around 0.9, so the fund was not an effective diversifier for a domestic stock portfolio. The fund’s steep front sales charge for smaller investment amounts certainly does not enhance its appeal.
To learn more about the Dreyfus Dynamic Total Return and other mutual funds, please register on our website.
November 10, 2014
REIT Correlations with Stocks
A recent article in The Wall Street Journal’s Investing in Funds & ETFs report discusses merger arbitrage mutual funds. According to the article, such funds
…may offer an attractive way to diversify away from the risks of stocks or bonds …[but] can’t replace bonds, because their returns aren’t certain and come mostly through any price appreciation, not yield. But held in tandem with bonds, they can offer a way to hedge against interest-rate risk and might cushion part of a portfolio against stock-market volatility
Let’s take a closer look at these statements with the help of a recently introduced Alpholio™ App for Android, and specifically its Portfolio, Correlation, Total Return and Efficient Frontier services. For the purposes of this analysis, the base portfolio consists of 60% SPDR® S&P 500® ETF (SPY) and 40% of the iShares Core U.S. Aggregate Bond ETF (AGG), i.e. a traditional balanced mix of stocks and bonds. Here is the baseline chart with statistics generated from total monthly returns of both ETFs and quarterly rebalancing of the portfolio:
The reason why the beta of this portfolio is not exactly 0.6 (i.e. equal to the 60% weight of the SPY) is threefold. Alpholio™ uses a broader definition of “the market” than just the S&P 500® index. Also, the correlation between the market and AGG is not zero. Finally, the portfolio is rebalanced quarterly, not monthly, which can lead to a temporary divergence of SPY/AGG weights from the original 60/40% level.
For reference, in the same time frame a portfolio consisting of just the SPY would have an annualized return of 8.52% with a standard deviation of 14.25%, Sharpe ratio of 0.55 and maximum drawdown of 50.8%. Adding AGG to such an equity-only portfolio decreases its return but reduces its volatility even more, thus improving the Sharpe ratio. The maximum drawdown is also significantly diminished.
The article quotes two merger arbitrage funds with substantial assets: The Merger Fund® (MERFX) and The Arbitrage Fund (ARBFX). To effectively diversify the balanced portfolio, should either fund replace a portion of stocks, a portion of bonds, or a combination of both? What should be the extent of such a replacement?
To answer the first question, let’s take a look at the correlation between SPY, AGG and either fund using the Correlation service of the Alpholio™ app. Here is a chart of the rolling 12-month correlation coefficient for monthly returns of SPY and MERFX:
The starting date of the chart stems from the earliest availability of AGG whose first full monthly return was in October 2003. The average correlation of 0.56 indicates that MERFX was a marginal diversifier for SPY (generally, a correlation of 0.6 or less is desirable). Here is a similar chart for AGG and MERFX:
The average correlation of just below zero indicates that MERFX was a much better diversifier for AGG than SPY. Similarly, the average correlation between SPY and ARBFX was about 0.42 and virtually zero between AGG and ARBFX. Therefore, to effectively diversify the base portfolio, it should generally be better to allocate more of SPY rather than AGG to MERFX or ARBFX. However, this would also suppress portfolio returns — as the following total return chart shows, MERFX and ARBFX had steadier but smaller cumulative returns than SPY:
To answer the second question: a portfolio with the highest Sharpe ratio (i.e. the tangency portfolio) would be mostly composed of AGG and MERFX. Here is an efficient frontier chart in which the current portfolio, depicted by a standalone marker inside the frontier, had 80% in AGG and 20% in MERFX but no SPY and was very close to the tangency portfolio:
Adding MERFX at the expense of SPY decreased the portfolio volatility and increased its Sharpe ratio, but resulted in lower returns. To illustrate this further, here is a chart and statistics for a portfolio that consisted of 45% SPY, 40% AGG and 15% MERFX, rebalanced quarterly:
Ultimately, it is up to the investor to trade off portfolio returns for risk — some may choose to optimize for the highest return per unit of risk, while others may strive for higher returns at the expense of a sub-optimal Sharpe ratio. The Alpholio™ app for Android provides a set of tools that facilitate the exploration of historical data and construction of desired portfolios, with the usual caveat that the past performance is not a guarantee of future results.
January 14, 2014
Alternatives vs. Bonds
Traditionally, real-estate investment trusts (REITs) provided a good diversification to other stocks in a portfolio. However, in the last several years, REIT returns have become highly correlated with returns of other equities. One theory, outlined in a Morningstar article, is that over the years REITs evolved from a small, illiquid and neglected to a mainstream and easility accessible asset class.
As an article in The Wall Street Journal indicates
From 1980 through 2006, stock performance of REITs moved in tandem with the broader market only 47% of the time, according to an analysis for The Wall Street Journal by Citi Private Bank in New York… Since then, as the bank’s research shows, REIT correlations have jumped to nearly 80%, erasing more than a quarter of a century in decoupling.
To illustrate that, Alpholio™ compiled the following chart of correlation between returns of the SPDR® S&P 500® ETF (SPY) and iShares U.S. Real Estate ETF (IYR):
The chart shows rolling correlations in trailing three- and four-year periods using total monthly returns of both ETFs since mid-2000. (As expected, thanks to a larger number of data points the latter curve is a bit smoother but lags the former one.) Either curve is characterized by four distinct phases:
- Through 2006, the correlation was indeed in the mid-40%
- From 2007 through 2008, the correlation gradually increased to about 70% and abruptly jumped to over 80% at the onset of the financial crisis
- From 2009 through mid-2013, the correlation stayed at about 85%
- Afterwards, the correlation decreased started to decrease.
The last two phases were caused, at least in part, by the Federal Reserve’s interest rate policy: a strong coupling of rising returns stimulated by low rates, followed by an indication of decoupling when rates rose. A better economic outlook is also a factor:
Improving conditions in the broader economy usually lead to lower real-estate correlations… In fact, correlations between the S&P 500 and REITs have dropped by about 10% since late last year.
Let’s take a look at the last phase in more detail, this time using trailing 18- and 24-month returns:
Here, thanks to shorter time windows the degree of decoupling in the last phase is more evident: the correlation reverted to about 50%. This would suggest that REITs might once again help with portfolio diversification. However, as the next chart shows, REIT returns are currently negatively correlated with the interest rate on a 10-year Treasury note:
With the prospect of rising interest rates this year, REIT returns are likely to continue to be depressed. At the same time, many analysts forecast 5-10% returns of the overall equity market (for example, S&P just increased its 12-month target for the S&P 500® index from 1895 to 1940, which implies an approx. 7% total return). Therefore, until interest rates stabilize, it may be too early to declare a structural decrease in correlation of REIT returns to those of other stocks. A permanent return to pre-2007 correlation levels would certainly help with portfolio construction.
July 26, 2013
Do Alternatives Diversify?
In light of a recent downturn in bonds caused by a perception of the Fed’s upcoming actions, a Barron’s blog post and a Morningstar article explore alternative investments with “bond-like” returns. However, it turns out that these alternatives behave mostly like stocks with poor return-to-risk characteristics, and thus do not provide diversification to a broader portfolio.
To illustrate, here are correlations to stocks and Sharpe Ratios derived from Morningstar’s statistics for mutual funds and ETFs mentioned in the post and article:
These three-year statistics indicate a high positive correlation to stocks coupled with sub-par risk-adjusted returns. This observation is corroborated by a new study from the Leuthold Group cited in The Wall Street Journal article that states:
“From 1994 through May, it found that hedge-fund correlations have slowly been inching up to 0.75, almost 36% higher than earlier levels. Since a measure of 1.00 represents lock-step movements, hedge fund returns are generally following the tendencies of stocks about three-quarters of the time… Funds with correlations to stocks of 0.6 or less are prized by investors since they can significantly reduce portfolio volatility and limit risks over full-market cycles.”
In the past month or so, these alternative funds held their value well relative to bond investments. This is supported by their negative or low positive three-year correlations to iShares Core Total U.S. Bond Market ETF (AGG), as estimated by Alpholio™:
|IQ Merger Arbitrage ETF
|IQ Alpha Hedge Strategy
For reference, the correlation of SPDR® S&P 500® ETF (SPY) to AGG over the same period is -0.33. Therefore, these alternatives do not provide a significant amount of diversification to a balanced equity-and-bond portfolio, but could be marginally helpful if the portfolio contains only bonds. However, even in the latter case they could be a drag on the risk-adjusted performance of the portfolio: at 1.31, the Sharpe Ratio of SPY is higher than that of any of the above funds.
July 26, 2013
A recent article in The Wall Street Journal discusses how wealth managers are increasingly investing their clients’ money in mutual funds that use hedge-fund strategies. The idea behind these “alternative” investments is a low correlation of their returns to those of the general market, which is supposed to protect portfolios during market downturns. Unfortunately, the price paid for this is a sub-par performance of such investments in normal market conditions.
Let’s take a look at correlations of some of the funds mentioned in the article. Correlation coefficients can be reverse-engineered from data provided by Morningstar:
|Natixis ASG Global Alternatives
|TFS Market Neutral
|Highbridge Statistical Market Neutral
The above figures are based on the most recent three-year period. As can be seen, correlations of these funds to the market are quite high.
For further reference, here are average correlations of three types of “traditional” alternative assets, i.e. REITs, commodities, and hedge funds, with stocks and Treasury notes, as calculated by Leuthold Group:
In the last four years, these correlations were much higher than their long-term historical averages.
Even institutional investors keep pursuing alternatives in the name of diversification. However, true diversification of a portfolio requires not only low correlations but also high returns of assets being added. While it may still turn out that alternative investments provide some degree of portfolio protection during the next market downturn, this assumption is becoming questionable.