AlphaClone Alternative Alpha ETF (BATS:ALFA) is a significant holding for me and my clients. It is a hedge fund replication fund, but instead of replicating the betas of a broad hedge fund index (like QAI, another favorite of mine and one that I use as a benchmark), ALFA seeks to "clone" the alpha of the best hedge funds by identifying the stock positions held in common by "established hedge funds." Some of the hedge funds and managers mentioned in their literature include such luminaries as Bridgewater Associates (Ray Dalio), Renaissance Technologies (Jim Simons), Greenlight Capital (David Einhorn), Pershing Square (Bill Ackman), and Paulson & Co. (John Paulson).
AlphaClone Founder and CEO Maz Jadallah developed the proprietary techniques used by the fund and the underlying index, the AlphaClone Hedge Fund Long/Short Index. Their raw database consists of the 13f filings that all large (over $100 million in assets) money managers are required to file with the SEC each quarter listing their long portfolio positions. AlphaClone's proprietary Clone Score rates the reward to cloning the manager's past 13f filings, and they focus on the top 300 hedge funds based upon this score. Subsequent performance of 13f holdings may be greatly affected by a manager's rate of turnover. Some, like Renaissance Technologies, have very high turnover, which limits the usefulness of 13f filings that have a delay of 45 days after quarter-end. However, Jadallah asserts that the average hedge fund has an average holding period of about one year, so even somewhat stale holdings information can potentially be quite valuable.
ALFA holds up to 100 positions, and will partially hedge its stock market risk with a 50% hedge if the SPDR S&P 500 Trust (NYSEARCA:SPY) is below is 200-day moving average. Most of the holdings are equally-weighted, but have an "overlap bias" meaning that a name held by twice the number of managers will have twice the weight. The index is adjusted quarterly, four days after the SEC 13f filing deadline. The 50% hedge is put on or taken off at calendar month-ends.
Does ALFA have alpha? To determine whether an ETF has historically generated an alpha, or risk-adjusted return, I start by measuring the sensitivity of ALFA's return to four risk factors that capture much of the risk common to most ETFs:
- Stock market risk (MKT), as measured by the S&P 500 Index
- Bond market risk (LTB), as measured by the 10 Year Treasury Benchmark Index
- Currency risk (DLR), as measured by the U.S. Dollar Index
- Commodity risk (OIL), as measured by the West Texas Intermediate Crude Oil Index
I use exponentially-weighted 36-month rolling multiple regressions to measure these risk factor sensitivities, or betas. ALFA started trading on May 30, 2012, but the underlying index goes back to February 29, 2000. I use the index returns to calculate risk factor sensitivities in order to take advantage of this longer history, since I know that as an index fund, ALFA's returns will be very close to its underlying benchmark index.
As shown in the graph above, ALFA's equity beta (red) and 10-year bond beta (green) have both moved around quite a bit. No doubt much of the movement in the MKT sensitivity is due to putting on and taking off the 50% S&P 500 hedge, making this fund's equity beta difficult to forecast. The growing negative sensitivity to LTB prior to May 2013 no doubt indicates that many hedge fund managers correctly anticipated the market's fear of a Fed tapering of QE3 and consequent rise in interest rates. They took profits and pulled back after bonds got clobbered in May and June of this year.
Since ALFA's May 30, 2012 inception, it has performed exceptionally well (see above graph). As expected, the fund (black) has honed very closely to its benchmark index (gray). Also as expected for an equity-oriented ETF, much of its return is explained by its MKT beta return (red). Using my 36-month regression methodology, the fund's MKT beta has averaged .57 since inception. To calculate the return from MKT beta, I multiply the previous month-end MKT beta times the monthly price return of the S&P 500. I use the same methodology for the other three risk factors. The residual return is the total return minus the return from the four risk factor betas.
It is rare to find an ETF with ALFA's level of risk-adjusted, or residual, return (orange), which has been running at rate of over 10% per year in the 18 months since its inception. Granted that's a short time period, but still, that's a lot of alpha!
However, the true MKT beta for this fund is very difficult to measure and forecast accurately because of the hedging technique used. To be conservative, in the graph above I adjusted the MKT beta to 1.00 and recalculated the residual return, shown at right. (This is almost certainly an over-correction.) Much more of the fund's total return is now explained by MKT beta (red), and much less residual return (orange) is left. At a beta of 1.00, the MKT line is the price return of the S&P 500. That is, we are subtracting 100% of the price return of the S&P 500 from the total return in our calculation of the residual return. However, even after this draconian adjustment, the residual return was still a very respectable 3.6% per year since inception.
If we apply the same draconian adjustment to the underlying index (gray) since its inception (see graph above), however, the return to MKT beta (red) does not loom nearly as large over this longer time period, and the residual return (orange) comes through with very impressive power and persistence.
The cumulative return of the index yields an annualized return of over 20% per year over the nearly 11 years of history available. A skeptic (like me) might assume that this is the result of data mining and back-fitting. Purveyors of new ETFs often seem to "discover" strategies with astounding historical back-tested performance, only to have their live performance fizzle out.
However, ALFA's live results have actually been better than its pre-launch index history. In the year and a half since its May 30, 2012 inception, the fund has returned roughly 50%, or about 30% per year! As shown above, roughly half (using 36-month regression for MKT beta) to three-fourths (using a MKT beta of 1.00) was due just to a rising stock market. However, there was still a lot of alpha (residual return) left either way. And the last graph of the history of the underlying index for the strategy shows that most of the return has come from sources other than my four risk factors over the long-term.
It sure looks like there is alpha here. What could I be missing? It could be that my risk factor sensitivities for ALFA are inaccurate. The most likely problem, if there is one, is with my estimate of MKT beta. We saw that raising the MKT beta from an average of 0.57 to 1.00 ate up quite a bit of the residual return. Perhaps the true underlying MKT beta of the holdings was higher than 1.00. I do not think that is likely but I can't prove it. A somewhat more likely scenario is that ALFA's holdings were highly sensitive to some other risk factors than my four, and that adding them to the risk model would explain some of the residual return. The suspects that come to mind are small cap, credit, or emerging market risk. I do not include these as risk factors in my model because 1) they are usually highly correlated to S&P 500 risk (causing statistical problems for my regressions) and 2) they are more expensive to hedge away than S&P 500 risk. But again, I can't prove that they wouldn't explain at least some of the return.
Even so, given the level and persistence of the residual returns using my four risk factors, I am comfortable concluding that yes, there does seem to be alpha in ALFA.
Disclosure: I am long ALFA. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: The information contained in this article has been prepared with reasonable care using sources that are assumed to be reliable, but I make no representation or warranty regarding accuracy. This article is provided for informational purposes only and is not intended to constitute legal, tax, securities, or investment advice. You should discuss your individual legal, tax, and investment situation with professional advisors.