- ALFA and GURU are presented as cheap alternatives to hedge funds.
- ALFA has a better risk-adjusted performance since inception date.
- But in fact SPY was better in risk-adjusted performance than both ALFA and GURU.
For a couple of years, strategy-based ETFs are flourishing. In a previous article, I have written about VQT and PHDG. This one looks into two ETFs that were designed almost at the same time and are based on hedge fund stock picks: the AlphaClone Alternative Alpha ETF (NYSEARCA:ALFA) and the Global X Guru ETF (NYSEARCA:GURU). Their business models rely on the idea that a "collective brain" of famous hedge fund managers should deliver a significant and steady alpha. Hedge Funds typically charge a 2% management fee and a 20% performance fee. Investors are submitted to various constraints, among them a minimum capital and planned redemption dates. ALFA and GURU are charging under 1% in net annual expense ratios, and shares can be bought and sold by any investor on any trading day. It makes them attractive for people who are not eligible as hedge fund customers. Each of them is based on a different rule-based strategy defined as an underlying index. More information about the methodologies can be found here and here.
The first table gives a summary of ETFs profiles:
Avg Daily Volume
87 (max 5.3%)
54 (max 2.32%)
GURU is more liquid, but ALFA trading volume is increasing and more than sufficient for individual investors. ALFA is more diversified with more total holdings, but more concentrated in its top holdings. Alpha may take a short position in a reference index when market-timing rules are triggered, whereas GURU is long only.
The next chart compares ALFA and GURU between 6/4/2012 and 3/21/2014. SPY is given as a benchmark.
Chart courtesy of Yahoo Finance.
Both ETFs have outperformed the benchmark. Their returns are similar, with a slight advantage for GURU.
The next table gives statistical data on the same period:
SPY x 1.25
ALFA and GURU provide a higher return at the price of a higher volatility. In fact, since their inception both have a lower risk-adjusted performance than SPY (Sharpe and Sortino ratios). The last row in the table shows that leveraging SPY by a 1.25 factor would have delivered a return close to GURU, a maximum drawdown close to ALFA, and a volatility lower than both. An annualized borrowing rate of 3% is taken into account in the calculation. In other words, for an investor with a margin account, SPY was the best of three. Moreover, using a leveraged S&P 500 ETF like SSO or UPRO with the appropriate weight makes margin unnecessary and borrowing rate pointless (read this article if you fear the decay of leveraged ETFs).
Since their inception date, the main point of ALFA and GURU over the benchmark is to offer a leverage of 25% with a higher relative risk. ALFA has the additional advantage to embed a timed-hedging strategy, which could not be efficiently tested on the period. Investors with experience might prefer managing the hedging position themselves (click here to learn how).
As a conclusion, the expense ratios of ALFA and GURU are cheap compared with hedge funds, but they are not when comparing their risk-adjusted performance with the benchmark. If I had to make a choice, I would prefer ALFA because it is less volatile.