Synthetic Hedge Fund Index Trackers: A New Way to Invest in Hedge Funds

In 2007, most of the big US and European financial institutions introduced their synthetic hedge fund index trackers (STs), which quickly became nicknamed ‘hedge fund clones’.The advent of STs follows years of academic research, which suggests that an important portion of hedge fund’s (HF) returns could be explained by an actively traded long-short portfolio of liquid financial instruments.

What Are Synthetic Hedge Fund Index Trackers?

STs are dynamically managed portfolios of liquid assets, such as futures and exchange-traded funds (also called replicating factors) that aim at minimising the tracking error with a target non-investable HF index. There are three conditions that a good ST must meet:

1. Target a good HF index.
2. Use a powerful tracking model.
3. Apply the model to a proper set of liquid replicating factors.

For any asset class, a perfect index includes the following characteristics:

• Representative of the investment universe.
• Transparency.
• Consistent weighting.
• Reported timely.
• Stable performance over time.
• Investable.

Most non-investable HF indices meet the first four criteria, but they all fail to meet the last two. First, HF net asset values are – to varying extents – based on appraisals that are subject to occasional revisions (some of them may be substantial). Second, some HFs no longer accept funds from new investors.

Non-investable HF indices also suffer from a number of biases that affect their historical track record. The main biases in question include survivorship, selection and backfill biases, and their cumulative upward impact on a HF’s returns history is believed to be between 2-5% per year.

So why not replicate an investable HF index? The reason is that investable HF indices are based on a much thinner set of HFs causing them to significantly under-perform their corresponding investable index. Since it is better to track an index having higher returns, most STs target a non-investable HF index.

A Powerful Tracking Model¹

If a ST wants to be representative of the HF universe, it must follow (track) the target HF index returns as closely as possible. Figure 1 summarises the main methods financial institutions currently use to build their STs.

Figure 1: Main Models Used to Build STs

It is widely accepted that the use of advanced filters is preferable because they have better tracking potential (most of them can achieve an annualised tracking error of 3% or less, which is quite good given the nature of non-investable HF indices).

Selecting A Proper Set of Liquid Replicating Factors

Regardless of how good the tracking model is, it will not perform well if it is applied to an inadequate set of replicating factors. HFs do not create money; they ‘simply’ make investment decisions on the same traditional financial markets most investors have access to. The advantage of HFs largely lies in the fact that their investment policies/mandates are not restricted as much as those of their long-only managers. These fundamental return-generating factors are:

• Inflation.
• The real interest rate.
• The interest rate slope (i.e. the term structure).
• The credit risk premium.
• The equity risk premium.

As a result, any sophisticated model will not be able to effectively track a representative HF index if it is not applied (fed) by liquid financial instruments that cover a wide range of the factors described above.

How Do Synthetic Hedge Fund Index Trackers Compare with their Non-investable Target?

Figure 2 breaks down the return components of STs and their non-investable HF index target. We can see that STs capture the systematic (beta and alternative beta) portions of HF returns, like stock index ETFs capture the passive (beta) portion of equity mutual funds.

Figure 2: Gross return Components of STs and of Non-investable HF Indices

As mentioned previously, non-investable HF indices’ returns cannot be matched (mostly because of biases and non-investability). Since it is difficult to isolate one returns component from the other, the breakdown is approximate, but there is consensus on the fact that, over the long term, STs should generate a gross excess return (over the one month LIBOR rate) between 3% and 4%, which means that their excess returns should lie somewhere around 2.5% and 3.5% net of all fees.

Is such a return sufficiently high to make STs interesting investments? Yes, because if we consider that most STs exhibit an annual volatility of 5%, their Sharpe ratio is approximately 0.5, which is about twice as much as the fundamental asset classes’ (bonds, real estate and equity) long-term Sharpe ratios². Also, as we will see in the next section, STs compare well relative to the investable HF indices.

For investors who are more concerned about capital protection, studies show that most STs offer a significantly smaller downside risk than equity indices, even after controlling volatility (i.e. after leveraging STs in order to raise their volatility to the level of the equity indices).

How Do Synthetic Hedge Fund Index Trackers Compare with Investable Hedge Fund Indices?

Table 1 summarises the fundamental differences between non-investable hedge fund indices, their corresponding STs and their investable versions.

As can be seen, relative to STs, investable hedge funds indices are less broad-based⁷ (which may be seen as a disadvantage) but they earn some additional return (liquidity risk premium and access to some exotic asset classes). If these advantages exceed the broad-base handicap, we should observe that over time the probability of STs out-performing investable indices should diminish with the investment horizon length or at least stay constant.

Figure 3 contains some results we obtained in another study we recently published⁸. They are quite unambiguous: they suggest high probabilities of the STs out-performing their corresponding investable indices for investment horizons longer than 12 months. Also, of the three models we used to estimate these probabilities, all of them indicate that the probability of STs out-performing investables uniformly converges towards 1.0 as the investment horizon is lengthened.

These results tend to show that the investable indices’ thinner-base disadvantage weighs more than their advantage of being able to capture some liquidity risk premium and additional diversification gains. They also suggest that an investor who considers a liquid exposure to HFs is better off investing in a SHFI rather than in an investable hedge fund index.

Figure 3: The probability that a ST out-performs its corresponding investable HF indices (IHFI) for various investment horizons


Source: Based on Innocap Investment Management’s SALTO MSCI HF Composite Synthetic Tracker versus the composite MSCI Hedge Invest Index

How Do Synthetic Hedge Fund Index Trackers Compare with Funds of Hedge Funds?

When first considering investing in HFs, most institutional investors look at traditional (rather than managed account-based) funds of HFs (FoHF) because the FoHF managers are assumed to be in a better position to identify the best performing HFs. The main advantage of investing in a FoHF rather than STs is that the FoHF, as a portfolio of genuine HFs, gives access to the (potential) alpha generation of the selected HFs. Hence, good FoHFs may generate alpha.

The disadvantages include the following factors:

• Not all FoHFs have above-average skill at identifying the best HFs. Hence not only can they generate no alpha, but they may also generate a negative alpha. It is clear that STs always generate a 0 alpha.
• FoHFs charge an additional fee layer to the fees paid to the HF managers. STs do not pay fees to HF managers and they usually charge smaller management fees (typically from 25 to 50 bp/y less) than FoHFs.
• An inexperienced investor unable to select the better performing HFs may not be in a position to identify the better performing FoHFs either.
• FoHFs do not offer daily liquidity (this is true for all traditional FoHFs and for most managed account-based FoHF).
• Traditional FoHFs often impose lock-up periods, SHFIs do not.
• Traditional FoHFs are considered opaque investments (because the underlying HFs are not managed accounts) so they are less regulatory-friendly.

In another study, we found that STs’ risk-return profile is very similar to that of the median traditional FoHF once we adjust for sampling biases and liquidity risk. Hence, an inexperienced investor roughly has a 50% of picking a traditional FoHF whose risk-return profile lies above that of typical STs. As a result, it is more and more widely accepted that STs constitute a better HF investment than traditional FoHFs, especially for first-time investors in the HF universe.

Other Features of Synthetic Hedge Fund Index Trackers

STs also feature two additional interesting characteristics that cannot be matched by most HF indices and most FoHFs. First, STs are easily scalable – if the investor is willing to take more risk, the ST can be leveraged to the desired volatility. Very few HF indices and FoHFs offer this possibility, unless the investor does it by itself.

Second, some financial institutions offer short STs. The short versions of a ST can be very interesting for institutions that have already invested in HFs and want to do some hedging on this asset class. This is far more advantageous than redeeming a HF when the investor seeks to temporarily reduce its exposure to HFs, because good HFs may not be able to accept the returning investor’s money when the time comes to raise the HF exposure again (i.e. another investor takes the available capacity following the would-be-returning investor’s redemption).

Conclusion

Synthetic hedge fund index trackers are to HFs what exhange traded funds (ETFs) are to mutual funds: a liquid, low-cost and transparent way to expose a portfolio to the asset class. STs that target a good HF index and use a sophisticated tracking model applied to a wide range of liquid and transparent financial instruments should exhibit an interesting risk-return profile, particularly investors who are liquidity risk averse.

STs currently constitute a small portion of all the HF assets under management because they have been introduced quite recently. Nevertheless, they now constitute a key element in the alternative assets offering of the biggest financial institutions. We should see their assets under management (AUM) grow significantly over the coming years, as using them in a core satellite approach to HF investing⁹ gains more traction among big institutional investors.