Imperfect Replication Beats Single Manager Selection 9/10 Times
Subscribe to Our Blog
One of the commonly raised concerns about Hedge Fund replication strategies is the tracking error to its index. While any replication is bound to be imperfect, it is important to wrestle with the alternative that most investors are choosing today - the significant tracking error that comes from concentrated investments in single managers.
When the two are compared, it's pretty clear that most investors would be better off accepting the model error of a low-cost, tax-efficient replication than investing in high cost, tax-inefficient single manager investments. Working through the numbers, it looks like replication could be about 9x more likely to outperform single manager investments over time.
To get a tangible sense of the trade-off between these choices, we compared the range of outcomes over the last 3 years between:
(1) the actual distribution of the net of fees returns of approximately 3000 single Hedge Fund managers (Source: Preqin)
(2) the estimated range of a Hedge Fund index replication strategy targeting the gross of fees index returns with a roughly 3% error standard deviation and 1% fees.
While the replication is imperfect, the range of its return is much narrower and the median outcome is a few hundred basis points higher. In roughly 70% of the possible head-to-head outcomes of a randomly selected head-to-head comparison, the replication beats the single manager approach.
It can be hard to intuitively picture the assumed tracking error from the above, so the chart below shows an example replication strategy that exhibits a similar type of modeled variance on a 3yr time frame. To align more with shorter-dated evaluation periods, we show the rolling 6m returns. The replication example is not perfect, but it’s a pretty good match. While the quality of replication in the future can’t be measured, what is shown is in the range of what is plausible using today’s technologies.
One of the advantages of modern Hedge Fund replication approaches is that they can be packaged in a much more tax-efficient ETF structure rather than typical LP structures. While every circumstance is different, below we assume that the replication tax structure is 25% based on capital gains and the single manager tax rate is 50% based on marginal income tax rates.
Accounting for the relative tax consequences increases the head-to-head outperformance of the imperfect Hedge Fund replication vs. the single manager LP positions to nearly 85% of the outcomes.
Most investors face the additional hurdle of being able to actually get into the very best funds either because they are closed, difficult to identify, or because the minimums are too high for even relatively well-off accredited investors. Even assuming a third of the funds in the right tail were uninvestable for one of these or other reasons means that by choosing an imperfect hedge fund replication strategy odds are 90%+ of outperforming single manager selection over a time frame of a few years. And that doesn’t consider any of the work required to make single manager investments - diligence meetings, paperwork, ongoing Schedule K-1s, etc.
What it comes down to is that while with a single manager there is the *possibility* of achieving higher returns than with a Hedge Fund replication strategy, the *probability* is that the replication will achieve a better outcome. The combination of single managers having a much wider distribution with much higher fees and taxes results in a distribution that is very wide and skewed lower than a lower cost, more tax-efficient replication structure.
There are few opportunities in investing to raise your probability of a better outcome by roughly 10x. That is the possible opportunity in choosing a low-cost, tax-efficient ETF relative to a concentrated single manager LP investment. Despite this possible risk-reward, only 0.06% of all money invested in Hedge Fund strategies are in low-cost replication ETF structures.