Hedge Funds as Diversifiers in Institutional Portfolios

Alternative investments come in many forms


In the not too distant past, hedge funds were often considered mysterious, exotic, and elusive investment vehicles that revealed little about their methods of generating return. As the industry has matured and standards of transparency have evolved, that perception seems to have mostly faded.

Meanwhile, other broad tags of disapproval, such as that hedge funds are expensive, glorified indexers – in essence, the opposite of exotic and mysterious – have more recently taken their place. While these extremes reflect some degree of hyperbole, they also reflect legitimate concerns and important considerations for hedge fund investors: avoid investments that are inaccessible and not able to be understood; avoid constructing a hedge fund allocation that is redundant and inefficient in a portfolio. But for most, the reality of hedge fund investing is very much in between: the inaccessibility and mysteriousness of hedge funds is often overstated, and their value as diversifiers is often understated.

  • The growth of the industry has led to an increased number of correlated managers, but it has also led to an increased number of uncorrelated managers. In fact, the single largest area of growth over time has been in the group of managers with the lowest percentage of their returns explained by those of the broad equity market. For a thoughtful, strategic investor, this diversity provides the opportunity to attain portfolio diversification independent of market direction.
  • The long-term, alpha-based case for hedge funds remains strong, despite recent declines in alpha coinciding with unusually adverse market conditions for security selection generally.
  • In addition to significant alpha, hedge funds also offer a diverse array of systematic or market exposures.
  • Hedge funds are not a monolithic entity, nor should they be considered a single and uniform investment class. They are highly diverse both among and within strategies.

Hedge funds continue to present a compelling value proposition to institutional investors as volatility dampeners, absolute return vehicles in certain cases, and as less correlated sources of return. As an investment class they have historically demonstrated the ability to generate superior, risk-adjusted returns to the broad market. Even in periods when hedge funds underperform the broad market they still provide measurable value to a portfolio. After all, if completely confident in the direction of the market, an investor would not need hedge funds or any other source of diversification.

Hedge funds as diversifiers in institutional portfolios
Investors allocate to hedge funds for a wide array of reasons. Some seek equity or fixed income complements or substitutes, others pursue volatility dampeners and sources of less correlated return, while others view hedge funds as a means of access to less liquid or more complex securities. All of these reasons can in some form be captured within a single category – risk. And for many, equities represent the dominant risk in their portfolios. According to the 2012 NACUBO-Commonfund Study of Endowments (NCSE), the average US endowment has 46% of its assets invested in equities, and significantly more on a risk-weighted basis. One of the primary roles of hedge funds for the typical institution, then, is often to provide a source of return that is only modestly correlated with the broad equity market.

Within the public sphere, however, assessments of hedge fund performance tend not to be as nuanced. Commentary like “…the S&P 500 was up 16% in 2012, but hedge funds, private investment vehicles run by what many consider to be top managers, were up only 8% on average…” is often characteristic of financial news commentary. While the particular favour of media attention to hedge funds partially reflects backlash against a perceived element of exclusivity, it is important to note that these characterisations come with embedded assumptions that may be inconsistent with many investors’ goals.

One implicit assumption, for instance, is that the success of a hedge fund – any hedge fund – should be viewed as a standing bet on the direction of the S&P 500 Index. This would presume that a hedge fund manager attempts to have long exposure to the equity market when it is going up and short when it is going down. Another is that hedge fund managers ought to be perpetually ahead of the most widely followed, passive equity market index given their elite and talented status. In both cases, the hedge fund universe is understood narrowly in terms of two monolithic entities: “Hedge Funds,” capital H, capital F, and “The Market,” capital M.

Actual (as opposed to theoretical) hedge fund investing, however, requires thoughtful, careful decision-making with significantly more context around strategies and their risk and return properties.

Correlation and the growth of the hedge fund industry
One charge often directed at hedge funds is that they offer little more than repackaged equity market risk even as sceptics acknowledge that hedge funds have historically delivered robust, uncorrelated returns within certain markets. However, many also assert that the growth of the hedge fund industry has built up industry-wide correlation on a large scale, with alpha increasingly scarce and fleeting, and with inevitably overlapping trades eroding diversification benefit. At a glance, the uptrend of broad hedge fund index correlation with the S&P 500 Index (here, the HFRI Fund-Weighted Composite Index) would seem to support these claims (see Fig.1).

The hedge fund universe has indeed grown several-fold since the “mystique” days of the early 1990s. Quantitative evidence suggests that many funds are largely vehicles for a form of broad market exposure (beta), and on an average basis the typical hedge fund universe constituent seems to have become increasingly correlated with the equity market over time. However, for the prudent hedge fund investor, the correlation properties of the average hedge fund – represented by the HFRI Index – are not necessarily relevant to a thoughtful, thorough selection process.

As the hedge fund universe has grown, competition for capital and the hunt for sources of return among individual managers have created an opportunity for more diversification, not less; in fact, a look within the broad hedge fund universe contradicts the notion that hedge funds are solely driven by the equity markets. For instance, a universe-wide look at correlations reasserts that the risk and return properties of a hedge fund allocation are not simply a function of the broad equity market. In Fig.2 are rolling 12-month R-Squared (a measure of the total share of return explained by the S&P 500 Index) of constituents of the HFRI Fund weighted Composite Index, representing all four major strategy groups – equity hedge, event-driven, macro, and relative value – to the S&P 500 Index.

Within the HFRI Fund Weighted Composite index, as the total number of funds has risen, the number of funds highly correlated with broad equity market has also risen – but so, too, has the number of uncorrelated funds. In fact, partitioning the manager universe on a 12-month rolling basis in terms of R-Squared to the equity market in 10% increments, from December 1992 to December 2012, there was a greater increase in the number of managers with less than 10% of return explained by the S&P 500 Index than any other group (see Fig. 3).

Regardless of the properties of the broad hedge fund index, then, for a very large number of constituent managers, the most dominant investment risk is not the direction of the S&P 500 Index. The assertion that hedge funds have evolved from superior diversifiers into equity market proxies loses some potency based on this measure of correlation. While the growth of hedge funds may be associated with greater correlation with equities at an index level, the ability to achieve diversification at the individual manager level has grown, not contracted.

Alpha generation
The HFRI Equity Hedge universe has the largest constituency of any hedge fund strategy and equity hedge has arguably been the strategy most challenged in the aftermath of the 2008 financial crisis. Fig.4 shows the rolling annualised alpha of the North America constituents of the HFRI Equity Hedge Index versus the S&P 500 Index. Over the long term, results demonstrate that hedge funds have, in fact, produced considerable alpha relative to the equity market.

Furthermore, segmenting the universe into alpha production quartiles, the 50th percentile has averaged nearly 5% annualised alpha while all quartile cohorts have suffered similar rates of decline at a time when broad equity market correlation was rising. Meanwhile, the top quartile has averaged nearly 10% annualised alpha production; even at its recent all-time low it produced 5% of positive alpha. These results highlight the potential difference in alpha derived from manager selection – finding a top performer can have a significant impact on investment results.

An additional step is to substitute an index representing “hedge fund beta” itself for the S&P 500 Index. The Goldman Sachs Absolute Return Tracker ETF (GJRTX) is one of the most recognised hedge fund replicators, and Fig.5 demonstrates that on a single-factor basis the large majority (approximately 84%) of the reporting hedge fund universe has produced positive alpha on a relative basis.

Does the result suggest that the overwhelming majority of hedge funds produces positive alpha, after accounting for market risk, to the rest of the hedge fund universe? That certainly seems unlikely. The strength of this result must be at least partially attributable to familiar database biases, like survivorship among reporting members, along with the replicator itself – designed to produce hedge fund beta and not alpha. The result does, however, strongly suggest that stating that the universes of investable hedge funds are either (1) simply interchangeable, or (2) simply vehicles for market risk, is greatly overstated.

Diversity of beta exposure
In addition to uncorrelated returns and alpha, the hedge fund value for institutional investors is diverse market (beta) exposure. Across the hedge fund universe, specific strategies and areas of specialisation draw managers into different markets using different methods. This results in significant diversity in the statistical properties of historical returns.

To illustrate this concept, Fig.6 and Fig.7 are two sets of multi-factor “beta” profiles for individual managers reporting to the HFRI database. The first chart shows results for equity hedge managers versus common equity factors, the second for relative value and event-driven managers versus credit-related factors. In both cases, the horizontal x-axis shows the “beta” or systematic exposure to one factor and the vertical y-axis shows simultaneous exposure to another.

In the equity hedge example, the first factor is sensitivity to a regional broad market (S&P 500, MSCI Europe, MSCI Asia ex-Japan) applied to managers based on their region of focus. In addition to broad market exposure, managers exhibit different biases to fundamental equity valuation factors. Exposure to such biases in stock selection is often considered a form of systematic risk; in the case of value it makes sense to consider this a “beta” to a regional value factor. In this case, the regional value factors are based on academic researcher Kenneth French’s Fama/French benchmark factor HML.

Particularly notable here is the dispersion of the systematic risk exposures. If managers were in fact best considered a monolithic undifferentiated mass, the plot points would be tightly concentrated in a single area. Instead, they are scattered across the surface of the chart.

The second chart (Fig.7) makes a similar case within the relative value and event-driven universes, showing sensitivity to two forms of systematic risk exposure within corporate credit spreads:

  1. A parallel shift down in spreads across the entire credit ratings spectrum (i.e., credit spreads are tightening to a similar degree in all ratings categories) and,
  2. A “fattening” in spreads (i.e., the spreads of lower-rated credits are tightening more relative to higher-rated credits, suggesting outperformance of riskier credits).

Managers to the right of the vertical axis have a statistical bias towards a broad credit rally, and managers above the horizontal axis towards outperformance in lower-quality credits (plot points in the upper right quadrant suggest positive sensitivity to both factors). While the majority of constituents in the two strategy universes are generally neutral-to-long credit (as opposed to short), and are largely expected to perform better in environments characterised by outperformance of lower-rated credits, the broad scattering of plot points also suggests significant dispersion in the beta risk profiles of the managers. This diversity of exposure is another critical aspect of the diversification benefit that hedge funds can offer institutional portfolios: hedge funds are diverse and come in many forms with respect to their systematic risks.

Endnote to HFRI Relative Value and Event-Driven Universe multi-factor “beta” profile
Credit beta factors used in scatter chart are based on principal components of correlation matrix of 10 option-adjusted corporate credit spreads, with Bank of America Merrill Lynch credit spread indices used as inputs. Loadings are represented below.

  • Parallel shift: spreads narrow across the ratings/quality spectrum.
  • Flattening: spreads on lower-rated credits narrow more than those of higher-rated, potentially with widening among higher-rated. Scenario may be interpreted to represent a migration to higher-risk, higher-yielding credit that is uncorrelated with the parallel shift factor.