A vexing issue with hedge funds is the lack of persistence of returns.  Last year’s winners generally are no more likely to be this year’s. Consequently, while quantitative screening tools help investors to identify strong historical performance, they are of little use in determining which funds will perform better going forward.

Beachhead’s proprietary research demonstrates that, while alpha overall does not persist, “tactical” alpha does.  Broadly speaking, hedge fund alpha can be broken down into two sources:  position alpha and tactical alpha.  Position alpha represents outperformance due to security selection, illiquidity and other factors, such as optionality.  Tactical alpha represents medium-term relative value trades across asset classes – such as investing in credit post-crisis or shifting back to developed markets equities in 2012-13.[1]  Both position alpha and tactical alpha have a low correlation to equities over time, and hence are valuable diversifiers.

In order to study the persistence of tactical alpha, we examined the performance of 679 funds over the seven year period from 2007 to 2013.  The key conclusions are as follows:

  • Managers that generated high tactical alpha (top quintile) in a given year outperformed the overall pool by 246 bps per annum on average in the following year.
  • Outperformance was driven by greater alpha, not higher leverage:  alpha to the overall pool was 275 bps per annum with comparable beta.
  • High tactical alpha funds had a much higher Sharpe ratio (0.84 vs. 0.59) over the full time period.
  • As expected, high tactical alpha funds performed materially better during the crisis, with aggregate outperformance over 2008-09 of almost 17%.
  • Relative value funds show the most consistent outperformance due to tactical alpha: although representing only 18% of all funds, they accounted for 30% of top quintile funds.

These results lead to several important conclusions about hedge funds.  First, unlike position alpha, tactical alpha tends to persist.  This makes sense:  high tactical alpha suggests the manager has a good understanding of the current market and, since tactical alpha positions tend to shift relatively slowly, this understanding provides a valuable edge in the near future.

Second, hedge fund managers can add material value through asset allocation.  One potential implication is that hedge funds with broader mandates may have greater flexibility to capitalize on alpha opportunities among markets.  It is noteworthy that equity long/short funds added less value through tactical alpha, although it is unclear if this is due to a narrower mandate or lack of skill.[2]

Consequently, in order to maximize returns, allocators to hedge funds may be better served by focusing on tactical alpha than position alpha.  First, for many hedge funds, position alpha is rarely greater than all in fees; therefore, most alpha generation after fees tends to be tactical alpha.  Second, since tactical alpha persists, a portfolio of high tactical alpha managers is more likely to outperform going forward.  Finally, tactical alpha exposures can provide important, and highly valuable, insights into how managers collectively view the relative attractiveness of markets.

Please contact Andrew Beer or Mathias Mamou-Mani at Beachhead Capital for additional information.

 

[1] Tactical alpha is distinguished from near term net exposure management, which generally does not produce consistent alpha.

[2] Interestingly, macro funds tended to add little value through tactical alpha, although this may be due to a difference in investment time horizon.

EmailPrintLinkedInEvernoteShare/Bookmark

By one estimate, the hedge fund industry managed $2.6 trillion in capital at year-end 2013.  Many expect growth to accelerate over the coming five years as institutional investors like US pension funds seek alternatives to investing in low yielding fixed income assets.  Deutsche Bank, for example, predicts that industry assets will grow by another $400 billion this year alone.

This growth comes in the face of widespread discontent about the cost of investing in hedge funds.  In the pre-crisis period, when hedge funds routinely outperformed traditional assets, the cost of investing was largely overlooked.  In recent years, though, as the average hedge fund delivered single digit returns, high fees increasingly have come under scrutiny.  To put the issue in context, investors paid approximately $95 billion in fees in 2013, or 44% of what their investors took home.  By our estimate, this figure is twice what it should be:  in other words, investors overpaid by a staggering $47 billion.  The same conclusion can be drawn for each year since the financial crisis.

Why is this?  In the post-crisis period, institutional investors and their consultants partially brought down the cost of investing by “dis-intermediating” funds of hedge funds – in essence, cutting out the middleman.  However, this represented only a small percentage of overall fees. Driven in part by risk aversion (you don’t get fired for hiring IBM), those same investors instead steered capital directly to the largest hedge funds – the $10 billion plus firms you read about in the press.  By some estimates, 100% of net capital inflows post-crisis went to the top 100 managers.  The big have gotten bigger and the small have struggled.

Why is this problematic?  Hedge funds charge two types of fees:  a management fee and a performance fee (typically 20% of profits).  When hedge funds were small, relatively high management fees (1.6% on average) were necessary to cover costs and to enable a manager to expand the research team, hire a head of operations, etc.  Performance fees were geared to better align the interests of investors and the manager, something that was sorely lacking in the mutual fund business.

As the industry matured, the fee structure didn’t.  A typical manager with $5 billion in assets under management earns $80 million in fees – most of it pure profit.  Incremental management fees don’t add to research coverage or a more robust investment process; they simply increase the manager’s take home pay.  Further, as funds grow, managers take less risk to preserve the value of the business.  Performance fees then no longer provide incentives for stellar returns, but rather become an annual tax on investors.  The absence of a hurdle rate on incentive fees meant that the 32% rise in the S&P 500 index last year resulted in 200bps of overpayment across the industry. When you overpay for beta, it is a dollar for dollar reduction in alpha: it’s why fee reduction is the purest form of alpha.[1]  Skill should be rewarded handsomely, not luck.

In the rest of the asset management industry, large investors get fee concessions; it’s why retail investors have the odds stacked against them.  However, when institutional investors fall over themselves to invest in a name brand fund, the manager has little incentive to cut fees.  Counterintuitively, institutionalization has failed to result in real cost savings for most large investors.  And as we know, the cost of investing is one of the most important factors in long-term investment performance.

There are four potential solutions.  As noted recently by the head of hedge fund allocations for a large pension fund manager, sophisticated investors will migrate to a combination of hedge funds and comparable, but far less expensive, strategies like dynamic and alternative betas, in which investors break down the drivers of hedge fund returns and invest in them directly.[2]  What you don’t get in manager talent, you save in lower fees.

The second option is for investors to band together to demand a material reduction in fees.  A group of investors that constitute half the assets of a $5 billion fund has a much better chance to bring down fees than if each $50 million investor negotiates separately.  The two areas of focus should be to bring down management fees (as a percentage of assets under management) as assets increase and to insist that managers only earn performance fees above a specified benchmark.

A third possibility is to steer capital to smaller hedge funds.  Interestingly, the have and have-not bifurcation of the industry may be a catalyst for reform.  Many smaller hedge funds, facing years of difficulty in raising capital, are much more willing to drop fees today.  An institutional investor who can invest $50 million with a $200 million manager can pretty much set its own terms.  Importantly, since smaller managers generally outperform larger ones over time[3], investors could see a dual benefit of a pick up in performance and lower fees.

A fourth area garnering attention recently is the alternative mutual fund space, where hedge fund strategies are managed in registered investment funds.  These products are in their infancy with short track records and an unproven business model.  While some products are offered at roughly half the cost of investing in hedge funds, the fee structure still is double or more that of traditional mutual funds.  The elephant in the room is whether the performance of these products will match those of the hedge funds they emulate.  Most are marketed off track records of hedge funds, not mutual funds, and the mutual fund structure, by design, has many more constraints.  It’s a bit like asking a champion boxer to fight with his shoelaces tied.  The irony is that the appeal of hedge funds twenty years ago was to take talented investors and free them of the (many) constraints of managing assets in a mutual fund structure; we now appear to have come full circle.[4]

Pension funds, other institutional investors and their advisors need to take a hard look at this issue. As one pension fund trustee reportedly observed, at some point you stop looking at fees as a percentage of assets and focus on the dollars involved.  A pension fund with $4 billion invested in hedge funds that overpays by 2% per annum throws away $80 million a year.  Over a decade that’s $800 million.  To paraphrase the famous quote, a million here and a million there and soon you’re talking about real money.

 

[1] See Fee Reduction and Alpha Generation

[2] See Hedge Fund Replication: A Practitioner’s Scorecard and Two Unanswered Questions About Alternative Betas

[3] See Performance of Emerging Equity Long/Short Hedge Fund Managers

[4] See Performance Drag of Alternative Multi-Manager Mutual Funds

By Andrew D. Beer, Chief Executive Officer of Beachhead Capital and Michael O. Weinberg, Adjunct Professor at Columbia University 

Well-documented advantages of alternative mutual funds include daily liquidity, lower all-in fees, greater regulatory oversight, lower minimum investment requirements and the absence of partnership K1s. These features make alternative mutual funds a viable investment for defined contribution plans and retail investors, an untapped multi-trillion dollar market for hedge fund managers.

The opportunity for fund of funds managers is clear. Post-crisis, funds of hedge funds faced a sharp decline in profitability due to a combination of disintermediation, declining fees and rising costs (e.g. customization). Gone are the days of managing a highly profitable co-mingled fund of hedge funds where each incremental dollar of revenue drops to the bottom line. Defined benefit pension plans – long-time investors in funds of funds – are in steady decline; by contrast, defined contribution plans are growing rapidly and current exposure to alternatives is de minimus. Alternative multi-manager mutual funds (AMMFs) could represent a new dawn for funds of funds:  co-mingled, highly scalable vehicles with strong potential investor demand.

However, mutual funds are far less flexible than hedge funds: all registered funds have structural limitations on leverage, shorting, illiquid assets, concentration and other criteria. This contradicts a core tenet of the hedge fund industry:  that talented managers perform better with fewer constraints. In fact, many early hedge fund managers were former mutual fund managers in search of a less constrained investment vehicle. By definition, structural constraints almost certainly will result in a performance differential over time:  the question is, by how much?  This paper looks at the evidence to date, some of the underlying causes, and raises some pertinent due diligence questions for potential investors.

1. Performance Drag: Evidence to Date

There is a limited amount of live data on AMMF performance. Of the thirteen funds that we currently track, only three have track records that extend back more than a year or two. Only five of thirteen have full year track records for 2012, while eight have full year track records for 2013. Any conclusions below should be taken in this context.

With that caveat, the chart below shows the full year performance of those five and eight funds, respectively. The magnitude of the performance drag is noteworthy:  the average AMMF underperformed the HFRI Fund of Hedge Funds index by 311 bps in 2012 and 420 bps in 2013, net of fees (the return figures are for institutional share classes only). The average AMMF also underperformed an index of liquid hedge funds (HFRX Global Investable Index) by approximately 200 bps in each year.

Full Year Performance Comparison 2012-13 

This is particularly surprising given that all-in fees in AMMFs are 150-250 bps lower than those in fund of fund funds; all things being equal, this should lead to higher performance. While the average AMMF charges slightly more than 200 bps per annum with no incentive fee,[1] the typical fund of hedge funds has all-in fees of 3.5% to 4.5% (75-150 bps per annum on top of underlying hedge fees of 1.6% and 20%, on average). On a fee equivalent basis, the performance differential rises to 448 bps in 2012 and 681 bps in 2013.

Estimated Fee Equivalent Performance Drag 2012-13

2. What Explains the Performance Drag?

When the HFRX GlobalInvestable Index was launched over a decade ago, performance drag initially was over 400 bps per annum, primarily due to adverse selection bias. That differential has come down over time, but still is persistently 100-200 bps per annum. Other liquid hedge fund strategies – investable hedge fund indices, managed account platforms, 130/30 funds, and UCITS products – have each underperformed hedge fund counterparts by around 200 bps per annum. This supports the general rule that liquid alternative strategies have a persistent long term performance drag relative to actual hedge funds.  (See The Performance Drag of Liquid Hedge Fund Strategies at www.beachheadcapital.com).

In the AMMF space, performance drag is likely to arise in two forms. First, following the HFRX example, hedge fund managers most eager to manage sub-accounts at reduced fees may be of a lesser quality. This may explain in part why the Hatteras Alpha Hedged Strategies Fund, the only fund with a ten year track record, underperformed the HFRIFOF since inception by 134 bps on an annualized basis since inception (including negative alpha of 1.8% given somewhat higher beta) and suffered much more pronounced drawdowns during the crisis (down 31.6% in 2008).[2]  On the surface, the current pool of subadvisors appears to be much more credible than the early participants in the investable hedge fund indices.

A more serious issue, then, is that high quality managers simply may not be able to deliver comparable returns within the constraints of the ’40 Act structure.[3]  While it is impossible to make an apples-to-apples comparison,[4] one indirect method is to compare the performance of AMMFs to a portfolio of hedge funds managed by the same subadvisors. Here we find that AMMFs have on average underperformed an equally weighted basket of the “flagship” hedge funds of the subadvisors by over 200 basis points.[5]  On a pre-fee basis, the differential rises to over 500 basis points.[6]

Why is this?  In addition to investment constraints, hedge fund managers have a strong incentive to avoid cannibalizing their core businesses; therefore, given the general absence of incentive fees, the most attractive or capacity constrained trades are likely to wind up in higher fee vehicles.[7]  Irrespective, given the importance of the hedge fund managers’ reputations in marketing AMMFs, it is reasonable for investors to require AMMF managers to break down performance by subaccount relative to the most comparable hedge fund managed by the same subadvisors.

A final and important distinction is that AMMF managers directly pay the subadvisors:  unlike in a fund of hedge funds, each dollar of subadvisory fees comes directly out of the pocket of the AMMF manager. AMMF managers therefore have a strong incentive to select subadvisors who will work for 1% or less. A fair question for any AMMF sponsor is how many subadvisors/strategies were rejected due to high fee expectations, and why this does not lead to adverse selection.

Conclusion

In conclusion, while it’s still very early in the game, our analysis indicates that AMMFs are likely to be subject to a persistent performance drag over time. Based on the more robust data pool from other liquid alternatives, our expectation is that the performance drag should be around 200 bps per annum, net of fees. Fortunately, since return information is readily available on ’40 Act funds, it will be much easier to make ongoing comparisons, and we look forward to updating this analysis in the coming year or two.

This does not mean that AMMFs are necessarily inferior to hedge funds of funds. The structural advantages are very real, and will be worth more than 200 bps per annum for many investors. Further, investors who previously have been precluded from investing in hedge funds may find AMMFs to be a valuable diversifier.

As of early 2014, the evidence suggests that the performance drag will be higher than many investors realize and that the (real) advantages of greater liquidity, lower minimums and reporting simplicity are likely to come at the cost of diminished returns.

 

[1] Most funds currently include fee waivers and/or rebates to bring down all-in fees. The average rebate among 13 funds included in the study currently is 48 bps per annum. Note that fee rebates generally have a finite life, so a relevant due diligence question is the extent to which fee rebates will continue going forward.

[2] The institutional share class of the Hatteras fund was not introduced until late 2011, whereas the retail share class was introduced in 2002. The retail share class is referenced here.

[3] This is the principal reason the SEC does not permit prospectuses to show the hedge fund track records of the subadvisors. In fact, only one of the AMMFs (Balter) requires the subadvisors to run portfolios substantially identical to their hedge funds and therefore is permitted to disclose the track records of the subadvisors’ hedge funds in the offering documents.

[4] Due to a lack of transparency of manager weights in AMMFs and strategy overlap between AMMF sub-advised accounts and predecessor hedge funds, among other issues.

[5] More precisely, of the eight funds with full year performance for 2013, we have return data for the flagship hedge funds on at least 70% of the managers of four of them. Those AMMFs underperformed an equally weighted portfolio of the available hedge funds by an average of 314 bps, net of fees. Assuming that the AMMF managers are paid 100 bps of the management fee, the fee equivalent differential is 214 bps on average.

[6] At a conference in 2013, one prominent multi-strategy hedge fund firm analyzed the expected performance differential in their flagship fund and canceled plans to create a ’40 Act fund after determining that they would suffer an 800 bps drag.

[7] Note that one possible window into this may be which managers advise assets of wholly-owned offshore subsidiaries. In general, offshore subsidiaries can pay management and even incentive fees to managers without disclosure at the fund level. This is a relevant due diligence question for any funds with offshore subsidiaries.

Alternative betas, or risk premia, are established investment strategies that are simple enough to automate with a computer but too complicated for most investors to implement directly. For instance, you can program a computer to buy “value” stocks and short “growth” stocks, but few investors choose to do this on their own. The same argument can be made for merger arbitrage, currency carry trades, momentum, trend following, commodity roll trades and other common trading strategies employed by hedge funds.

The advertised appeal of alternative beta products is that they have a low correlation to traditional assets and have a high-expected return – the practical definition of a valuable diversifier. With widespread pressure to bring down the cost of investing, many investors are considering whether to invest directly in alternative betas to avoid high hedge fund fees and improve liquidity. Having examined a broad range of these products, we conclude that there are two (big) unanswered questions. Each arguably undercuts the diversification thesis.

1. What are expected returns?

There is a paradox in the alternative beta diversification thesis. Expected returns for alternative risk premia are supposed to be high since most investors do not or cannot invest in them directly. However, the proliferation of products should lead to capital inflows and hence drive down returns over time.

This is not a small issue. Take benchmark equity strategies, where we have two decades of data since they were “discovered.”  In 1993, Fama observed that buying value and shorting growth delivered substantial excess returns over 1963-1990. During that period, the Sharpe ratio was 0.57, somewhat higher than the long term Sharpe ratio of the equity markets. Similarly, Carhart showed in 1997 that momentum over the same period offered similarly compelling returns – an even higher Sharpe ratio of 0.85.

These strategies did not remain undiscovered for long. Numerous firms soon offered products and funds specifically to provide access to these strategies. The results were entirely predictable. As capital flowed in, returns declined. In the past ten years, the Sharpe ratios of value and momentum have been 0.15 and 0.02, respectively.

A related issue is that few products actually have live track records. Those that do were often launched in the past several years. Most products are marketed as indices, for which reporting standards are very low:  simply labeling a pro forma track record an “index” sidesteps most of the onerous disclosures that apply to investment products. As an investor, you often have to dig to learn when the index was formally launched; anything prior to this should be discounted for back-fill bias. Consequently, the pro forma returns, or a portfolio of them, will generally be overstated.

The obvious risk of these strategies is that inflows into a capacity constrained strategy will drive down returns to the point where they are no longer an effective source of diversification. By analogy, as discussed previously in a note, CTA returns have been low for almost a decade, and the most likely culprit is the influx of capital.

2. Do They Reduce Idiosyncratic Risk?

“Beta” strategies generally reduce idiosyncratic risk. When you invest in the S&P 500, you don’t need to worry about whether a manager bets wrong and goes off the rails. The term “beta” also implies something that can be clearly defined. In this context, the moniker “alternative beta” is something of a misnomer.

Modeling Risk:

Alternative beta products substitute “manager” risk with “model” risk. Due to the complexity of setting numerous parameters, one firm’s alternative beta will look very different from another’s. We addressed this in the merger arbitrage space and found very different results among providers. The implication for investors is that alternative beta products are more akin to low cost, systematic single strategy funds rather than a beta product per se.

Whereas some firms offer products that have been optimized, others are offered piecemeal. Barclays reportedly offers close to one thousand discrete products. For the former, investors need to evaluate the quality of a complicated optimization; in the latter, the range of products is overwhelmingly broad. In either case, the products fail to provide the clarity and simplicity that most investors associate with the term “beta.”

Implementation Risk:

Due to their complexity, alternative beta products typically are offered in swap form. There are several material issues with this. Liquidity for OTC products can be contingent upon stable markets: a standard ISDA agreement has sixteen pages covering market dislocations. The ability to execute is dependent upon getting the counterparty to repurchase the contract which, as many discovered during the crisis, may be difficult during a chaotic market. Buyers bear the credit risk of the counterparty that may or may not be priced into the product. All in pricing can be opaque at best, especially when it is dependent upon execution. Trading and transactions costs can be very high, such as for the execution of roll trades outside of specified windows.

Conclusion

Based on our review, there is a fundamental disconnect in how the products are positioned and the reality of investing in them. While the headline cost of investing may be low, overall pricing is much more opaque. As noted, liquidity is more constrained than advertised.

However, the biggest issue – and the most difficult to answer – is whether alternative betas are subject to arbitrage over time. No one knows where the tipping point is, and it’s nearly impossible to gauge contemporaneously. Because of this, product providers focus on historical returns, which are easy to calculate but unlikely to reflect returns going forward. As Kahneman observed in Thinking Fast and Slow, when asked a difficult question, human beings have a tendency to pose themselves an easier question and answer it instead. Well said.

Introduction

There is a great deal of confusion about how to measure the success of hedge fund replication – in a sense, to answer the question, “Has hedge fund replication worked?”  In this note, we provide a candid assessment of the successes and failures of the space and introduce a framework of five criteria – a “scorecard” – as a guide for potential investors.

On the one hand, replication products overall have fulfilled the original promise of delivering “hedge fund returns” but with much lower all-in fees and daily liquidity.  Specifically, over the past five plus years an array of replication products has delivered returns comparable to funds of hedge funds and, more importantly, outperformed managed account platforms, UCITS funds and investable hedge fund indices.

On the other hand, potential investors often are put off by the complexity and opacity of many such products – especially those offered by investment banks.  We attribute this frustration to unrealistic expectations, set by the banks themselves, that the strategy should be as simple and predictable as “investing in the S&P 500 index” – that is, a default allocation that requires minimal due diligence with de minimus tracking error and no idiosyncratic (manager) risk.  As discussed in detail below, this was misguided and has hindered more widespread adoption.

In the chart below, we offer a scorecard on standard hedge fund replication products along five criteria: performance, liquidity, transparency, fee reduction and “index-like” alternative:

Chart 1

 “Replication” Defined

Note:  For compliance reasons, we do not include the performance of Beachhead products since this would limit the distribution of this paper to a small subset of investors for whom it might be valuable.  Furthermore, at Beachhead we have migrated away from the “standard” approach described herein in an effort to capture both hedge fund alpha and beta through fee disintermediation and other more sophisticated strategies.

Replication is a broad term.  For our purposes, hedge fund replication is defined as a factor-based approach to delivering the returns of a pool of hedge funds.   The pool might be an index or an actual portfolio.[1]  It’s based on the (now proven) concept that hedge funds derive the majority of returns from market forces, that exposures to different markets change over time, but that these shifts are slow enough that a “backward looking” model can keep up.  Therefore, a dynamically–adjusted portfolio of market betas (equities, bonds, currencies, etc.) can do a very good job of delivering returns that look a lot like those of the pool of actual funds.

Chart 2

When discussing the replication “universe” below, we focus only on funds and indices with live performance; pro forma index returns are subject to backfill bias.  As noted, it is often challenging to try to make “fee equivalent” comparisons across strategies.  Consequently, we specify when likely fee adjustments would impact the conclusions.

Importantly, we differentiate hedge fund replication from rules-based trading strategies, which seek to capture “alternative betas” like momentum, currency carry, merger arbitrage, etc.   These strategies can be an effective way to get exposure to a particular strategy at a lower cost and with greater transparency.  However, alternative betas represent a small portion of overall hedge fund returns.  Therefore, while the rules-based approach can provide additional liquidity or reduce fees at the margin, factor-based replication is necessary to approximate the returns of a diversified portfolio or index.

Evaluating the Five Criteria of Hedge Fund Replication

Performance:  Good Relative, Low Absolute, Returns

The principle criticism of replication products is that returns over the past five years have been “mediocre.” As shown below, the primary cause was the decline in hedge fund returns, not the failure of replication per se.

Remember that most replication products were designed to match the performance of funds of hedge funds, net of fees, at a time when trailing hedge fund returns had been exceptionally high on a risk-adjusted basis.  As of mid 2008 funds of funds had returned 8% per annum over the preceding five years – equal to equities and much higher than bonds.  It was taken as a given that performance would remain strong.  Therefore, the products were designed to offer comparable net returns but with superior liquidity and transparency – which were scarce at the time.  In the subsequent five years, however, hedge funds significantly underperformed expectations, especially relative to stocks and bonds.  As shown below, this explains virtually all of the “mediocrity.”

Chart 3

Interestingly, replication products had much lower drawdowns during the financial crisis than hedge funds and funds of funds.  As with funds of funds, the replication strategies modestly underperformed most direct portfolios since the crisis, as represented below by the HFRI Fund Weighted Composite.

Chart 4

Given the superior liquidity of replication, a fairer comparison is to judge the performance of replication strategies against liquid “alternatives” to hedge funds.[2]  Replication had a similar drawdown during the crisis, but has outperformed recently.

Chart 5

The results above should provide comfort to investors that, as hedge fund returns improve over time, so too should replication results.

Liquidity:  Daily, No Gating Risk

All replication products provide daily liquidity and have no gating or suspension risk.  In a market dislocation, investors should be able to efficiently cut exposure to replication products – a valuable benefit that is not captured in historical returns.

Arguably, though, the “value” of liquidity has declined since the crisis.  Many hedge funds now provide investors with shorter redemption cycles and there has been a marked expansion of more liquid alternatives, like UCITS funds and managed account platforms.  However, overall there is little doubt that replication products can provide far more reliable liquidity than direct hedge funds and most other liquid hedge fund alternatives.

Transparency:  Available, But Often Difficult to Interpret

Position level transparency enables investors to monitor current holdings and provides a window into exposures as a risk management tool.  Unfortunately, while many replication products technically offer transparency, the information value is limited.

For example, the top holdings on Bloomberg for the Goldman Sachs Absolute Return Tracker are Treasurys and repurchase agreements; market exposure instead is held through derivatives that are “off balance sheet.”  As an investor, this raises a series of questions, such as who are the counterparties, what are the actual underlying costs vs. investing in straightforward market instruments, and whether there are conflict of interest (self-dealing) issues.  In a similar vein, while the IndexIQ ETF invests mostly in other ETFs, its holdings include a variety of overlapping funds, presumably to meet mutual fund diversification and position limit requirements. Many of those fixed income ETFs in turn invest through derivatives and have the same issues related to underlying counterparty risks, all-in costs and potential drag on performance.

These portfolios are typical in the space:  they place an unnecessarily high due diligence burden on investors, complicate portfolio level analyses and make it more difficult to monitor results over time.

Fees:  Lower (Headline) Fees, But Not Better Performance

The fees for replication mutual funds are in the range of 100 bps to 200 bps – a significant reduction relative to hedge funds – and vary considerably between the retail and institutional markets.  In general, replication fees are 50% or less than hedge fund fees, and lower relative to indirect investors (through funds of funds, platforms, etc.).  Lower fees, however, have not translated into a commensurate increase in performance.  Here’s why:

As noted above, almost every established hedge fund replication product is designed to target the net of fee performance of the relevant index or portfolio of funds.  Before the crisis, hedge fund fees were roughly 30-40% of gross returns; post-crisis this figure was much higher.  Consequently, as industry returns have declined, this assumption has been called into question.

Going forward, a better approach may be to target the pre-fee, or gross, returns of the hedge fund portfolio.  Part of the problem is that investors often unnecessarily pay away a substantial portion of alpha by overpaying for beta.  Replication strategies, if properly designed and implemented, should be able to recapture a portion of these lost fees and help investors to better evaluate whether a fee structure is fair and reasonable.

“Index-like” Alternative:  Cut Some, But Added Other, Idiosyncratic Risk

The final value proposition of replication was to provide an “index-like” alternative to investing in individual hedge funds.  “Index-like” implies several attributes:  investable, liquid, relatively low cost and a significant reduction in idiosyncratic risk.  Investing should be relatively simple and efficient and hence require substantially less due diligence.  In 2007-08, the analogy most often used was that of the S&P 500 – how investors could use replication as their “default” allocation and concentrate due diligence and monitoring resources on individual managers.

As noted, this comparison was misguided for a few reasons.  At a basic level, there was not and is not a widely accepted definition of the “hedge fund industry”; consequently, different products track different populations of funds.  Some track indices with well-established data biases; others are constructed opaquely from internal databases (e.g. prime brokerage clients) and raise conflict of interest questions.  Some view index returns as, by definition, mediocre and hence unattractive in an industry where allocators constantly seek to identify outperformers.

Second, given the “approximation” approach of replication models, performance by product will vary depending on choice of factors, window length, investment vehicles, fees and a variety of other constraints.  What this should translate to in practice is that while replication models can successfully reduce hedge fund manager risk (e.g., Madoff/fraud and/or headline risk), investors still need to carefully evaluate any replication product – in a sense, the idiosyncratic risk of one replication product/provider versus another.

In reality, potential investors should consider replication to be a low cost investment strategy and not an index.[3]  As with any strategy that relies on a quantitative model, there is a great deal of human judgment that goes into building the model and interpreting the results, and this should not deter investors from seeking the other valuable benefits that the strategy can offer.

Conclusions:  What to Expect Going Forward?

The most important thing about hedge fund replication is that it adds another (and very powerful) tool to the arsenal of investors.  Hedge fund replication proves that the primary drivers of hedge fund returns can be identified and that these exposures are stable enough to foster a reliable investment strategy.  The ability to deconstruct hedge fund returns enables investors to make more informed decisions about the merits of different strategies and to better identify where true alpha resides – and, hence, isolate when it’s worthwhile to sacrifice liquidity and pay high fees for portfolio diversification.

The hedge fund industry overall has undergone a wrenching change over the past six or seven years.  The composition of the investor base is far more institutional today.  The entire fund of funds industry effectively has restructured, with perhaps 80% of pre-crisis funds now closed, and many of its traditional investors now invest directly.  More widespread and comprehensive due diligence practices have reduced the risk of individual manager fraud or blow-up risk.  Investors are much more sensitive to gating and suspension issues, and this shows up in better redemption terms and much more liquid underlying portfolios.  In addition, many more liquid investment vehicles – whether UCITS funds, managed account platforms or now alternative mutual funds – are available today.  Disappointing returns have sharpened the focus on hedge fund fees.

In this context, several leading investors now predict that replication-based products are the “next wave” for pension plans and other fee-sensitive investors. Many highly sophisticated investors already incorporate replication strategies into their investment portfolios and 2012-13 saw a marked increase in interest among leading consultants and institutional investors.

In conclusion, though, we assert that “standard” replication products offer an effective yet only partial solution.  Much of the real promise, we believe, lies in a “customized” approach where replication methodologies are adapted to the specific, stringent and changing demands of investors.  One promising approach is to target pre-fee returns and hence improve performance through a form of “fee disintermediation.”  For some portfolios, this can improve performance by 200 bps or more annually.  Another solution is to replicate the return profile of actual “high alpha” portfolios – those with a higher returns, concentration and turnover.  There is strong evidence that even 300-400 bps of persistent outperformance can be “captured” by a carefully designed and implemented replication strategy.  These and other enhancements lay the groundwork for the next generation of replication products and should help sophisticated investors to achieve their long-term goals of maximizing net of fee performance while minimizing risk.



[1] This is true as well for alternative beta strategies, which essentially are single strategy funds.  See Merger Arbitrage Replication:  Does it Work?

[2] The composite consists of investable hedge fund indices, UCITS products and managed account platforms sponsored by Lyxor, RBC, HFR, Alix Capital, and Dow Jones/Credit Suisse (discontinued in May 2013). Note that we only use live data and that there may be differences in reported numbers and actual net of fee returns realized by investors.

[3] Replication generally does not work for individual funds because the exposures of individual funds are generally less stable, and hence less predictable, than those of a portfolio.

 

“After the wheel, God’s greatest invention was the carry.”
Private Equity Titan

In this note, we examine the relationship between the hedge fund fee structure and how it impacts alpha.

In the early days of the industry, higher management fees were designed to cover costs of a deep and rigorous research and investment process; performance fees were meant to reward the manager for alpha generation.  The standard 2/20 fee structure made sense when hedge funds were smaller and either truly “hedged” –offsetting long and short positions and hence little market exposure – or focused on markets like commodities where beta alternatives were not obvious.

Over the past decade, several changes in the industry have drawn attention to the issue of whether the standard hedge fund fee structure is equitable.  Today, a good portion of the industry – event driven, equity long/short  – has consistent and identifiable exposure to equity market beta; likewise, as we’ve gained a more comprehensive understanding of hedge fund performance, it has become clear that more diverse forms of beta explain the majority of returns.  This raises the question of whether investors are overpaying for sources of return that can be obtained more cheaply and efficiently elsewhere. Finally, the concentration of capital among larger funds has created windfall profits for managers as management fees no longer just cover costs but have become a valuable profit center.

As shown below, we argue that high management fees can be a direct transfer of valuable alpha from investors to the managers.  In fact, a reduction in management fees results in a dollar for dollar increase in expected alpha.  In this way, fee reduction is the purest form of alpha.

Less intuitively, we also argue that performance fees can be equally problematic.  As beta returns rise, fund returns generally increase as well.  However, the absence of a hurdle rate means that investors often pay incentive fees on beta.  Consequently, as markets rise, alpha received by investors can actually decline.

Breakdown of Hedge Fund Returns

Investors today are much more knowledgeable about the composition of hedge fund returns.  A framework that includes multiple forms of beta has supplanted the simple model of equity beta/alpha.  The very definition of beta has broadened to include benchmark strategies and other investment programs designed to efficiently deliver returns from more exotic risk premia.  The net effect of this is that over time, betas have come to explain a greater and greater portion of returns, which leaves less and less in the “pure alpha” category.

The current thinking is that there are four primary sources of returns:  static beta, dynamic beta, alternative beta and alpha.  A brief description of each is included in the box below.

In practice, when looking at an individual fund, it can be difficult to cleanly distinguish between different categories. Fee Paper 1Should we consider the decision to cut risk prior to a market drawdown alpha, dynamic beta, or simply luck?  As the firm evolves and markets change, at one point does a shift in static betas represent a form of dynamic beta?  Into which category should we place dynamic allocations to alternative betas?  More broadly, as investor sophistication grows, will we continue to move more and more sources of alpha into defined beta categories?

In the chart below, we order the different sources of returns according to expected correlation to traditional assets and Sharpe ratio.  Static beta clearly has the highest correlation to traditional assets and a low expected Sharpe ratio.  Alternative betas have a higher expected Sharpe ratio and much lower correlation to traditional assets, which is precisely why they used to be categorized as alpha.  Dynamic beta – shifts in exposures and asset allocation weights – is much more variable.  Alpha stands on its own with both a very low correlation and very high expected Sharpe ratio.

Fee Paper 2

Clearly, the most valuable portion of the return stream is alpha.  This is what investors seek when they invest in hedge funds:  a reliable source of returns that is noncorrelated to the rest of their portfolio.  The justification of a high fee structure is grounded in the belief that a talented manager can generate excess returns over time, and that these returns will be utterly uncorrelated to overall market movements.  In fact, the very statistical definition of alpha is most easily visualized as the return that the fund should generate when the market return is precisely zero.

In this light, the expected alpha of a portfolio should be highly stable and noncorrelated.  This can seem counterintuitive at first.  After all, every manager has good and bad years; market conditions at times are better and worse for a given strategy.  The point is that investors expect alpha to be noncorrelated and that there is no logical reason why the excess returns should be driven directly by market conditions.  After all, if alpha predictably increased in rising markets, then by definition we would classify a portion of it as beta.  Therefore, alpha generation must be truly independent of the various forms of beta.

Alpha, Hedge Fund Returns and Fees

With this framework in mind, assume we have a simple hedge fund that has a net exposure to the S&P of precisely 0.50 and delivers 600 bps per annum of alpha before fees.  The manager does not employ alternative beta strategies and market exposure does not change over time. As noted, alpha generation does not vary with market returns.  Due to the fund’s remarkable consistency of outperformance, the manager is able to charge a 2% management fee and 20% carry.

Given the stability of alpha, all variation in fund returns will be driven by moves in the market.  In the chart on the left, we show net fund performance at market returns of 0% to 20%.  In the chart on the right, we show how total fees paid decline as returns increase, which is what investors expect.

Fee Paper 3

For instance, with the equity market up 10%, the fund returns 7.2%, 220 bps of which is alpha.  Certainly, almost 35% of gross returns were paid to the manager, but roughly half of this consisted of performance fees which are paid only when the fund performs well.  Most investors would be content with this.

But what is the effect on alpha received by investors?  When the market returns zero, the manager earns 6% before fees.  Even though we pay away 2.8%, we’ve earned 3.2% in a difficult year for equities and our manager has delivered 320 bps of alpha.  Simply, the manager generated 600 bps of alpha and we were willing to pay away 47%.  Expensive for sure, but alpha is highly valuable, and not many funds can consistently deliver it.

But what happens when the market returns 10%?  As noted, the fund returns 7.2% and investors received 220 bps of alpha – again, a very respectable performance.  Performance has increased, yet alpha has declined.  At a market return of 20%, the fund gains 11.2% net while alpha has declined to 120 bps.  What’s going on here?

The issue is that performance fees are paid on both alpha and beta.  As beta returns increase, the investor pays a higher performance fee without a commensurate increase in alpha.  The chart on the left below shows how the percentage of alpha paid away rises from 47% to 80% as market returns rise from 0% to 20%.  All of this higher payout is due to higher performance fees.  The chart on the right looks at the same question from a different angle:  how much alpha (before fees) does the manager need to generate to deliver 250 bps of alpha (after fees)?  The more the market rises, the higher alpha needs to be. In an up 20% market, alpha must be over 750 bps in order for the investor to net 250 bps; here, over two thirds is paid away due to performance fees on beta.

Fee Paper 4

Implications:  Why Fee Reduction is the Purest Form of Alpha

At multibillion dollar hedge funds, management fees have become a profit center – in many cases, a more important contributor to firm profits (and firm value) than performance fees.  This represents an enormous transfer of wealth to the managers.  The markets understand this.  When analysts (or strategic investors) seek to value an alternative asset manager, profits derived from management fees are valued at roughly twice those of performance fees.  Why?  Because management fees are stable and are paid irrespective of whether the market and/or fund is up or down.  Management fees are the pure “alpha” of the hedge fund management company.

Performance fees, while intuitively appealing to many investors, often do not result in a more equitable sharing of risk and reward.  In the example above, we might be comfortable paying 80 bps of performance fees when the market returns zero and the fund has returned 6% gross, but it should give us pause that we pay away another 200 bps, now 80% of alpha, simply because the market rose 20%.[1] As investors, we bore that risk and its benefit should inure to us.

Conclusions

There are two obvious ways to make the hedge fund fee structure more equitable over time:

  • Management fees should scale downward as fund AUMs increase.  When management fees become a profit center at large funds, this results in a direct transfer of the most valuable portion of the return stream from investors to managers.
  • Incentive fees should have a hurdle based on the appropriate measure of fund beta (or betas).

We have repeatedly made the point that “fee reduction is the purest form of alpha.”  In practice this means that investors need to consider the idiosyncratic nature of a given hedge fund when deciding which type of fee reduction is likely to be most valuable over time.  For a fund with high beta strategy (e.g., activist, event driven), the net benefit of a hurdle rate on performance fees might far exceed that of a modest reduction in management fees.  For smaller funds, a higher management fee might be necessary to support operational stability and depth of research, but early investors might insist that this scales down as AUMs increase.

This begs the question, what is an equitable split between managers and investors for pure alpha generation?  One extremely sophisticated family office recently offered that they were content paying away 40%, provided that it truly was for alpha and not a disguised form of beta.  This seems like a rational starting point for an ongoing debate among investors and managers, and certainly is preferable to some of the adverse outcomes outlined above.

 


[1]              From a technical perspective, the hedge fund manager has been given a free call option on the market that is equivalent to a one year European call option struck 4% in the money on notional equal to 10% of our investment in the fund.  The present value of this one year option is between 80-100 bps.  Investors hand this to the manager each January 1.

The recent launch of several mutual funds of hedge funds (or, more accurately, funds of managed accounts) geared to the US retirement market has turned attention to the question of how much investors should expect to sacrifice in returns for daily liquidity and pricing.  Since these products mostly have very short track records, the answer to this question is best found by looking at past efforts to introduce greater liquidity into the hedge fund investment model.

The historical record is not particularly promising.  Investable indices were first launched over a decade ago but initially suffered from adverse selection bias that caused one index to underperform its non-investable counterparts by 300-600 bps during its first four years of live performance.[1]  More recent innovations like managed account platforms and UCITS structures appear to have a lower performance drag, but making apples to apples comparisons is notoriously difficult.  In a recent study, Cliffwater sought to calculate the annual return difference between 148 hedge funds and their direct liquid counterparts, such as managed account platforms, 40 Act funds, bank platforms and UCITS structures.  They concluded that investors in the liquid alternative sacrificed approximately 1% per annum due to a combination of fees and lost alpha.  They noted that the annual cost was greatest for Event Driven managers (2.3% per annum) and lowest for Managed Futures (0.5%) and Macro (0.2%); Equity Long/Short was close to the mean at 1.1%.   However, the performance drag in the Cliffwater study – conducted in March 2013 and based on liquid alternatives that were still in operation at that time – likely is understated since it excludes liquid alternatives that were launched over the past decade but subsequently shut down due to poor tracking.

In this note we seek to estimate the performance drag associated with liquid alternatives to investing directly in hedge funds during the post-crisis period.  Rather than analyze individual managers, we look at the recent performance of several indices that track liquid alternatives.  As shown in the chart below, a composite of liquid alternatives underperformed the HFRI Fund Weighted and Fund of Funds indices by 3.08% per annum and 1.29% per annum, respectively, over the three and a half year period from January 2010 through June 2013.  Factoring in some of the data biases discussed below and variance due to fee structures and strategies, we estimate that it is reasonable for a typical investor in hedge funds to expect a performance drag of 150-250 bps per annum when electing to invest in highly liquid hedge fund strategies.

Performance Drag 1 (2)

Data Considerations

At the hedge fund index level there are two principal issues with the data.  The HFRI Fund Weighted index is slightly overstated due to reporting bias; that is, funds have a window in which to decide whether to report a given month’s returns.  Second, the HFRI Fund of Funds index includes a second layer of fees, which accounts for some of the recent performance differential.  In general, we assume that the HFRIFWI is a reasonable proxy for the direct hedge fund portfolio of an institutional investor, while the HFRIFOF is more representative of the returns expected by smaller investors.

With respect to the liquid products, we ideally would analyze the full universe of managed account and UCITS products that are (and have been) designed to track the performance of a given hedge fund so that we can make apples to apples comparisons on a fee equivalent basis – an expanded version of the Cliffwater study.  However, proprietary platforms generally restrict access to data and hence make such comparisons impossible or, alternatively, present the data in a self-serving manner.  Further, institutions that have negotiated special managed accounts typically do not publish comparative results.

Given these limitations, there are three principal hurdles in the data available to us:

  • Fee Equivalency.  In general, there is very little transparency into whether certain indices (e.g. Lyxor) include platform level fees.  Some products, such as the Credit Suisse AllHedge Solutions, are actual investment products, while others merely aggregate reported data.  UCITS and registered funds may have materially different fee structures than the actual hedge funds.
  • Selection Bias.  When first introduced, indices are often backfilled and historical results are replete with survivorship bias.  Hence, we focus only on live performance.  Further, indices that perform poorly can be shut down, as was the case with the Dow Jones Credit Suisse Core Hedge Fund index in May 2013.
  • Strategy Bias.  A liquid index may overweight certain strategies that are inherently more liquid and therefore more conducive to being offered in a liquid structure.  Certain strategies like managed futures have underperformed others (e.g. credit) and therefore may depress results.  To address this we look at a group of indices rather than focus on a specific provider.

We hope to overcome these limitations by looking at a broad enough sample of indices so that any fund level or platform specific bias has a limited impact on overall results.  In light of this, the data shows that there is a consistent and significant performance drag associated with liquid alternatives.

What Causes the Performance Drag?

There are several likely causes of the persistent performance drag.  Daily (or near daily) liquidity should limit the investment universe to positions that, by definition, lack an illiquidity premium.  If illiquid assets command a 3-4% return premium over time, we might expect a performance differential of perhaps 1% over time if we assume that a typical hedge fund with restrictive liquidity terms might hold a quarter to a third of its portfolio in less liquid instruments.  Note that certain liquid hedge fund proxies – such as the Lyxor managed account platform and factor based replication models – outperformed materially during the crisis, which supports the conclusion that at least a portion of the differential is due to a mismatch in the liquidity of the underlying portfolios.  However, this clearly does not explain the entire differential.

As noted, adverse selection among managers was very pronounced in the formative years of the investable index business.  More recently, established and highly regarded hedge fund managers appear much more willing to run liquid alternative products and the performance differential has narrowed.  The persistent performance drag even in recent years suggests that investment restrictions in more liquid products cause a portfolio-level form of adverse selection.

Another cause, as noted by Callan in a recent research report on alternative mutual funds, is that registered funds and similar vehicles restrict leverage, which can be a key contributor to returns in certain strategies.  A fourth likely cause is that managers need to retain a cash buffer to manage more frequent inflows and outflows.

The mix of factors clearly will vary from product to product and strategy to strategy.  Cliffwater’s conclusion that the performance differential is lowest for managed futures is consistent with the ease of running such strategies in multiple vehicles and the leverage available through the futures markets.

Conclusions

The initial appeal of the hedge fund model was that it enabled talented and motivated managers to pursue investment opportunities outside the constraints of registered investment products like mutual funds.  Managers were relatively unfettered to pursue compelling investment opportunities as market conditions warranted.  A talented merger arbitrage specialist in the late 1980s might have evolved into a distressed investor by 1990 and a buyer of nonperforming real estate portfolios within a few years thereafter.

The institutionalization of the business over the past decade has gradually introduced a series of new constraints into the investment model.  Institutional investors are wary of style drift and value consistency in the investment process, sometimes even in the face of an inferior opportunity set.  The concentration of capital among large firms post-crisis naturally has narrowed much of the investment universe to situations where managers can deploy hundreds of millions of dollars in a single position.  Post-crisis aversion to gating/suspension risk has led to intense scrutiny of the liquidity of underlying portfolios.

A rational question for investors is whether the structural constraints of liquid vehicles cause the underlying managers to deviate from the core investment strategy to such a degree that it undermines the original investment thesis – that is, to invest in strategies with a higher expected risk adjusted return than, or low correlation to, traditional investments.  Based on a cursory analysis, a recently launched mutual fund of managed accounts product appears to have underperformed the flagship hedge fund counterparts run by the same managers by 600 bps on a fee equivalent basis over just the first half of 2013.  If anything, this suggests that investors will be sorely disappointed if they expect the mutual fund to approximate the performance of a portfolio of those hedge funds over time.



[1]  The adverse selection issue was most acute with the HFRX Global Investable index versus either the HFRI Fund of Funds index or HFRI Fund Weighted index.  Credit Suisse’s Blue Chip investable index did not suffer from nearly as much of a performance drag; however, this product was never designed to offer daily or weekly liquidity.

[2]  The Dow Jones Credit Suisse Index, designed to track the universe of managed account and UCITS products, was launched in early 2011 but ceased reporting in May 2013.  Consequently, comparative statistics are for the live period only.

The Newedge CTA index (“NEIXCTA”) has returned a disappointing 0.82% per annum over the past five years.[1] Many explanations have been given, including the lack of a trending market.  This explanation implies that, as markets normalize, CTA performance should revert to higher historical levels – the 6%+ per annum realized in the five years preceding the crisis.

CTA 1

However, this analysis ignores the role of LIBOR in managed futures portfolios: CTAs earn a return on investor cash posted as margin for futures positions. Consequently, an evaluation of CTA returns should account for LIBOR.  In this light, the historical performance of the index excluding LIBOR over even the preceding ten years is distressingly low: approximately 1.5% per annum, or less than half the reported figures, and slightly negative over the past five years.

CTA 2

Viewed on an annual basis, the impact of LIBOR on pre-crisis returns is strikingly clear:

CTA 3

This analysis suggests that the performance issue for CTAs may be systemic.  Despite this, CTA assets under management has increased five-fold over the past ten years.[2]

CTA 4

The question is why. The most compelling explanation is that investors place a premium on assets with a low correlation to equity markets. What’s often overlooked is that a key assumption underlying modern portfolio theory is that each asset has the same expected risk-adjusted return. From the above, it is difficult to conclude that the expected returns for CTAs should be comparable to those of assets with clearly defined risk premia – like equities or credit.

A secondary explanation is that CTAs are often sold as a “hedge” with the expectation that they will perform well in a material drawdown in the equity markets, as they did during 2008.  The return standard for insurance-like investments may be somewhat lower than that for others:  it may be given a “pass” when it loses money as long as other assets are rising.  A third possible explanation is that CTAs generally have lenient liquidity terms, a valuable feature post-crisis.

The primary beneficiary of the increase in assets under management appears to be the managers themselves.  Using an assumed fee structure of 2/20, slightly less than half of gross returns were paid away to managers over the past ten years.  Adjusted for LIBOR, the numbers are more pronounced:  fees were roughly double what the managers made above collateral interest.  Over the past five years, virtually all of gross performance was paid away.

CTA 5

The figures raise a fundamental issue of fairness in the managed futures space and argue for a material change in the industry fee structure.  Simply put, investors should question whether managers should be rewarded for warehousing cash in interest bearing accounts.  At a minimum, a LIBOR based hurdle is warranted.

The analysis raises a serious question about whether the pre-crisis returns of managed futures funds were an anomaly or are likely to recur.  Given performance over the past ten years, it appears that investors should expect CTAs to earn 1-3% over time in a low interest rate environment.  Moreover, the performance of systematic strategies is dependent on the ability to capitalize on market anomalies – such as evidence of excess returns from momentum or mean reversion.  Recent poor performance implies that many anomalies are subject to being arbitraged away over time and that the recent underperformance of the industry is likely to persist.


[1] Source: Bloomberg, net returns

[2] Source: BarclayHedge

As institutional investors search for ways to reduce fees in hedge fund portfolios, attention has turned to the relative merits of investing directly in hedge funds vs. seeking to replicate the performance of a given strategy by investing directly in the underlying market exposures, or risk premia.  The idea has intrinsic appeal:  why pay hedge funds 2/20 if returns can be delivered cheaply and efficiently by investing directly in the underlying exposures?

Factor-based replication is highly effective at delivering the results of most hedge fund strategies – especially equity long/short and more directional strategies – but is arguably less effective with strategies with low exposure to traditional asset classes.  For merger arbitrage specifically, it may make more sense to try to replicate the underlying trading strategy itself:  that is, by acquiring a representative sample of corporate acquisition targets and, for stock-based deals, shorting the acquirer.  In theory, this should enable a sophisticated investor to derive similar returns but with greater liquidity and transparency, lower fees and other benefits.

In this note, we examine three indices that have been constructed to deliver a liquid, investable alternative to investing in merger arbitrage hedge funds.  We compare the returns of the live results against the performance of actual merger arbitrage hedge funds and conclude that the approach indeed can be effective at delivering comparable returns, with a few important caveats:

  • First, index design is extremely important – one of the three indices clearly fails at its stated objective and it’s not clear how to anticipate which index will perform well going forward.
  • Second, the indices do not appear to improve returns – that is, eliminating the 2/20 does not result in higher returns.
  • Third, and surprisingly, the ability to short appears to have very little impact on returns.

For comparative purposes, we also explore two alternatives to rules-based trading:  an established merger focused mutual fund and factor based replication.

Merger Arbitrage Index Construction and Results

The S&P Long Only Merger Arbitrage (SPARBM) and the IndexIQ Merger Arbitrage (IQMNAT) indices were launched prior to the financial crisis, while the Credit Suisse Liquid Alternative Beta (CSLABMN) index was launched in January 2010.  Each index seeks to provide a broad representation of the merger arbitrage space by investing in companies that are subject to takeover offers. A review of the index construction methodology of each index provides a window into the complexity of designing a set of coherent and consistent “rules”:

  • which markets to include, especially whether to include emerging market targets;
  • minimum transaction or target size;
  • when to initiate the position, how to size it and when/how to rebalance;
  • which types of transactions to exclude, such as CVRs or non-control tenders;
  • whether/when to short and which instruments to use;
  • whether the target needs to be at a discount to the announced price or not; and
  • how to treat new offers or stale deals.

The following chart provides a comparison of key parameters and recent top five holdings:

1 (2)

Given similar objectives, it’s a bit surprising that no top five position is shared by all three indices.  In fact, two of the top five CSLABMN positions do not appear at all in the IQMNAT holdings (unfortunately, the full CSLABMN position list is unavailable, so it’s impossible to undertake a full portfolio comparison).  The S&P index appears to have more of a midcap bias, with a median daily trading volume of roughly one fourth those of the other two indices.

The differences in construction can have a material impact on returns.  (Note:  Since we are skeptical of back-filled index data, we use only the results from when the sponsor began to publish live results.   Consequently, we examine the two former indices from the beginning of 2008, and include all three after January 2010.)  Surprisingly, IndexIQ’s long/short product had materially greater drawdowns during the crisis relative to the (long only) S&P index and has performed relatively poorly since.  By contrast, CS’s product performed similarly to the S&P index until 2012, when it underperformed by approximately 600 bps.

1 (2)

As noted, the S&P index above is a long only index.  In late 2012, S&P introduced a long/short version of the index, which has had very erratic performance since inception.   The following chart shows the performance of both the long only and long/short indices since September 2012:

S&P Long Only Merger Arb

The variability in returns between the different indices (IndexIQ vs the others) and even among providers (S&P Long Only vs S&P Long/Short) highlights the fact that the specification of rules for an “alternative beta” strategy like  merger arbitrage is far more complicated than that for traditional indices.

Index Performance vs. Hedge Fund Performance

Due to the relatively erratic performance of the IndexIQ index and the recent (and very poor) performance of the S&P Long/Short index, we have excluded them from the following analysis under the assumption that few investors would opt to use them as merger arbitrage proxies at this point.  Instead, for simplicity and clarity, we compare the results of just the S&P (the live period) over the past five years to the performance of the HFRI Event-Driven:  Merger Arbitrage index[1]:

S&P - HFR

Since January 2010, when the CSLABMN was introduced, both the SPARBMN and CSLABMN have outperformed the hedge fund indices. That said, year to year differences can be significant:  the indices outperformed materially during 2010 when many hedge funds deleveraged during the inception of the European fiscal crisis, while the CS index underperformed by approximately 600 bps during 2012.

S&P - HFR - CS

It’s important to note that the indices do not include management fees (although the CSLABMN does include a 50 bps index calculation fee), while the hedge fund indices are reported after hedge fund level fees.  In rough terms, the indices returned around 4% per annum gross over the past three years, while actual hedge funds returned around 3% net.  A more accurate comparison would be to look at the net returns to investors of each approach.  If we assume 100 bps of management fees for the index products, the compound returns over the past three years are comparable.  Based on this, it seems reasonable to conclude that the majority of the hedge fund returns are driven by an underlying risk premium. While an investor might not have realized a material increase in returns over the past three or five years, the rules-based approach may still provide materially better liquidity and transparency.

Alternative Mutual Fund Approach

The recent growth of the alternative mutual fund industry raises the question of whether investors can realize similar returns to hedge funds but in a more highly-regulated, potentially lower cost structure.  Most alternative mutual fund strategies do not have sufficiently long track records to allow for effective comparison; in the merger arbitrage space, we fortunately can analyze the returns of the Merger Fund, a mutual fund that was launched in 1990 with a mandate to focus exclusively on takeovers.  With a large asset base ($4.5 billion), highly diversified portfolio (78 longs and 16 shorts), and a narrow focus, MERFX serves as an interesting proxy for the merger arbitrage sector.  The chart below shows performance from January 2008 to the present vs. the HFR index.

The Merger Fund

The outperformance of the Merger Fund during the crisis may have been attributable to an outsized weighting in the BoA-Merrill transaction, which had a material impact on merger fund returns.

MERFX charges no incentive fee, but the all-in management fees and expenses are similar to those of a typical merger arbitrage fund (1.33% per annum excluding trading and other investment related expenses).  The absence of incentive fees, which for the merger arbitrage hedge fund averaged less than 1% per annum over the past five years, did not appear to translate into higher returns, although some investors may draw comfort from investing in a mutual fund structure.

Factor-based replication

Another approach is to use a factor model to seek to replicate the returns of the merger arbitrage hedge fund indices.  In the following charts, we examine the results of a replication of the HFR index over the past five (left) and three (right) years.[2]

HFR index over the past five (left) and three (right) yearsThese results are based on simulated or hypothetical performance results that have certain inherent limitations. Unlike the results shown in an actual performance record, these results do not represent actual trading. Also, because these trades have not actually been executed, these results may have under-or over-compensated for the impact, if any, of certain market factors, such as lack of liquidity. Simulated or hypothetical trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profits or losses similar to these being shown.

 

We find that a factor-based approach was not effective during the financial crisis; however, since 2010 the factor model approach would have materially outperformed actual hedge funds with comparable volatility but relatively low monthly correlations.  Despite the recent results, as practitioners, we would be disinclined to use factor based replication for a portfolio that consists exclusively of merger arbitrage funds due to the consistently low market exposure.

Conclusions

Any conclusions from the analysis above are constrained by the lack of a robust pool of data.  We have only two merger arbitrage indices that extend back through the crisis; a single large diversified mutual fund; and the hedge fund indices themselves are replete with data biases.  With those caveats in mind, there are several interesting conclusions and questions that follow.

The most important conclusion appears to be that the construction of merger arbitrage indices is complicated and introduces its own form of idiosyncratic risk.  This likely is true for any rules-based “alternative beta” index.  While the results above show that comparable returns potentially can be achieved (with greater liquidity, lower all in fees, etc.), it is by no means obvious how to determine – in advance – which complex set of rules will provide the most effective means to capture the merger arbitrage risk premium.  An investor in IndexIQ’s product in 2008 would be sorely disappointed today, as would an investor in the recently launched S&P Long/Short index, or even an investor in the CS product in early 2012.  Consequently, investors who expect this approach to materially reduce tracking error, monitoring costs or other risks are likely to be disappointed.  Further, the complications involved in developing and running such a program will limit fee savings, as highlighted by the difficulty of making net of fee comparisons between indices and actual investments.

If the rules-based indices, on average, underperform actual hedge funds, then the question is why.  One possible explanation is that by focusing on larger, simpler deals the indices over-allocate to more efficiently priced transactions – where the merger arb risk premium is low.  Discussions with actual managers suggest that traditional sellers – long only mutual funds – are now more willing to make their own assessment of deal risk, which undoubtedly has improved with experience and better dissemination of information.  This should naturally compress the premium over time and force managers to seek excess returns from more complicated transactions – those with variable consideration, the likelihood of a topping bid, capital structure opportunities, etc.

A broader question that arises is whether “alternative betas” are stable over time. Investors generally believe that certain risk premia – for equities, for credit, for illiquidity – should persist indefinitely; however, even traditional risk premia dramatically compress (equities during the 2000s) and widen (illiquidity post-crisis) in different market environments.   Since more complicated risk premia – such as those for benchmark strategies like value, momentum and “insuring” takeovers – are based on the principle that less sophisticated/more constrained investors will consistently transfer value to more sophisticated/less constrained investors, the key question is whether more widespread knowledge of these strategies – and how to efficiently implement them – will inexorably lead to capital inflows, more educated sellers and, by definition, a compression of excess returns over time.


[1] We also compared the results to the Dow Jones Credit Suisse Risk Arbitrage hedge fund index.  The results are very similar to those of the HFRI Merger Arbitrage index; therefore, for ease of comparison we have limited the analysis in this note to the latter.

[2] Note that the HFR Merger Arb replication performance presented here is gross of replication fees in order to make the returns comparable to the rules-based merger indices.

This note re-examines two frequently cited studies on factor-based hedge fund replication:  Hasanhodzic and Lo’s seminal paper, “Can Hedge Fund Returns be Replicated?:  The Linear Case” (“Lo”) and Amenc et al., “Performance of Passive Hedge Fund Replication Strategies” (“Amenc”). Lo was the first to articulate that a linear, factor-based model could successfully replicate the returns of various hedge fund strategies. Amenc, on the other hand, was highly critical of the approach and sought to disprove its effectiveness.

As outlined below, the most important finding of the Lo paper is often overlooked:  that the simple five factor model appears to have done an even better job of replicating the returns of the sample than the authors articulate. The Amenc paper, on the other hand, was highly critical of the approach and concluded that the replication results were consistently inferior to those of actual hedge funds.  However, the study’s conclusions were severely undermined by poor factor specifications which distorted the results.

Hasanhodzic and Lo, “Can Hedge Fund Returns be Replicated?:  The Linear Case” (2007)

This important paper, first released in 2006, introduced the concept of using a 24 month rolling-window linear regression to replicate hedge fund returns out of sample.  In many ways, this seminal paper launched the factor-based hedge fund replication business.  Remarkably, though, the authors overlooked the most important conclusion:

  • Using a simple five factor model, the replication of an equally-weighted portfolio of 1,610 funds appears to capture all or virtually all of the returns over almost 20 years, adjusted for survivorship bias.

In other words, the simple clone’s performance exceeded all expectations and is consistent with the performance of actual hedge fund replication indices since 2007. Remarkably, this pro forma performance of the clone was approximately equal to the performance of the S&P 500 over the same period, but with materially lower volatility and drawdowns. This is a startling result that is lost in the paper’s forty pages of formulas, text and tables. Here’s why:

The data set used was based entirely on “live” funds in the TASS database as of September 2005 – 1,610 funds.  Invariably, “live” funds have outperformed “dead” peers by a wide margin:  in the HFR database, for instance, by more than 400 bps per annum. Inexplicably, Hasanhodzic and Lo assert that “any survivorship bias should impact both funds and clones identically,” and therefore can be ignored. This simply is incorrect. We know today that these kinds of data bias, by definition, are “non-replicable.”  Therefore, the clone should be compared to actual realized performance – i.e. adjusted for survivorship bias. This is why replicators are often benchmarked against indices the like HFRI Fund of Funds index that are more representative of actual investor returns.

From Figure 5 in the paper, we can infer that the equally-weighted portfolio of sample funds returned between 13% and 14% on a compound annual basis over almost twenty years. This clearly is unrealistically high: hedge funds as a group simply did not outperform the S&P by 200-300 bps per annum on a net basis during a twenty year bull market in which stocks returned 10% per annum. Assuming several hundred bps of survivorship bias, the hedge fund portfolio would have slightly underperformed the S&P 500, but with materially lower drawdowns and volatility.  And, in fact, this is precisely how the simple clone performed. See Figure 5 reproduced below with commentary added.

Factor based photo 1

In this context, the performance of the linear clone (around 10% per annum) is remarkable and should have been highlighted more prominently.

A secondary issue is the use of a factor set that is missing important market exposures. The study employs only five market factors: the S&P 500 total return, the Lehman AA index, the spread between the Lehman BAA index and Lehman Treasury index, the GSCI total return, and the USD index total return. More recent studies, including our own, have demonstrated that emerging markets, short term Treasury notes and small capitalization equities are important factors since they enable the models to incorporate, respectively, volatility expectations, yield curve trades and market capitalization bias. Conversely, while the inclusion of the GSCI has intrinsic appeal, it does not appear to be additive over time to out of sample results. Consequently, the overall results arguably would have been even more compelling with a slightly more robust factor set.

Amenc et al., The Performance of Passive Hedge Fund Replication Strategies (2009)

In response to the paper by Hasandhozic and Lo and the launch of several factor-based indices, EDHEC released several papers that were highly critical of the concept during 2007-09.  In the first paper, “The Myths and Limits of Passive Hedge Fund Replication:  An Attractive Concept… Still a Work-in-Progress,” the authors seek to redo the rolling linear model employed by Hasandhozic and Lo, but apply it to the EDHEC hedge fund database. Since there is very little explanation of the underlying data, it is impossible to estimate the effect of survivorship bias or other sampling issues.

The more relevant paper was published in 2009, “The Performance of Passive Hedge Fund Replication Strategies.” It is difficult to read this paper without the sense that the authors, who are closely tied to the fund of hedge fund industry (and funded by Newedge), had a predetermined agenda.  The end result is a paper that includes some very helpful analysis – for instance, that Kalman filters and non-linear factors don’t improve out of sample results – but whose conclusions are undermined by selective omission.  For instance:

  • Even though there was over two years of live data from replication indices that showed strong results with high correlation through the crisis, the authors neglect to include this and focus instead on re-doing the Lo analysis with the admittedly incomplete five factor set.
  • When the authors do in fact acknowledge that Lo’s factor base should be expanded to include emerging markets, small cap stocks and other factors, they test each strategy with an unreasonably narrow subset of factors even though it was well established by this time that a more robust factor set was critical.  This is discussed in detail below.

In Section 3.2, the authors “test whether selecting specific sets of factors for each strategy leads to an improvement in the replication performance. Based on an economic analysis and in accordance with Fung and Hsieh (2007), who provide a comprehensive summary of factor based risk analyses over the past decade, we select potentially significant risk factors for each strategy.” The factors identified are quite reasonable, such as the spread between small and large capitalization stocks, emerging markets, and other fixed income spreads.

In the table below, the five factors on the left side represent the original Lo portfolio, while the five on the right represent the Fung & Hsieh additions.

Factor based photo 2

The logical next step would be to test whether the results of the Lo five factor set is improved by the addition of one or more of the factors.  Instead, the authors only use 1-4 factors for each strategy and throw out most of the original factors. Remember that at this time it was well established that a narrow factor base was insufficient to replicate most hedge fund returns.  This is why Merrill, Goldman Sachs and others all used 6-8 factors, not 1-4. To use one example, in order to seek to replicate the macro space, the authors used only the Lehman AA Intermediate Bond index – a single factor – with a 24 month rolling window. For distressed, the one factor is the spread between a BAA index and Treasurys. For risk arbitrage, it’s only the S&P 500. For long/short equity and funds of funds, it’s the S&P 500 and the small cap-large cap spread.

To underscore the point, the debate at the time was not whether one or two factors could reasonably replicate sector returns, but whether a diversified portfolio of market factors could do so. By starkly reducing the factor set, the authors essentially designed an experiment that was bound to fail.  Consequently, investors should seriously question the validity of the authors’ conclusion that “the performance of the replicating strategies is systematically inferior to that of the actual hedge funds.”