Why Emerging Managers Now?

Transcription

1 Why Emerging Managers Now? Infusion Global Partners Whitepaper May 5, 2015 Abstract: Traditional asset classes appear to offer uninspiring beta returns at present, and recent years hedge fund returns have disappointed both in magnitude and diversification benefits, likely reflecting capacity pressures associated with the concentration of AUM and inflows with larger funds. We argue that, by contrast, Emerging hedge funds offer a rich opportunity set with far fewer capacity issues where skilled managers with concrete competitive advantages in less efficient, smaller capitalization market segments can generate better, more sustainable and less correlated excess returns. Emerging managers do involve more investment and operational risk than larger peers; to that challenge we offer some suggestions on a thoughtful and rigorous approach to constructing an Emerging Managers allocation and balancing effective due diligence with scalability. 1

2 Jan 2006 Jul 2006 Jan 2007 Jul 2007 Jan 2008 Jul 2008 Jan 2009 Jul 2009 Jan 2010 Jul 2010 Jan 2011 Jul 2011 Jan 2012 Jul 2012 Jan 2013 Jul 2013 Jan 2014 Jul 2014 Jan Introduction: Searching for Beta and Alpha Most investors would agree that the current market environment makes their return goals particularly challenging. Equity markets are at or near all-time highs and relatively expensive by most metrics. Prospects for bond returns continue to be severely repressed by central bank policy to negative yields in several cases! and credit spreads offer little compensation for default risk that is likely to grow as the current expansion ages and becomes more fragile. Real estate has been significantly bid up in the global search for yield and shows signs of overcapacity in a growing number of markets and segments, and the supply-demand imbalances that have pressured commodities lower over the past year look set to continue for some time. With beta returns thus expected to offer meager risk-adjusted returns over the near term, many investors have increased their focus on alternatives, and hedge funds in particular. In fact, assets continue to flow into hedge funds; hedge fund AUM is estimated to have grown at an average rate of 16% since 2008 and reached a peak of $2.5 trillion in 2014 according to Eurekahedge. 1 However, this dynamic may have a self-defeating aspect in at least two ways. Larger asset volumes are more likely to face capacity pressures that limit return potential (which we will discuss in more detail), and to increase correlations with broad market indexes and thus reduce diversification benefits. Both trends are evident in the data from recent years: Figure 1 below 25.0% 20.0% 15.0% 10.0% 5.0% Figure 1: Hedge Fund Returns Lower in Recent Years 0.0% -5.0% -10.0% -15.0% -20.0% Eurekahedge Main Hedge Fund Index, Rolling 12-month returns. Source: Eurekahedge. shows that average hedge fund returns fluctuated in a range mostly above 10% until 2011 (with a large dip through the Global Financial Crisis), but since that time have languished around or even below 5%. Figure 2 illustrates the deterioration of diversification benefits, as hedge fund correlations with major market indexes now exceed 0.75 in some cases. 1 Hedge fund reporting is voluntary, so figures vary across data providers, but other estimates are similar. 2

3 Jan 2006 Jul 2006 Jan 2007 Jul 2007 Jan 2008 Jul 2008 Jan 2009 Jul 2009 Jan 2010 Jul 2010 Jan 2011 Jul 2011 Jan 2012 Jul 2012 Jan 2013 Jul 2013 Jan 2014 Jul 2014 Jan Figure 2: High Hedge Fund-Market Correlations Rolling 36-month correlation between Eurekahedge main Hedge Fund Index and S&P 500 One attractive alternative may be to focus on Emerging hedge funds those with smaller assets under management that are much less likely to be hampered by capacity issues. The opportunity set of emerging managers is many times the size that of larger funds, with literally thousands of funds under $200 million in assets under management (a common but arbitrary definition). Evidence shows that smaller managers have historically outperformed larger managers on average, and above-average emerging managers have outperformed correspondingly above-average (e.g. 60 th percentile in respective universes) larger managers more substantially. Several rationales support the logic of an emerging manager focus, including the ability to operate, and focus more heavily on best ideas, in smaller capitalization segments of markets which tend to be less efficient and where concrete competitive advantages are more plausible and viable. Smaller funds do entail additional investment and operational risk, which we address in the final section of this whitepaper. 2. Are Smaller Capitalization Market Segments Less Efficient? Market efficiency is a multi-dimensional concept about which investors have widely differing opinions. A market s efficiency is thought to be correlated with its trading volume, depth of analyst coverage, transparency of pricing, and other factors. Most of these factors are correlated with each other, providing support for the view that larger capitalization markets are relatively more efficient and smaller capitalization markets less efficient. However, is it possible to measure market efficiency objectively and quantitatively? The most widely accepted academic definition is that a market is efficient if it is not possible to earn excess risk-adjusted returns using publicly available information. This is commonly approximated by testing for statistical significance in the prediction of returns using public information. A simple measure of such predictability is autocorrelation, the correlation of a security s returns with its own lagged returns. 2 To compare the efficiency of large vs. small cap US stocks, we computed autocorrelations using daily returns over the period , for each constituent of the Russell 2000 and S&P 500 universes. The expected value for each autocorrelation under the assumption of market efficiency is zero, with a standard deviation of about Figure 3 below plots the resulting distributions of the 2000 and 500 coefficients for the two universes, along with the confidence interval where estimated coefficients are not significantly different from 2 This measure was used in Fama s seminal 1970 paper on testing market efficiency. 3

4 zero. It shows that the small cap universe has a much larger percentage of stocks (and number of stocks, since the universe is four times larger) with significant autocorrelations, consistent with the smallcap market being much less efficient. Figure 3: Distribution of Autocorrelation Coefficients 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% 95% confidence interval Russell 2000 S&P 500 A broader view of market efficiency comes from academic and practitioner research on the many anomalies, including value, price and earnings momentum, and other factors, that have been shown to generate excess risk-adjusted returns in a wide range of markets and time periods. 3 Researchers have consistently found that returns to these anomalies are larger and more statistically significant in smaller cap market segments. Figure 4 is a recent example from a paper by Israel and Moskowitz, showing long/short returns for Value (measured by book-to-price ratio) and Momentum (measured by past one-year returns, excluding the most recent month) in US stocks divided into five groups by market capitalization. The dark section of the bars shows the return from the top 1/5 of stocks ranked by the factor, and the lighter section represents the return from shorting the bottom 1/5 of stocks in each cap bin. For momentum, the returns in the largest cap bin (approximately the S&P 500) averaged 300 to 400 basis points lower than the returns in the smaller cap bins, and for value the returns in the two largest cap bins were far below those in the smaller cap bins, again consistent with larger capitalization markets being more efficient. 4 3 Amongst a very lengthy list of references, see Asness, Moskowitz and Pedersen, Value and Momentum Everywhere [2010] and Fama and French Dissecting Anomalies [2008]. 4 The smallest cap bin is primarily microcap; bins 2 and 3 correspond to the smaller and larger half of smallcap, bin 4 to mid-cap, and bin 5 to large cap. See Israel and Moskowitz (2013) for more details. 4

5 Figure 4: Returns to Momentum and Value in US Equities by Size Category 3. Benefits of Being Small a) Concrete Competitive Advantages Every active investment manager strives to have some edge that will allow them to outperform. These competitive advantages are usually a function of proprietary information that the manager is able to obtain or generate, about the securities and forces affecting them in the target universe. Larger capitalization markets, almost without exception, have greater analyst coverage, newsflow, and other readily available information, making it harder for any manager to have a concrete edge in that segment of the market. We believe that areas with less analyst coverage and newsflow, where information is scarcer and more proprietary, offer greater opportunities for a manager to have an edge. These markets usually have lower capitalization and less liquidity, which constrains the size of managers able to operate in those market segments. b) Best Ideas and Market Niches Most managers have a formal or informal process for ranking their candidate investments, and believe that their top ideas are likely to outperform others further down the spectrum. This is supported by academic evidence confirming that best ideas outperform other portfolio constituents, across a wide range of investment styles and time periods. 5 It almost goes without saying that smaller managers are more able to operate in smaller capitalization market segments, both in terms of the positions they hold and in their trading. To illustrate, consider a large fund with AUM of $1b and an emerging manager with $100m, both considering a $500m capitalization small cap stock as a best idea. 5% is a common figure both as an upper bound of the amount a manager would want to hold of any company s stock, and a portfolio weighting large enough to be a meaningful high-conviction position. 5% of a billion dollars is $50 million, or an unacceptably high 10% of the stock s capitalization; 5 Best Ideas, Cohen, Polk and Silli, MIT working paper

6 % building such a position would also entail either a long time period or taking a large majority of daily trading volume for several days which would likely move the price so much that it degrades the return opportunity (and present issues in exiting the position as well). The small manager, by contrast, can amass a 5% portfolio position in the stock by owning only 1% of the outstanding shares, and since volume for many stocks exceeds 1% of outstanding shares daily, trading in and out of such a position would be quite feasible. 4. The Emerging Manager Opportunity Set Having established that smaller capitalization market segments tend to be less efficient, and that smaller funds are more able to operate and have an edge in those segments, we now take a look at the emerging manager opportunity set. Figure 4 is a scatter plot of Eurekahedge s database of 16,000+ funds annualized returns against their assets under management, with each point in the cloud representing a fund. 6 The dominant feature in the data is the 50 Figure 5: Annualized Return vs. AUM Source: Eurekahedge USD millions much greater dispersion of returns for smaller managers vs. larger peers. While some of that dispersion is on the downside, we see rich potential in the thousands of funds in the northwest corner of the chart. Specifically, there are more than 2500 funds under $200m with annualized returns over 10%, and only 186 funds over $1 billion with comparable performance. Many of those high-performance, smaller funds operate 6 We exclude all funds with AUM listed as either 0 or missing, and outlier returns above +50% (of which there are 93) or below -30% (of which there are 61), resulting in 14,302 datapoints. Returns are from inception through March 2015 (or last reporting date data includes dead funds to limit survivorship bias); AUM figures are as of the end of the return period. Vertical clusters are often different currency share classes of a fund. 6

7 in less efficient market segments with concrete competitive advantages. We now turn to some suggestions for how to validate those characteristics and mitigate associated investment and operational risk. 5. Constructing an Emerging Managers Allocation a) Inefficiencies, Structural Advantages, and Risk/Return Evaluation As mentioned above, we believe that skilled managers are more likely to be able to generate a sustainable competitive advantage in less efficient market segments, which underlies our preference for small cap. However, capitalization is only one dimension of efficiency. Here we briefly touch on some other characteristics of markets associated with inefficiencies. One relates to the investor base. If a substantial percentage of the capital in a space is not focused on return maximization, to the point of being the marginal buyer or seller, that is likely to leave opportunities for alpha-seeking participants. A large and important such investor group is the world s central banks, with their involvement in bond and currency markets. In our estimation, fund manager returns from trying to predict central banks effects on these markets have been quite mixed, but there is no question that the magnitudes, breadth and persistence of those effects have been sizeable. A second class of market participant that is often not return optimizing, and that has grown dramatically in recent years, is ETFs. Many ETFs follow rules for trading and security inclusion/exclusion that create opportunities that others may exploit, such as mechanically rolling futures contracts or pro-rata purchases of eligible offerings. A third theme that has been important in recent years is demand for yield with much less regard for total return, which has fueled appreciation of high-dividend shares. Market efficiency can also be hampered by impediments to the flow of information about prices and trading volumes. For example, real-time pricing is available for most equities, government bonds, and foreign exchange, but US municipal bonds (a $3.7 trillion market) have no centralized source for intraday pricing and a large percentage of the thousands of bonds in that market do not trade on a typical day. Similar conditions exist for structured products such as residential and commercial mortgage-backed securities and some high-yield bonds. This provides an opportunity for funds with superior price estimation capabilities perhaps based on data or modeling to have a concrete edge over competitors. While Emerging Managers typically have short track records, we believe it is still important to evaluate as rigorously as possibly their live performance, not only on an absolute and risk-adjusted basis, but also in the context of the investment environment. This may be facilitated by relevant and verifiable track records from previous organizations, which can partially ameliorate the challenges of short track records. In our view, however, there is no substitute for a thorough understanding of the manager s investment process from idea generation through to trade execution. One measure we find quite useful is unlevered real alpha, which we define to be the returns generated after adjusting for leverage and any passive exposure to a market. For example, if a manager generated returns of 18%, using 200% gross exposure and a beta to the relevant market of 0.5 in a year where the market was up 10%, the unlevered alpha would be 6.5% = (18 0.5*10)/2 since 5% of the returns were attributable to beta and half of the remaining 13% came from leverage. Unlevered real alpha is a useful reference value for comparing managers underlying return generation capability and, when broken down in different dimensions such as long vs. short, per trade, by sector, etc. can greatly deepen understanding of a strategy. 7

8 b) Investment Risk Management Emerging Managers are viewed as being substantially more risky than their larger peers, which makes sense, since smaller portfolios can take positions in smaller, more volatile securities than large funds, smaller portfolios may be less diversified, and small funds likely have more limited resources for independent risk and middle/back office functions. In fact, however, the difference is often not large: for example, the median Eurekahedge fund under $200m AUM has a volatility of returns of about 10%, while that of the median fund larger than $200m is about 7.5%. It is not difficult to construct a portfolio of 5-10 emerging manager strategies with volatility below 7.5%, which is more in the range of bonds than equities. However, volatility is a crude proxy for the downside investment risk that most investors are truly concerned with, in the case of emerging as with larger managers. Direct, statistical analysis of downside risk using measures such as Sortino, MAR and Omega ratios is helpful but often provides limited benefit even with several years of data to work with, given that important risks can lie in tails and are rarely observed. As with the investment process, we believe there is no substitute for a good understanding of the key sources of investment risk: exposure concentration, leverage, and liquidity. It is essential to validate that constraints and procedures are in place to mitigate excessive risk exposure, such as limits on individual position sizes and exposures to the overall market and particular industries, sectors, countries, or currencies. Leverage is another key source of risk that can be challenging to understand and compare across strategies since it may be defined, employed, and obtained in different ways depending on the context. For example, futures and options contain embedded leverage and a strategy may implicitly have considerable leveragerelated risk despite a large percentage of AUM kept in cash (low margin to equity ratio). In cash-instrument strategies, a given amount of leverage may lead to dramatically different risk depending on how hedged the resulting leveraged positions are. For example, one fund may run at 500% gross exposure ($5 invested for each $1 of AUM) but keep the portfolio very close to neutral exposure in all key dimensions, and be less risky than a more directional fund with little or no leverage. Liquidity is a third key source of risk that has been the driving factor in many hedge fund failures. In our view, it is crucial to understand and validate that the liquidity of portfolio holdings is consistent with the liquidity terms of the fund, so that managers do not find themselves needing to be a forced seller to meet redemptions, or even to gate. Of course this needs to be evaluated under both normal and stress conditions, since in some asset classes liquidity can deteriorate quickly and dramatically (though in equities and many futures markets, trading volumes typically rise in a crisis). A key tool in the risk kit is transparency. At its essence is the idea that to control something you need to be able to measure it. Managers need to know their cash positions and exposures in order to accurately assess risk, and analytic risk software packages take position or exposure level data as inputs. Historically, hedge fund managers have maintained considerable secrecy about the details of their portfolios, in part to protect their intellectual property. However, with the increase in concern about risk in recent years, and the shift in hedge fund AUM from a primarily high net worth/family office base to a more institutional orientation, investors are demanding, and managers are providing, more transparency. We believe that portfolio transparency can greatly mitigate risks of emerging managers in two important ways. The first is in facilitating understanding of each manager s behavior, where a key concern is style drift. Generally a manager has been hired based on his or her expertise in a particular area pairs trading, merger arb, fundamentally valuing US healthcare stocks and any deviation from past patterns of holdings, trading frequency, or risk exposures should be questioned and understood (and may be cause for withdrawal). The second benefit of transparency is in understanding the aggregation of exposures across the portfolio to any 8

9 market, industry, currency, commodity, or liquidity. We advocate the use of separately managed accounts whenever possible, both for the transparency they provide and for their operational and liquidity advantages. c) Operational Risk Management A significant component of the perceived risks of investing with emerging managers relates to operations. With small teams, short track records, and niche strategies which may be located outside the US and more difficult to understand than more standard approaches like long/short equity, investors are rightly concerned about middle and back office matters. We believe that it is essential to perform operational due diligence to confirm that the team has effective and professional processes for settling and confirming trades, prime broker and administrator relationships and processes, meets regulatory requirements, has controls to mitigate fraud risk, and disaster recovery procedures amongst the many aspects of Op DD. d) Constructing an Emerging Managers Allocation Overall, we believe that emerging hedge funds offer a rich opportunity set without the capacity issues impacting the broader hedge fund market, where skilled managers with concrete competitive advantages in less efficient, smaller capitalization market segments can generate better, more sustainable and less correlated excess returns. The investment and operational risks should not be underestimated, but evaluated in the same way alternative risk/reward choices are traditionally assessed: by estimating the degree to which the risk can be mitigated and diversified, and then whether the additional expected return justifies the increased risk. The resource commitment required to assess this risk-return tradeoff, when combined with the limited capacity that is inherent in the more attractive emerging manager strategies, may present a challenge to investors who either need to allocate substantial minimums or have limited staff availability or expertise. In this case, it may make sense to outsource the emerging manager allocation to a specialist who can identify, diligence and monitor best-in-class funds, and combine them in a portfolio to achieve both diversification benefits as well as more scalability than individual funds would be able to offer. References Asness, C.J., T.J. Moskowitz and L.H. Pederson, Value and Momentum Everywhere, Journal of Finance Cohen, Polk and Silli, Best Ideas, MIT Working Paper, Fama, E., Efficient Capital Markets: A Review of Theory and Empirical Work, Journal of Finance Fama, E. and K. French, Dissecting Anomalies, Journal of Finance Israel, R. and T.J. Moskowitz, The role of shorting, firm size, and time on market anomalies, Journal of Financial Economics