To say that hedge funds have had a tough time navigating the world of activist central banks and pervasive central-planning, would be a vast understatement.
As Barclays writes in a recent report titled "Against All Odds", looking at hedge fund alpha and returns, from 1993 – May 2016, HFs produced cumulatively ~134% of alpha. However, peak cumulative alpha over the period was 139%, achieved in 2011, which suggests that in the last almost 4.5 years, HFs actually generated negative cumulative alpha starting around 2011. Incidentally, that is around the time when both central banks fully took over capital markets, when "expert networks" were busted, and when trading on inside information became virtually impossible.
More specifically, Barclays calculates, the average monthly alpha has declined to -0.07% (annualised ~0.8%) from 2011 to May 2016 compared to an average of +0.48% (-5.9% annualised) for the entire period analysed (1993 to May 2016).
Barclays then compared the risk and returns of HFs from the same time frame.
Figure 3 illustrates the relationship between the 36-month trailing excess returns versus the 36-month standard deviations of those returns (we use this as a proxy for risk) over various market cycles (i.e., ’93 – ’98, ’99 – ’02, ’03 – ’07, ’08 – ’11, ’12 – ’16). As we looked at the various market cycles we found an interesting ‘pattern’ in the relationship between risk and returns. If you exclude the first market cycle in the early to mid ’90s, when HFs were able to generate significant excess returns quite steadily, there is a clear relationship between the excess return HFs generate and the standard deviation of those returns, a measure of the ‘active risk’ taken by HFs. Since 1993, average excess returns have been decreasing rather steadily, while risk has gone up / down over the various periods – in the last cycle, starting in 2011, risk has gone down to its lowest levels, which may help explain / contribute to the industry’s recent underperformance.
In its final return analysis, Barclays compared the excess returns across HF strategies. Figure 4 shows the regression of monthly HF returns by strategy against the same key market indices used previously. The results tell two distinct stories. The annualised excess returns across all strategies from 1993 to 2011 were robust. Since then, however, excess returns across all strategies have declined. The results are particularly striking for Equity Hedge and Event Driven strategies, which provided investors 810 bps and 660 bps of alpha respectively in the earlier period while they created negative annualised excess returns from 2011 – May 2016 (i.e., Equity Hedge generated -200 bps of alpha and Event Driven -80 bps).
Indeed, as Barclays politely puts it, "regardless of the way we analysed recent HF performance, it appears that HFs have been underperforming lately."
But why? What do hedge funds blame their own underperformance on: that is the question Barclays posed to its investors.
- Figure 5 shows that the most common driver perceived to adversely impact recent HF performance (chosen by almost 75% of investors) has been that the HF industry has become too big relative to the opportunities available.
- More than half of respondents (57%) indicated that macro conditions (e.g., political, central bank, etc.) have contributed to the underperformance.
- The balance of factors (i.e., changes to market behaviour, ‘high’ fees, ‘other’) suggest that there are many other perceived explanations for the poor results, though these are much less widely held.
- One of the key considerations when discussing underperformance of HFs is the reference point for investors’ expectations. As one interviewed investor mentioned, the HF market has changed considerably (e.g., new regulations, monetary intervention, increased operational burden, etc.) and it might make sense to change the expectations of HFs commensurately. While this will not relieve HF managers from the pressure of pushing fees down for instance, it might change the conversation with investors regarding what the objectives / expectations of an allocation should be.
A detailed look at each of these excuses:
HF AUM growth
Regarding the most common driver referenced above, ‘the HF industry has become too big relative to the opportunities available’, we decided to evaluate how the HF industry has grown. Figure 6 shows us that the overall CAGR for HF AUM from 2009 – 2015 was 10% and that the individual strategy components each went up by between 9% and 12%. Across the various strategies, it appears that, on average, asset growth in the individual funds account for two-thirds of the overall growth by strategy while the number of new funds accounts for only one-third. This suggests that, on average, HFs are growing their AUM quite considerably and that the bigger HFs are in fact getting bigger.
Despite the HF industry’s significant growth since 2009, it is still very small relative to the pool of global financial assets. Figure 7 shows that the estimated value of all financial assets in 2015 was $305tn and that HFs at $2.9tn account for just under 1% of the total. Although over the period HFs had a CAGR of ~10% they only increased their portion of the pool by about 300 bps, from 0.7% to 1%. Furthermore, Public Equities and Public Debt had CAGRs of 7.5% and 6.6% versus 4% for all financial assets and therefore did a much better job of keeping pace with the growth in HFs. This suggests that the issue is likely not the growth in size of the overall HF industry, as there appears to be an ample supply of assets. The issue may be, however, the growth in size of many individual HFs, which are pursuing similar strategies leading to crowding.
While there are many advantages that larger HFs have (e.g., access to talent, institutional infrastructure, etc.), there are also several drawbacks. One such drawback is that as HFs become larger, their investable universe can often be diminished (e.g., due to position limits) as it is often not ‘worth it’ to invest in smaller situations that can hardly move the PnL needle. Figure 8 illustrates how since the middle of 2012 position crowding in US Equities has increased significantly – rising from 2.2% at 3Q12 to 5.2% in 3Q15. Historically, investing in crowded names has generated positive returns, particularly in stable, rising markets. However, when the reverse happens, it tends to be sharp and painful. The reverse, in this case, started in 3Q15, with indices composed of crowded names dropping significantly. We discussed this phenomenon in our recent piece ‘Hanging Tough: Landscape and Recent Developments in Event Driven Strategies’.
What is worthwhile to mention here is how this coincides with a significant underperformance of larger funds, evident in the difference between the 12-month rolling returns of the HFRI Fund Weighted (i.e., equally weighting all funds) and Asset Weighted (e.g., all funds are not counted equally) indices. In Figure 9, we can see that from July 2011 through July 2015 the results were variable and inconsistent, whereas since July 2015 the pattern seems to be much clearer and it appears to be trending down across strategies. This suggests that the Asset Weighted indices significantly underperformed the Fund Weighted indices. The most severe decline comes from Event Driven where the asset weighted index underperformed the fund weighted index by nearly 6% over the last 12 months through June 2016.
The second most commonly mentioned driver of underperformance by investors was that macro conditions have been working against HFs. Macro managers have been mentioning this issue for a while, but Figure 10 appears to also confirm this thesis with regard to Equity strategies. The chart on the left shows the amount of alpha HFs were able to generate in different market conditions from 2000 – 2015, specifically when there was high and low levels of realised correlation (e.g., the degree to which different stocks in the S&P 500 move in the same direction) and of dispersion2 (e.g., the difference in stock movements regardless of whether they are going in the same direction or not). HFs generate almost 10% of alpha when dispersion is high and correlation is low, conversely, when dispersion is low and correlation is high HFs only generate 0.8% of alpha. The charts on the right show how correlation and dispersion have moved over time and it appears that we have been experiencing an adverse scenario for HFs over the last two years or so – high correlation and low dispersion.
Thus, it is apparent that the respondents in our survey have a pretty clear grasp of the issues facing HFs as there appears to be evidence behind the two common causes of the recent underperformance cited by investors.
* * *
Excuses or not, the reality is that the infatuation with hedge funds is fading fast. Barclays predicts that based on recent HF performance and the increased challenges to launching an HF, "we estimate that there would be a net decrease in the number of funds by YE 2016." This means that 2016 will be the first year since 2009 with net outflows from the HF industry.
HF launches and closures
In order to estimate the projected levels of HF launches and closures, we took a look at historical results. Figure 17 depicts the percentage of HF launches and liquidations (the number of each relative to the existing total number of funds) as well as the estimated percentage of launches and liquidations for the full year of 2016. According to HFR, the number of new HF launches and liquidations (as a percentage of the existing HFs) has been fairly stable from 2010 through 2015 (other than 2011). Over this time frame, on an annualised basis, new launches were about 12% – 13% and liquidations were about 10%, which results in 2% – 3% additional new funds annually. In 2016, based on our ongoing conversations with managers, investors, and external research, we expect the launch rate to decline to ~8% and the liquidation rate to rise to ~12%, resulting in a net decline in the number of HFs by ~4% (i.e., ~340 funds). There have been several established HFs with relatively long track records, particularly in the Fundamental Equity L / S space, that have shuttered their doors recently in addition to the normal attrition across the less prominent, newer HFs. As a point of reference, in 2008, when the industry was under severe stress, the liquidation rate increased to 21%.
Industry size in 2016
In ‘Bracing for Impact’, published earlier this year, we estimated that 2016 will be the first year since 2009 with net outflows from the HF industry. To be precise, we estimated $30bn of net outflows with a range between zero and $50bn. To assess at what AUM level the industry will settle at the end of 2016, we also need to take performance into consideration. Our base scenario is that performance will be muted (0% – 4% return) and therefore the industry will likely end up in 2016 with roughly $2.9tn AUM, roughly the same (or slightly higher) compared to the beginning of the year. In Figure 18, we look at three potential scenarios for YE 2016 based on the estimated flows and HF returns:
- Optimistic – HF performance rises to a healthy range of 4% – 8% in 2016 and flows are flat. Overall, AUM is likely to rise by $180bn to $3.1tn, driven by performance gains.
- Baseline – HF returns improve from 2H15 and 1Q16 levels, but only to a low positive level (i.e., 0% – 4% return for 2016), while investors redeem a net of $30bn, resulting in a slight overall increase in AUM from a year ago.
- Pessimistic – Industry returns deteriorate (i.e., -4% – 0% return for 2016) and net outflows increase to $50bn, which would result in industry assets declining by over $100bn for YE 2016.
For those who are not quite ready to write off hedge funds as a dying industry, here are the HF strategies currently preferred by investiors (a good idea may be to avoid the most overexposed ones).
Investors’ HF strategy preferences
Our final analysis was derived completely based on investors’ feedback with regard to their strategy preferences for 2016. Figures 19 and 20 show our respondents’ intentions with regard to increasing / decreasing their allocations to various HF sub-strategies for the second half of 2016. The increase versus decrease preference difference is listed on the right hand side of the respective figures and we use this as a crude proxy for investor interest. The arrows on the far right hand side of the chart show the change in sentiment toward the sub-strategy between 4Q15 and 2Q16 (e.g., a blue arrow shows an increase in the level of interest in 2Q16 versus 4Q15). Key takeaways:
- It appears that systematic strategies, Systematic / CTA (+25%) and Quant Equity (+24%), are not only the most in favour (on a net basis) over the next six to 12 months, but also are the strategies that gained the most in investor interest since the end of 2015.
- Equity Market Neutral, Equity Emerging Markets, Macro, and FIRV all had net positive interest at 2Q16, but appear to have less net interest from investors since YE 2015.
- Meanwhile, Equity L / S and Event Driven have the lowest level of investor interest (on a net basis) and have actually seen the level of interest decline since YE 2015.
In conclusion, here are Barclays' parting thoughts and considerations for HF managers and investors:
- 2015 was a tough year for HF managers and 2016 continues to be challenging.
- While there are some exceptions, HFs in general continue to face performance challenges.
- Although there are some diversification benefits to adding HFs to liquid portfolios, the industry’s recent underwhelming performance has heightened investors’ concerns about their HF allocations – as evidenced by recent industry headlines.
- Investors are clearly dissatisfied with HF performance, but it does not seem that they have given up on HFs.
- While more than half of surveyed investors said they are dissatisfied with the performance of HFs in their portfolios, only a minority are looking to reduce their HF allocations. Most are looking to reallocate to other HFs, seek fee discounts from their current HFs, or just leave their current allocations unchanged.
- Many investors are also seeking alternative ways to improve their return profile or terms (e.g., by investing with emerging or small managers).6
- Investor interest in certain strategies remains strong (e.g., Systematic / CTA, Macro, Distressed, Emerging Markets).
- Our baseline estimate is that the size of the industry in 2016 (versus 2015) is likely to remain unchanged at $2.9tn.
- Our analysis shows that HF performance impacts flows with a lag of one to three quarters, which we believe will result in net outflows for 2016, at least in part, to reflect the performance challenges of 2H15 and early 2016.
- However, our projection of a modest increase in HF returns (to around 0% – 4%) will likely result in assets remaining flat versus 2015 as performance and flows balance each other out in 2016.
- That said, there are also optimistic and pessimistic scenarios which could see the industry grow assets by $180bn or see a reduction in assets of $110bn, respectively.
However, when one cuts through the chase, what the above really boils down is two things: over the past 5 years, it has become far more difficult to trade on inside information following the SAC debacle, but more importantly, central banks have grotesquely manipulated markets to the point where no fundamental analysis matters, and where only quant strategies (i.e. HFT frontrunning) remain profitable. As for all those other millionaires suddenly finding themselves in a Darwinian fight for survival in this paradoxical new normal, our condolences, violins and all.