1
Chapter 4:
Does Manager Offshore Experience Count in the Alternative UCITS Universe
*?
B. Dewaelea, I. Markovb, H. Pirottec, N. Tuchschmidd Abstract
This article examines the performance of alternative Ucits on the basis of managers’ offshore experience. We compare Ucits in terms of their alpha, i.e. their added value over and above that of the market as captured by Fung and Hsieh’s [2004] eight regressions factors. Managers with offshore experience are defined as companies managing offshore hedge funds in addition to managing Ucits.
For a sample period from 2008 to 2011, we find that for Ucits with such managers there is some evidence, although weak, of the added value of offshore experience. On the other hand, we find no evidence of added value in the case of non-offshore experienced managers. Motivated to further refine our results, we separate Ucits with offshore experienced managers into two groups: those with equivalent offshore hedge funds (replicas) and those without (new funds). This time, Ucits with no offshore equivalents show low volatility and a strongly positive alpha. Ucits with offshore equivalents on the other hand bring no added value and, not surprisingly, bear no substantial differences in their risk profile with their paired funds offshore. Offshore experience therefore plays a significant role in creating positive alpha, as long as it translates into real innovations. That could be explained by the fact that a Ucits structure brings in additional costs. If the fund is a pure replica, this cost represents a handicap that is hardly compensated.
* We would like to thank Alix Capital for providing access to their alternative Ucits database. We would especially like to thank Mr. Louis Zanolin and Mr. Dravasp Jhabvala of Alix Capital for providing much
additional information, help and insightful comments, Mr. Daniel Mosca for his dedicated work on our database, Mr. José Galeano and Mr. Fayçal Zerigue of SYZ Asset Management for their research assistance.
a Centre E. Bernheim, Solvay Brussels School of Economics and Management, Université libre de Bruxelles, av. F.D. Roosevelt 21, CP 145/01, B-1050 Brussels, Belgium.
b Research assistant at Haute Ecole de Gestion de Genève, Campus de Batelle, Bâtiment F, 7 route de Drize, 1227 Carouge, Switzerland. Contact author : [email protected]
c BNP Paribas Fortis Professor of Finance at Centre E. Bernheim, Solvay Brussels School of Economics and Management, Université libre de Bruxelles, av. F.D. Roosevelt 42, CP 144/03, B-1050 Brussels, Belgium
d Partner at Tages Capital LLP, 100 Pall Mall, London, SW1Y 5NQ, United Kingdom
2
4.1 Introduction
The European Union’s (EU) Directive on Undertakings for Collective Investment in Transferable Securities (UCITS) was a framework started in 1985 and aimed at creating a single European market for the distribution of investment funds registered in a particular EU member state. The framework contained various provisions on diversification, leverage, liquidity and the use of derivatives with the purpose of instilling confidence and maintaining a high level of investor protection.
The severe restrictions of the initial Ucits framework, however, made it a qualified success and not many Ucits-compliant funds were launched. Anderberg and Bolton (2006) point that some of the restrictions seemed sensible, for example the restriction on investing in precious metals, but others were less so, for example the rule preventing the holding of cash and money market instruments (MMI), which effectively ruled out Ucits cash funds. In the early 2000s the European Commission (EC) issued the Management Directive and the Product Directive, collectively known as Ucits III, which provided for the smoother cross-border marketing of Ucits and permitted a broader spectrum of eligible investment instruments and hedge fund-like strategies in the Ucits framework.
In 2004, the EC gave the Committee of European Securities Regulators (CESR, replaced by the European Securities and Markets Authority - ESMA on 1 January 2011) a mandate to publish guidelines on what constitutes eligible investments (CESR (2008)). Murphy and O’Shea (2008) summarize that depending on the jurisdiction and subject to strict risk management provisions, Ucits can invest in assets such as structured financial instruments as transferrable securities, transferrable securities embedding derivatives, closed-ended funds as transferrable securities, financial derivative instruments (hedge fund indices, credit derivatives, commodity indices, 130/30 strategies) and MMI (also bank loans as MMI in Ireland). Anderberg and Brescia (2009) add that the Ucits IV directive should give Ucits even more latitude with the ability to establish master-feeder structures and perform cross- border mergers and should further ease the cross-border marketing of funds.
The present article is concerned with the performance of Ucits III funds that employ hedge fund-like strategies, the so-called Ucits hedge funds, alternative Ucits1 funds or Newcits.
Some of them are established by traditional offshore hedge fund managers, while the majority is run by non-offshore experienced managers. In referring to managers here, we imply the management companies, not the persons managing the funds. The purpose of this research is to analyze whether concurrent or previous manager offshore experience has an impact – and, if confirmed, in which direction – on the performance of alternative Ucits. Our analysis tests for the so-called skill factor, in that offshore experience has a positive effect on implementing hedge fund investment strategies under a Ucits wrapper, versus the alternative where offshore experienced managers would use a Ucits structure to access a new market after a mediocre track record.
We compare our alternative Ucits samples by building equally-weighted monthly- rebalanced indices and computing each index alpha, i.e. its added value over and above that
1 We refer to these funds as alternative Ucits or simply Ucits when there is no danger of confusion.
3
of the market as captured by Fung and Hsieh’s (2004) eight regressions factors. What we find is that for a sample period from June 2008 to May 2011, alternative Ucits with offshore experienced managers have positive alpha2 with a p-value3 of 0.12 which provides weak evidence of superior skills. Deeper analysis of the Ucits with offshore experienced managers, however, shows that their sample is not homogeneous. Their performance is driven by funds that do not explicitly track offshore equivalents. These are new funds started directly on the Ucits framework rather than being replicas of existing offshore products. They provide positive significant alpha throughout the sample period.
The rest of the article is organized as follows. First we discuss the Ucits directives, how they affect investment strategies and how they are perceived by investment professionals. Then we offer a brief overview of relevant academic research on alternative Ucits and hedge funds. Next we present our database and our methodology which are followed by a comparative analysis of the added value of offshore experienced and non-offshore experienced managers. In the penultimate section we provide some risk metrics of the samples. Finally, we offer our concluding remarks.
4.2 Market Analysis
Most of the existing literature on Ucits focuses on several main areas of interest – the historical development of the Ucits legal framework and its influence, the growth of the industry, and the constraints on hedge fund-like strategies. Very few studies conduct statistical analyses on the performance of alternative Ucits or comparisons of the latter to the performance of hedge funds in general. The scarcity of such literature should not be a surprise as the average industry history is relatively short. On the other hand, however, there is a myriad of journal, magazine and newspaper articles which offer diverging views on the current state and the outlook of the industry. We present these views as a confirmation of the need of a more in-depth analysis of the alternative Ucits industry. We also describe the main provisions of the Ucits framework and how they pertain to hedge fund-like strategies.
As summarized by Tuchschmid, Wallerstein and Zanolin (2010), in most jurisdictions Ucits are not allowed to invest in commodities and commodity certificates. Commodities should be accessed through indices. Investments in private equity and real estate are also prohibited.
Ucits are, however, allowed to invest in non-eligible assets as long as those do not amount to more than 10% of the total portfolio, the so-called trash ratio.
Even within the realm of eligible instruments, however, Ucits are constrained by the types of investment strategies they can use. Markov and Tuchschmid (2011) outline the main provisions of the Ucits directives and their impact on the suitability of different hedge fund strategies under the Ucits framework. Long/short equity strategies, for example, are constrained by the restrictions on short selling. The use of synthetic short selling entails additional costs such as the need for active collateral management due to the restriction on counterparty risk. Moreover, the difficulty in establishing a synthetic short position in non-
2 The constant term (or unexplained returns) measured by a multifactor regression (Fung and Hsieh [2004]).
3 Due mostly to the small sample problem, relative to the heterogeneity of the data and the need to subsample for the purposes of our study.
4
equity instruments makes strategies such as fixed income arbitrage very difficult under the Ucits framework. More generally, Gabbert (2005) points out that under the Ucits III regulations, Ucits can attain a maximum short exposure of 30%, the so called 130/30 strategy, which is not enough to replicate a matched-pair strategy where one security is held long and another one is held short. In essence, only 60% of the portfolio (30% long and 30%
short) can be traded on a matched-pair basis.
Beaudoin and Olivier (2010) moreover explain that the strict diversification requirements affect hedge fund strategies that presuppose more concentrated portfolios, such as emerging markets funds for example. Leverage requirements also have an impact on strategies, such as fixed income arbitrage, which may involve highly leveraged positions. The strict liquidity requirements imposed on Ucits were beneficial during the financial crisis of 2008-2009, but they also carry a cost. As Beaudoin and Olivier (2010) summarize it, Ucits are effectively barred from investing in less liquid securities such as microcaps and distressed securities or long-term strategies such as deep value investing. Convertible arbitrage strategies may also be difficult to employ because of inefficient pricing due to market illiquidity. Ucits must therefore focus on the most liquid part of the market, which offers fewer opportunities. The liquidity requirement therefore has a price.
Many of the abovementioned restrictions can be overcome though. As Markov and Tuchschmid (2011) point out, an alternative Ucits fund may replicate an offshore hedge fund’s performance through the use of contracts for difference (CFD) or total return swaps (TRS). Such replication strategies are not risk-free, however. First of all, they can be detrimental to the fund itself with investors becoming displeased with possibly large tracking errors between the onshore and the offshore funds. Secondly, as Danaher (2010) observes, even though technically allowed, such strategies are outside the realm of transparency endorsed by the Ucits directives. Johnson (2011) comments that “the emergence of ever more complex and esoteric vehicles” may tarnish the Ucits brand. He gives the example of the launch of a precious metals Ucits fund, 82% of whose portfolio is invested in just 4 metals. He also cites professional opinions that many of the smaller fund managers are pushing the Ucits envelope and not abiding by the rules 100%. What is even more worrying is that some regulators seem more interested in providing regulatory cover for such funds after a first “barrier check” than in ensuring the original objective of investor protection during the life of the fund.
In response to these developments, ESMA published a discussion paper (European Securities and Markets Authority (2011)) treating the possible impact on investor protection and market integrity that complex Ucits pose through their use of hedge fund-like investment strategies previously prohibited to them. ESMA also set out to examine the possible measures to mitigate these risks which go as far as warnings to retail investors or limits on the distribution of certain products to retail investors.
With regard to commodity indices, ESMA is pushing for stricter diversification rules both in terms of component weights and commodity composition. A strategy index must also be a benchmark for its reference market and not simply an investment strategy wrapped in an index. The rebalancing frequency should be such that it does not prevent investors from being able to replicate the index. Intra-day or daily rebalancing clearly does not satisfy this criterion. Ucits must also ensure that any valuation of the swap and the financial index
5
must include an independent assessment. ESMA’s cost-benefit analysis of the proposed changes and their potential impacts shows that some structured Ucits would not be in compliance and therefore would shut down.
Despite these controversies, according to HFN Industry Research (2010), the growth of alternative Ucits has been significant in recent years, especially 2009 and 2010 following the havoc that the financial crisis wrought in the hedge fund industry. Differing opinions exist as to whether this will be a sustained growth backed by the high investor protection that the Ucits directives offer or a transitory event. John Bohan of Apex Fund Services says that the demand for offshore products based in the Cayman Islands and Bermuda will persist as those centers will retain more flexibility but that they nevertheless need to increase their level of regulation (HFMWeek (2010)). Dominique Lecocq of Lecocqassociate adds that the EU AIFM directive, which will harmonize the management of non-Ucits hedge fund managers with AUM in excess of 100 million EUR, is expected not to increase the demand for Ucits (HFMWeek (2010)). It is expected to give investors sufficient comfort that the hedge fund industry is subject to proper regulatory surveillance. Investors may thus prefer hedge funds to paying higher TER ratios for Ucits employing costly OTC swaps to access strategies that are otherwise ineligible under the Ucits framework (HFMWeek (2010)).
In this line of thought, Harris (2010) suggests that the recession has been the real reason for the slowdown of inflows into the hedge fund industry. A report by Hedge Fund Research (2011)) may serve as a confirmation. According to this report, in Q1 2011 the total capital invested in the global hedge fund industry surpassed 2 trillion USD for the first time in history. There is, therefore, evidence that managers are not dropping the traditional hedge fund model and are in fact returning to it as the grip of the financial crisis is softening or as the regulatory scrutiny on banks hardens pushing creative managers to join funds.
Harris (2010) notes that more and more managers are establishing parallel structures, i.e.
Ucits as complements to offshore hedge funds instead of replacements. In addition, a report by RBC Dexia and KPMG reveals that in establishing parallel structures, Ireland’s Qualified Investment Funds (QIFs) and Luxembourg’s Specialized Investments Funds (SIFs) are gaining popularity versus the Ucits format (HedgeWeek (2011)). QIFs and SIFs are said to offer more flexible liquidity and transparency rules.
The growth of the alternative Ucits industry should not be underestimated though.
According to Alix Capital4, a Geneva-based investment boutique, from the beginning of 2006 to May 2011, the total number of funds increased more than 6-fold and the total AUM increased more than 10-fold from 10.8 billion EUR to 112.3 billion EUR. For the last two years of this period, the total AUM increased more than 3-fold, compared to that of the global hedge fund industry, which increased by a little more than 50%. In addition, the Ucits brand has become globally recognized.
Even though Ucits are not necessarily conservative investment vehicles, they are perceived as safer compared to hedge funds. In addition, the establishment of alternative Ucits by large hedge fund managers such as Man Group, Brevan Howard and Paulson & Co may serve as a sign that this is not a passing trend. The efficiency gains to be brought by the Ucits IV directive are expected to have a positive impact on the European Ucits market. As
4 Alix Capital’s website: http://www.alixcapital.com
6
already mentioned, the establishment of master-feeder structures, the possibility of cross- border mergers and the facilitation of cross-border marketing will further streamline the European Ucits market. As far as the EU’s AIFM directive is concerned, there is evidence that the level of uncertainty that has surrounded it coupled with manager impatience about its final provisions have worked in favor of the number of Ucits being established (Varriale (2011)).
4.3 Literature Review
One of the first studies to examine manager offshore experience is Agarwal, Boyson and Naik (2009). The authors conduct a comprehensive analysis of the performance of hedged mutual funds, which bear many similarities to alternative Ucits. Alternative Ucits are Ucits that employ hedge fund strategies but are regulated by the European Securities and Markets Authority (ESMA). In the same vein, hedged mutual funds are mutual funds that employ hedge fund strategies but are regulated by the US Securities and Exchange Commission (SEC).
Agarwal et al. (2009) test three hypotheses comparing hedged mutual funds to hedge funds and traditional mutual funds. The regulation and incentives hypothesis states that hedge funds should outperform hedged mutual funds due to the lighter regulatory environment in which the former operate and their performance-based compensation. The strategy hypothesis posits that hedged mutual funds should outperform traditional mutual funds because of their flexibility to employ more diverse investment strategies and profit from both long and short positions. The skill hypothesis predicts that hedged mutual fund managers with concurrent or previous hedge fund experience should perform better than hedged mutual fund managers without hedge fund experience. For a sample period from 1994 to 2004, Agarwal et al. (2009) find evidence in support of all three hypotheses. Regarding the last hypothesis, the authors point that hedged mutual fund managers with hedge fund experience should be more adept at implementing hedge fund strategies under a mutual fund structure.
Darolles (2011) tests the Agarwal et al. (2009) skill hypothesis on a sample of alternative Ucits from the Morningstar database covering the period June 2004 to May 2011 and finds that Ucits with offshore experienced managers outperform Ucits with non-offshore experienced managers. Unlike Darolles (2011), however, in our study we calculate the Fung and Hsieh (2004) alpha of both groups of funds in order to extract the true manager added value in both cases.
The analysis of the alternative Ucits space cannot be complete without a comparison to the hedge fund industry. After all, our goal is to analyze the effect of manager offshore experience on the performance of alternative Ucits. Tuchschmid et al. (2010), one of the first academic studies to compare alternative Ucits and hedge funds, finds mixed results as far as the alpha of alternative Ucits in comparison to that of hedge funds is concerned. For a sample period from 2006 to 2010 the Ucits Alternative Global Index5 outperforms the HFRI
5 A leading comprehensive equally-weighted alternative Ucits index provided by Alix Capital:
http://www.ucits-alternative.com.
7
fund of funds and the HFRX Composite Index6 but underperforms the HFRI Composite Index7. For a shorter sample period starting in 2008 the Ucits Alternative Global Index outperforms the HFRI composite index as well. It should be pointed out that the second period is much more affected by the financial crisis. The authors find significant differences in the investment opportunity set of alternative Ucits and hedge funds, with the latter having significantly higher dispersion of return and risk. The authors also find that large differences exist with regard to strategy level. In some strategies, hedge funds outperform alternative Ucits in terms of absolute return, while in others hedge funds perform worse. In general, hedge funds appear to be a much more heterogeneous group than alternative Ucits.
In an extension of this work, Tuchschmid and Wallerstein (2013) could confirm the absence of significant lag between the performance of Ucits and hedge funds even though the former showed generally lower risk or volatility than their less-regulated counterparts.
In a recent book chapter, Markov and Tuchschmid (2011) point out that during the recent financial crisis alternative Ucits perform better on average, which could be attributed to the stricter risk management process and the enhanced liquidity requirements. This is in line with the aforementioned findings that during the crisis period hedge funds performed worse compared to more conservative investment vehicles. During the period April 2010 to March 2011, however, hedge funds outperform alternative Ucits in 6 out of the 10 strategy classes under investigation and the global index. These include long/short equity and fixed income, two of the strategies with the highest value of AUM of alternative Ucits.
Tuchschmid et al. (2010) and Markov and Tuchschmid (2011) analyze the performance of alternative Ucits vs. hedge funds but do not subsample within the alternative Ucits space.
In this article, we test Darolles’ (2011) results on manager offshore experience though a different methodology and on a larger sample of funds. The Fung and Hsieh regressions we perform allow us to measure the true added value of managers with offshore experience.
Moreover, we further analyze offshore experienced managers as their sample appears not to be homogeneous. Their performance appears to be based on the types of funds they manage – whether these are new alternative Ucits funds or replicas of already existing offshore vehicles.
4.4 Data
Before describing our database, we present a visual comparison of the alternative Ucits industry, the hedge fund industry and the equity markets. Figure 4.1 presents the cumulative excess returns8 of four indices: the Ucits Alternative Global Index, the HFRI Fund Weighted Composite Index, the S&P 500 and the Euro Stoxx 50, from December 2005 to May 2011. It is obvious that the financial crisis of 2008-2009 had a strong impact on the relative performance of the indices. Prior to 2008, the equity indices outperform both the alternative Ucits and the hedge fund index but afterwards they drop precipitously. In fact, both equity indices have a negative cumulative excess return at the end of May 2011. The hedge fund index performs better compared to the Ucits index in the beginning of the
6 An investable hedge fund index provided by Hedge Fund Research.
7 A non-investable benchmark hedge fund index provided by Hedge Fund Research.
8 Over the Libor rate of the respective currency.
8
sample period. In July 2008 the Ucits index starts dominating but in August 2010 the hedge fund index emerges on top again.
During the worst of the recession, the Ucits index, probably in part due to the industry’s relatively more conservative nature in a more regulated environment, dominates both the hedge fund index and the two equity indices. Moreover, it is the least volatile during the sample period. But later on, the hedge fund index recovers with a much faster pace to provide a cumulative excess return of 6% at the end of May 2011 compared to that of 0.1%
for the alternative Ucits index.
Figure 4.1 Alternative Ucits vs. hedge fund and equity indices
This figure presents the cumulative excess returns of the selected indices when the end of December 2005 is taken as a base of 100%. The sample period ends at the end of May 2011.
To carry out our analysis we used a database provided by Alix Capital and consisting of the monthly net-of-fees returns of 550 alternative Ucits funds which charge both management and performance fees and actively take short positions. They are run by a total of 259 different managers whose expertise in offshore funds has been individually analyzed. The analysis was carried out by cross-referencing the managers to managers in the BarclayHedge and Lipper TASS hedge fund databases, by additional internet searches, by consulting Alix Capital and by contacting some of the managers. The breadth of the search was motivated by our desire to find all the management companies in the alternative Ucits database that also manage offshore hedge funds regardless of age of the hedge funds. We are therefore confident that our database is as complete as possible.
Table 4.1 gives the top 5 countries of domicile of the alternative Ucits, henceforth simply Ucits, in our database. As shown in the table, Luxembourg is the preferred Ucits domicile.
Table 4.2 moreover provides some statistics pertaining to the Ucits and their managers.
40,00%
50,00%
60,00%
70,00%
80,00%
90,00%
100,00%
110,00%
120,00%
130,00%
déc.-05 mars-06 juin-06 sept.-06 déc.-06 mars-07 juin-07 sept.-07 déc.-07 mars-08 juin-08 sept.-08 déc.-08 mars-09 juin-09 sept.-09 déc.-09 mars-10 juin-10 sept.-10 déc.-10 mars-11
Ucits Alternative Global HFRI Fund Weighted Composite
S&P 500 Euro Stoxx 50
9 Table 4.1
Alternative Ucits domiciles
This table presents the top five countries of domicile of the alternative Ucits in our sample.
Country of domicile Number
Luxembourg 257
Ireland 107
France 98
United Kingdom 38
Germany 21
Table 4.2
Alternative Ucits and manager statistics
This table presents the number of alternative Ucits and the numbers of managers in total and by offshore experience in our sample.
Funds and managers Number
Alternative Ucits 550
Managers 259
Offshore experienced managers 81 Non-offshore experienced managers 178
Table 4.3
Strategy breakdown of alternative Ucits by manager offshore experience
This table presents the strategy breakdown of alternative Ucits with offshore experienced and non-offshore experienced managers.
Strategy Managed by offshore
experienced managers
Managed by non- offshore experienced
managers
CTA 11 20
Macro 33 106
Long/Short Equity 57 80
Equity Market Neutral 10 29
Event Driven 9 5
Fixed Income 28 43
Emerging Markets 21 24
Commodity 1 13
FX 8 29
Multi-Strategy 4 19
Total 182 368
10
Table 4.3 presents the strategy breakdown of Ucits in terms of manager offshore experience.
Table 4.2 and 4.3 indicate that 81 offshore experienced managers manage 182 Ucits and 178 non-offshore experienced managers manage 368 Ucits. Table 4.3 moreover shows that the number of funds in some strategies is small. Results in the empirical sections, therefore, need to be viewed in light of this.
Our database should not be significantly affected by the usual hedge fund database biases – the survivorship bias, the instant history bias and the selection bias. The instant history bias is controlled by the fact that Ucits are required to publish their performance with the regulator (ESMA) on a regular basis. As far as the survivorship bias is concerned, our database contains Ucits that defaulted after mid-2009. This is due to the fact that Alix Capital started collecting data in 2009 and their database contains no funds that ceased operations prior to June 2009. As shown by Tuchschmid et al. (2010), however, Ucits have exhibited a very low attrition rate. Coupled with the fact that our sample period starts in mid-2008, we are convinced that the survivorship bias does not significantly affect our database. The third bias is controlled by the fact that the funds in the Alix Capital database have to comply with certain criteria in order to be included. In particular, they have to ask for both management and performance fees and actively take short positions (and not just claim they could). The compilation does not rely on the voluntary disclosure or contribution by funds as is the case with most hedge fund databases. And there are no other ex-ante conditions or requirements for the inclusion of funds in the database.
4.5 Methodology
We analyze the differences between Ucits with offshore experienced and non-offshore experienced managers both in terms of the alpha, or manager added value, of each group and in terms of their risk characteristics. Even though we compare collections of funds, we refer to the alpha in this case as manager added value, not fund added value, since we split the sample according to the managers’ offshore experience. To perform the comparisons we combine the funds of each group into equally-weighted monthly-rebalanced indices. A particular issue in our database is that due to the short history of many of the Ucits, our indices may initially have a smaller number of funds and then funds are added and the index is reweighted as new return data appears. We chose this methodology over one where we take a sample period for which all (or the majority of) funds have available return data or one where we select the funds the length of whose return series is equal to or longer than a predefined sample period length. The first case would result in a very short sample period while the second case would leave a very small number of funds. The short sample period of the return series can be explained by the fact that the boom in the launch of alternative Ucits started in 2009 and the industry therefore has a very short average history.
In order to conduct a coherent comparison between indices of funds denominated in different currencies, we expressed all return series as excess returns over the 1-month Libor rate of the respective currency. Using the funds’ excess returns in their own currency is equivalent to comparing investment projects where the investor is fully hedged towards currency risk.
Thus our comparison lingers on the usual covered interest rate parity assumption being perfectly verified.
11
To measure the alpha of each index, we used the Fung and Hsieh 8-factor model which includes the original 7 asset-based style factors from Fung and Hsieh (2004) as well as the emerging markets factor they suggested later. As we use series of excess returns over the Libor rate, the 8 regression factors from Fung and Hsieh, where appropriate, have also been expressed in excess of the 1-month USD Libor.
The resulting factors thus are: (1) the S&P 500 excess monthly total returns, (2) the difference between the Russell 2000 and the S&P 500 monthly total returns, (3) the change in the difference between the 10-year Treasury constant maturity yield and the 1-month Libor (month end-to-month end), (4) the change in Moody’s Baa yield over the 10-year Treasury constant maturity yield (month end-to-month end), the excess returns to lookback straddles on (5) bonds, (6) currencies and (7) commodities and (8) the MSCI Emerging Market excess monthly total returns9. The eight regression factors could best explain our choice to work with Ucits indices as opposed to samples of individual Ucits. Since many of the Ucits have relatively short histories, it would be unsuitable to apply an eight-factor regression to their returns. Nevertheless, the Appendix provides further results on a per- fund basis for those Ucits covering the period from June 2008 to December 2010.
As we are only interested in the alpha of each index, analysis of the betas is not included.
We should also mention that we performed Engle and Ljung-Box tests for residual heteroskedasticity and residual autocorrelation for various numbers of lags in the test statistics. The tests indicate that the latter are not issues in our regressions. The results should nevertheless be viewed with care due to the small number of observations.
4.6 Alpha Estimation
The comparison of Ucits based on manager offshore experience tests for the so-called skill factor in implementing hedge fund investment strategies in a Ucits framework. This comparison is similar to the one conducted by Agarwal et al. (2009) in which they compare hedged mutual fund managers with concurrent or previous hedge fund experience and those without. They find that hedge-fund experienced managers perform significantly better than non-hedge fund experienced managers.
Our sample is European (by definition) and much larger since it contains 182 Ucits with offshore experienced managers and 368 Ucits with non-offshore experienced managers10. At the time of writing, data on the lookback straddles used in the Fung and Hsieh regressions was only available through December 2010, which explains the sample period’s end in this section.
Table 4.4 compares the two samples for a period of 5 years from January 2006 to December 2010. However, since the alphas we obtain are not significant, we cannot conclude on the offshore manager added value during this period.
9 The hedge fund trend-following factors (5,6,7) based on Fung and Hsieh [2001] were collected from David Hsieh’s website: http://faculty.fuqua.duke.edu/~dah7/DataLibrary/TF-FAC.xls.
The rest of the data was obtained from Datastream and the Federal Reserve’s website:
http://www.federalreserve.gov/.
10 Compared to only a few dozen hedged mutual funds in Agarwal et al. [2009].
12 Table 4.4
Manager offshore experience (2006-2010): monthly alpha
This table presents monthly alphas with 90% confidence bounds and adjusted R-squared for Ucits with offshore experienced (182 funds) and non-offshore experienced (368 funds) managers. p-values are provided in brackets.
January 2006 – December 2010 90% Confidence Interval
Monthly Alpha Lower Bound Upper Bound Adj. R-Squared W/o offshore exp. -0.06%
-0.16% 0.04% 0.77
(0.34) With offshore
exp.
0.01%
-0.11% 0.13% 0.80
(0.87)
In an attempt to find influences induced by the recent financial crisis, we contracted the sample period to June 2008 – December 2010 and performed the same regressions on the remaining 31 observations for each index. In Table 4.5, we see that the alpha of Ucits with offshore experienced managers is higher and is the only one with a p-value close to the 10%
significance level, at 0.12. This is a very important observation as it tells us that over the 31- month period ending in December 2010, which is undoubtedly marked by turbulent markets, there is some evidence that offshore experienced managers exhibit skill. Similar to Agarwal et al. (2009), we can infer that offshore experience appears to affect performance, even when the managed vehicle itself is not an offshore one.
Table 4.5
Manager offshore experience (2008-2010): monthly alpha
This table presents monthly alphas with 90% confidence bounds and adjusted R-squared for Ucits with offshore experienced (182 funds) and non-offshore experienced (368 funds) managers. p-values are provided in brackets.
June 2008 – December 2010 90% Confidence Interval
Monthly Alpha Lower Bound Upper Bound Adj. R-Squared W/o offshore exp. 0.05%
-0.06% 0.17% 0.88
(0.42) With offshore
exp.
0.15%
-0.01% 0.31% 0.89
(0.12)
Despite being important, statistically speaking this evidence is still quite weak. Therefore, we attempted to carry the analysis further in quest for more significant results. Offshore influence can manifest itself in the Ucits space from another angle. Within the sample of Ucits with offshore experience, we can split the funds along yet another line – those with offshore equivalents and those without. The offshore equivalents themselves are offshore hedge funds managed by the same management company whose performance the Ucits track. Our purpose here is to see if there is a discrepancy between those funds that are started directly in the Ucits framework (new funds) and those that are created as replicas of existing offshore products and whether this discrepancy introduces any substantial heterogeneity amongst the Ucits with offshore experienced managers.
13
To avoid confusion we should stress on the following points. First, the fact that a Ucits fund is managed by an offshore experienced manager does not automatically mean that it tracks an offshore equivalent. Offshore experienced managers can manage both Ucits with and without equivalents. Secondly, the fact that a particular Ucits does not have an offshore equivalent does not mean that the management company has no offshore experience. It may manage other Ucits that track offshore equivalents, or indeed offshore funds that are not tracked by any Ucits. On the other hand, non-offshore experienced managers can, by definition, manage only Ucits without offshore equivalents.
Figure 4.2 illustrates the decomposition of the Ucits sample into Ucits with offshore experienced and non-offshore experienced managers. The former are subsampled into Ucits with and without offshore equivalents. The separate box represents the offshore funds that are tracked by the Ucits with equivalents.
Figure 4.2 Alternative Ucits sample decomposition
This figure presents the decomposition of our sample of Ucits by manager offshore experience. Ucits with offshore experienced managers are additionally split into subsamples of Ucits with offshore equivalents and Ucits without offshore equivalents.
Performance data on the offshore hedge funds is collected and merged from BarclayHedge, Lipper TASS, Bloomberg and fund prospectuses and is net of fees. The majority of them are domiciled in the Cayman Islands. Other domicile locations such as Bermuda, the British Virgin Islands, the Bahamas, Guernsey and even Ireland and Luxembourg are also present in our database. Deutsche Bank’s Ucits platform, for example, allows managers such as Winton Capital and Man Group to replicate their offshore hedge funds onshore through a swap agreement. Alternative Ucits of this kind are established after the offshore hedge fund has already been in existence for a period of time. One possibility is that the Ucits are established by successful managers to offer competitive performance in an onshore vehicle in order to diversify their client base. Alternatively, they might be established by managers who are not so successful with the purpose of gaining access to the European investment market and raising assets under a presumably safe structure. Indeed, the Ucits framework facilitates fund marketing and distribution. This is especially useful in the post-crisis period when investors seek regulated collective investments.
What we find, using equally-weighted monthly-rebalanced indices for the period June 2008 to December 2010, is that the alpha of Ucits without equivalents is positive and statistically significant at the 95% confidence level. The Ucits with offshore equivalents, on the other hand, exhibit no significant alpha. We can conclude therefore that the performance of Ucits
Ucits with non-offshore experienced managers
Ucits with offshore experienced
managers Ucits with offshore equivalents (replicas)
Ucits without offshore equivalents (new funds)
Offshore equivalents
14
with offshore experienced managers during our sample period is driven by the performance of Ucits without equivalents. After removing the polluting factor of replicating Ucits, offshore experienced Ucits managers do add significant value to the performance of their funds.
This would lead us to believe that Ucits with equivalents are set up as replicators of second- tier offshore products. Such a conclusion in no way undermines our result on the added value of offshore experience. However, it puts a big question mark on the long-term viability of this marketing and distribution strategy. In any case, our results should be viewed with regard to the impact of the financial crisis on the hedge fund industry’s performance during our sample period, as indicated by Figure 4.1. The need for further research within the hedge funds space is thus highlighted. Moreover, we should not forget about the additional cost layer of the Ucits structure as compared to the tracked offshore vehicle.
4.7 Risk Profiles
Having estimated the alphas, it is important to compare the alternative Ucits indices in terms of various risk metrics. The sample period for the comparisons below is from June 2008 to May 2011. Unlike in the alpha estimation above, here we are not restricted by the available data on lookback straddles and we are able to cover a period of 3 years. Table 4.6 shows that Ucits with offshore experienced managers are significantly more volatile. They are also riskier in terms of downside deviation and maximum drawdown. Downside deviation refers to the volatility of index returns that lie below the index average. Even though the performance in terms of net-of-fees returns is not significantly different, it leads to a much higher Sharpe ratio for the Ucits with offshore experienced managers.
Table 4.6
Manager offshore experience: monthly risk metrics of the indices
This table presents the risk metrics of the equally-weighted monthly-rebalanced indices of Ucits with non-offshore experienced (368 funds) and offshore experienced (182 funds) managers.
June 2008 - May 2011
Mean Volatility Downside
Deviation
Maximum
Drawdown Sharpe Ratio
Non- offshore exp. 0.01% 0.95% 0.77% 7.94% 0.01
Offshore exp. 0.10% 1.37% 1.44% 9.82% 0.08
t-, F-test p-values 0.73 0.03
To present the influence of offshore replication in a more explicit manner, in Table 4.7 we compare Ucits with and without offshore equivalents. Again, there is no meaningful difference in the returns but Ucits with offshore equivalents are significantly more volatile and perform worse also in terms of downside deviation. The difference in maximum drawdown is especially conspicuous and very much in line with the results from the previous section. We can again confirm that offshore experience adds value only in the cases where the Ucits have no offshore equivalents. It boosts their alpha and provides for a lower risk profile.
15 Table 4.7
Ucits with and without offshore equivalents: monthly risk metrics of the indices
This table presents the risk metrics of the equally-weighted monthly-rebalanced indices of Ucits without (119 funds) and with (63 funds) offshore equivalents.
June 2008 - May 2011
Mean Volatility Downside
Deviation
Maximum
Drawdown Sharpe Ratio
W/o offshore eq. 0.16% 1.20% 1.20% 7.51% 0.13
With offshore eq. -0.26% 2.57% 2.91% 23.64% -0.10
t-, F-test p-values 0.38 0.00
Comparing Ucits and their offshore equivalents in Table 4.8, we can see no differences in the mean returns or their volatility. Nevertheless, the Ucits exhibit sizably larger maximum drawdown and their negative return translates into a negative Sharpe ratio. This could be attributed to the additional costs that the Ucits charge and the imperfect offshore tracking.
Table 4.8
Ucits and their offshore equivalents: monthly risk metrics of the indices
This table presents the risk metrics of the equally-weighted monthly-rebalanced indices of Ucits (58 funds) and equivalent offshore hedge funds (58 funds).
June 2008 - May 2011
Mean Volatility Downside
Deviation
Maximum
Drawdown Sharpe Ratio
Ucits (replicas) -0.32% 1.95% 2.17% 20.17% -0.16
Offshore funds 0.10% 1.94% 2.25% 13.49% 0.05
t-, F-test p-values 0.37 0.96
In an attempt to investigate further differences in the last case, we calculated the tracking errors between the Ucits and the offshore hedge funds. The tracking error for each pair of funds is defined as the standard deviation of the return differentials. The average tracking error is 1.8% in monthly terms, or 6.1% annualized. In order to observe whether the tracking errors are related to strategy, we grouped the pairs into three generalized groups by strategy. Equity market neutral, fixed income and event driven were grouped as relative- value strategies. CTA, macro, long/short equity, emerging markets and FX were grouped as directional strategies. Multi-strategy was left on its own.
We find that the average tracking error of the relative-value strategies is the lowest, the average tracking error of the directional strategies being significantly higher and the highest. The average tracking error of the multi-strategy lies in between those two extremes but is not meaningfully different from each of them taken individually. The performance of Ucits with equivalents therefore also depends on the replicability of the hedge fund strategy under a Ucits framework.
16
4.8 Conclusion
The alternative Ucits industry is relatively new but it is developing fast and attracting many new managers. We attempted to assess the added value of these managers based on their offshore experience – the so-called skill effect of applying hedge fund strategies under a Ucits framework. We found that Ucits with offshore experienced managers exhibit significantly more risk but on the other hand have a positive alpha with a p-value of 0.12.
Even though not acceptable at traditionally established significance levels, it prompted us to carry out further analysis within the sample of Ucits with offshore experienced managers.
To better understand the performance of these managers, we split their Ucits into two groups – those that track offshore hedge funds and those that do not. We find that Ucits with offshore experienced managers that do not track offshore vehicles provide positive significant alpha. A consequence of these findings is that the alpha of Ucits with offshore experienced managers is due to the Ucits without equivalents. Moreover, not only do Ucits without offshore equivalents provide positive alpha, they are also significantly less volatile.
Offshore experience therefore plays a significant role in creating positive alpha, as long as it translates into real innovations. That could be explained by the fact that a Ucits wrapper brings in additional costs. If the fund is a pure replica, this cost represents a handicap that is hardly compensated. Therefore, even though offshore experience appears to be an important factor, the space of Ucits with offshore experienced managers is far from homogeneous.
Ucits with offshore equivalents bring no added value and bear no substantial differences in their risk profile with their paired funds offshore. Nevertheless, we see a degree of divergence as shown by the non-negligible tracking error due to liquidity, leverage and diversification issues and the additional costs that the Ucits charge. Some Ucits only replicate the liquid part of the offshore portfolio. Even the swap structures are not perfect replications due to issues such as performance drag and the additional costs of the Ucits wrapper.
Empirical analysis always depends on the length and the constituents of the sample. The entry and exit of funds, the changing economic conditions, and the evolution of regulations will thus provide opportunities for further research. Whether we will see the alternative Ucits space become more homogeneous or more segmented is a question of time. In the same vein, the future will tell us if alternative Ucits can supplant offshore funds. For investors, however, the quality of the due diligence process remains the key driver when it comes to performance and our results indicate that selecting offshore experienced managers is already a first step in the right direction.
17
4.9 References
Agarwal, Vikas, Nicole M. Boyson and Narayan Y. Naik, 2009, Hedge Funds for Retail Investors? An Examination of Hedged Mutual Funds, Journal of Financial and Quantitative Analysis 44, 273- 305.
Anderberg, Karen L. and Laurence E. Bolton, 2006, Ucits: A Developing Model for the Future of the European Investment Fund Marketplace, Euromoney International Investment & Securities Review, 13-18.
Anderberg, Karen L. and Jessica A. Brescia, 2009, Ucits IV: Reforms to the Ucits Directive Adopted by the European Parliament, Euromoney International Investment & Securities Review, 17-21.
Beaudoin, Denis and Christophe Olivier, 2010, Ucits III Funds: One Size Does Not Fit All, Swiss Hedge Magazine 2nd Half 2010, 20-24.
Committee of European Securities Regulators, 2008, CESR’s Guidelines Concerning Eligible Assets for Investment by Ucits (CESR/07-044b).
Danaher, Simon, 2010, Total Return Swaps Expose Ucits Investors to Unknown Risks, Portfolio Adviser.
Darolles, Serge, 2011, Quantifying Alternative Ucits, Working paper Université Paris IX Dauphine.
European Securities and Markets Authority, 2011, Discussion Paper: ESMA’s Policy Orientations on Guidelines for Ucits Exchange-Traded Funds and Structured Ucits (ESMA/2011/220).
Fung, William and David A. Hsieh, 2001, The Risk in Hedge Fund Strategies: Theory and Evidence from Trend Followers, Review of Financial Studies 14, 313-341.
Fung, William and David A. Hsieh, 2004, Hedge Fund Benchmarks: A Risk-Based Approach, Financial Analysts Journal 60, 65-80.
Gabbert, Dale, 2005, Europe’s Chance to Put Right Ucits Errors, International Financial Law Review, Harris, Joanne, 2010, On Again, Off Again: Onshore vs. Offshore, Hedge Funds Review
Hedge Fund Research. 2011, Hedge Fund Industry Surpasses $2 Trillion Milestone.
HedgeWeek, 2011, Co-Domiciled Hedge Funds on the Rise as Non-Ucits Onshore Structures Gain Traction.
HFMWeek, 2010, Ucits Funds, March 2010.
HFN Industry Research, 2010, Ucits Compliant Hedge Funds, HFN Industry Research, November 4, 2010.
Johnson, Steve, 2011, ‘Creativity’ May Tarnish Ucits Brand, Financial Times, May 15, 2011.
Markov, Iliya and Nils S. Tuchschmid, 2011, Can Hedge Fund-Like Returns Be Replicated in a Regulated Environment?, In G. Gregoriou, M. Kooli, eds., Hedge Fund Replication.
Basingstoke, UK: Palgrave-MacMillan, 2011.
Murphy, Kevin and David O’Shea, 2008, The Eligible Assets Directive under Ucits III: Something for Everyone, Butterworths Journal of International Banking and Financial Law, 137-138.
Tuchschmid, Nils S., Erik Wallerstein and Louis Zanolin. 2011, Will Alternative Ucits Ever Be Loved Enough to Replace Hedge Funds?, Working paper Haute Ecole de Gestion de Genève.
Tuchschmid, Nils S. and Erik Wallerstein, 2013, Ucits: Can It Bring Funds of Hedge Funds On- Shore?, Journal of Wealth Management 14, 94-109.
Varriale, Gemma, 2011, Ucits Gaining from AIFMD Uncertainty, International Financial Law Review.
18
4.10 Appendix to Section 3.5
Given the particularity of alternative Ucits returns, many of the funds in our sample have short histories not covering the entire period from June 2008 to December 2010 or from June 2008 to May 2011. Our decision was, therefore, to analyze equally-weighted monthly- rebalanced indices for each subsample by adding funds to the indices as soon as they start reporting returns.
Nevertheless, for the sake of completeness and robustness check, we provide below some regression results on a per-fund basis. These results are only computed for those funds that cover the entire period from June 2008 to December 2010, i.e. 232 funds in total. This limits the selection bias, which is clearly diluted when computing indices, and other types of potential biases when dealing with incomplete panels. As a consequence, the total number of funds in each subsample will differ from the number of funds included in the indices analyzed in the text.
Table 4.9 reports the number of funds with significant positive or negative alpha in the samples of Ucits managed by offshore experienced and non-offshore experienced managers.
Two important ratios stand out from the table. First, the proportion of funds in the total sample with significant positive alpha is much higher for Ucits with offshore experienced managers (0.33 vs. 0.13). Secondly, the ratio of the number of funds with significant positive alpha over the number of funds with significant negative alpha is much higher for Ucits with offshore experienced managers (6.33 vs. 1.69). Even though one might argue that these results are not conclusive for the complete samples, they still provide good support for our analysis.
Table 4.9
Manager offshore experience (2008-2010): significant alpha at the fund level
This table presents the Ucits with non-offshore experienced (174) and offshore experienced (58) managers, that have a significant positive or a significant negative alpha. The analysis was performed on the 232 funds that cover the whole sample period from June 2008 to December 2010.
June 2008 - December 2010
Sign. positive
alpha
Sign.
negative alpha
Current
sample Full Sample
Non-offshore exp. 22 13 174 368
Offshore exp. 19 3 58 182
Table 4.10 presents the results from a panel regression of all funds where offshore experience is treated as a dummy variable. The goal here is to see whether offshore experience has any significant role when considered together with the 8 Fung and Hsieh asset-based style factors. We can observe a positive significant alpha for the entire sample of funds. What is more important, the coefficient of offshore experience is positive with a p- value only slightly above 0.10. Therefore, we can consider offshore experience as a significant factor boosting the performance of alternative Ucits.
19 Table 4.10
Panel data for offshore experience
This table presents the results from a panel regression of the 232 funds that cover the whole sample period from June 2008 to December 2010. There is a dummy variable for offshore experience in addition to the 8 Fung and Hsieh factors. The regression’s adjusted R-squared is 0.11.
June 2008 - December 2010
Estimate p-value
Alpha 0.00071 0.08
Offshore exp. 0.00127 0.10
Table 4.11 treats offshore experience from a somewhat different point of view. It presents the results of the second step of a two-step regression “à la Fama-MacBeth” where the individual fund alphas are first computed by fund time series, and then regressed cross- sectionally on a dummy variable representing the offshore experience. It shows that the effect of offshore experience on the alphas is positive and significant, although it explains only a small part of the alpha variability itself.
Table 4.11
Alpha vs. offshore experience
This table presents a regression of the individual alphas of the 232 Ucits that cover the whole sample period from June 2008 to December 2010, against a dummy variable for offshore experience.
The regression’s adjusted R-squared is 0.01.
June 2008 - December 2010
Estimate p-value
Intercept 0.00068 0.05
Offshore exp. 0.00139 0.04
20
21
