There has been a phase of the multi-manager hedge fund industry when firms in it competed through privileged access to superior managers. Next, selectors of hedge funds would brag of the size of their research teams and deep understanding of the investment strategies they invested in. There followed a period when all the above were a given, and multi-manager hedge fund businesses marketed their risk measurement and risk management and portfolio construction techniques as if they were an edge. So the trends have been towards more resource, more process and more quantitative tools and inputs to decision-making.
As one would expect of a selector of hedge fund managers with an institutional client base, the processes of FERI Trust GmbH are well structured and thorough. The firm has a track record going back to 1999 in advising/managing hedge fund portfolios for institutional investors. In particular, the firm focuses on giving its pension fund and insurance company clients in Germany, Austria, North Italy and Holland bespoke solutions through its managed account platform (MAP).
Both the top-down and the bottom-up parts of the investment management process have unusual features. The two parts will be examined in turn.
The first key to the top-down part of the construction of portfolios of hedge funds at FERI is in Fig.1. The labelling of the components of the universe by those who select hedge funds is indicative, and it is usually reflected in this type of illustration somewhere in the presentation.
In the hedge fund industry the scope of the broad labels varies from investor to investor. For example, for some investors “equity hedge” managers are long-biased equity hedge managers, and market-neutral managers are in another box. For others, if a manager invests only in equity they are in the “equity hedge fund” category however they shape their balance sheet.
For FERI “event driven” is the classic combination of distressed investing, merger arbitrage, and special situations investing. “Equity hedge” for FERI is the first variation given above, and so for Storr and colleagues quantitative equity market-neutral comes under relative value. The bias that FERI has in equity hedge is to avoid set-ups with multiple risk takers, and FERI has bias towards managers running concentrated portfolios. Together these two strategy buckets (ED and EH) constitute the directional elements of the portfolio construction.
“Tactical trading” includes systematic CTAs and global macro managers. Non-systematic managers of three types are included in this strategy bucket: commodity managers (say managers who spread trade commodities), non-systematic (discretionary) macro managers, and managers with variable/opportunistic exposures to markets.
“Relative value” includes fixed income arbitrage, convertible bond arbitrage and volatility arbitrage. The tactical trading and relative value risk buckets diversify fund of funds-level risk from the return impacts of markets directions. Under this view of the hedge fund investment universe the top-level portfolio role of the managers is more important than their idiosyncratic methods of delivering returns (the specific investment process undertaken).
One of the consequences of this universe view/portfolio construction approach is that multi-strategy managers do not fit in easily. If a fund of funds manager or alternatives advisor like FERI wants to enact top-down views in a portfolio of hedge funds then managers that at their discretion vary exposures to strategies are an impediment. There is no point in having a strong positive opinion on CB arbitrage only to find a large relative value manager allocating away from the strategy towards volatility arbitrage or RMBS positions.
Marcus Storr, head of hedge funds at FERI Trust, says “There are two ways of approaching allocating to a fund: look at the fund’s merits and how it manages the portfolios, or look from a macro perspective.” In the second way the question asked is in which environment is this strategy suited to work?
Over a period of years FERI Trust has built models to answer this question. A large number of factors have been tested to see which ones correlate well with hedge fund returns at the strategy level. “For example,“ says Storr, “we did some research in volatility – and showed that it was positive for CTAs. For event-driven strategies high volatility in markets is not good, but ED funds do well in normal volatility conditions in markets.”
The work mapped economic and market conditions, and currently uses four factors. For this purpose the best lead indicator for global economic growth is the OECD leading indicator. On the market side factors include inputs like the Merrill Lynch Crowded Trade Index and the Credit Suisse Risk Appetite Index, and the VIX. “Through modelling we get to a point of being able to indicate the attractiveness of the hedge fund sub-strategies,” explains Storr. The summary output of this tactical strategy allocation (TSA) is shown in Table 1.
CLICK IMAGE TO ENLARGE
The attractiveness is scored regularly so that time series have been created, and the scores can be compared on a cross-sectional basis at any one point in time. That is, a score of 0.5 in one sub-strategy does mean it is less attractive at that point than a sub-strategy with a score of 0.75 at that point. However, the quality of the signal generated varies through time. Consequently a high score for attractiveness may be undermined by a low confidence in the score. Partly because of the TSA system indications, in recent times CB arbitrage has been the most attractive strategy. Marcus Storr gives other reasons that have made CB arbitrage very investable:
Storr and his team put some effort into understanding the state of the macroeconomic and market environments through reading research, including that from FERI EuroRating Services AG. Like all those who invest in hedge funds, FERI Trust benefits from the insights of the hedge fund managers who they track for views on economic activity and traded markets. Many managers provide good commentary and context for what they do, and a few, like Bridgewater Associates, provide great insight.
Through internal discussion and debate the hedge fund investment team at FERI constructs a macro view. Within this framework some strategies will be expected to do better than others – which gives a bias towards allocating capital to some strategies and away from others. But FERI Trust does not stop there. Rather, Storr and his senior colleagues, Ilya Kutsarov, Dr Thomas Maier, and Tao Wan, will use the TSA to test the rigour of their own organic views. Storr puts it thus: “We make up our own mind on the macro environment, and then use the TSA as a guide to decision-making, but it is not the determinant.” If you like, the TSA is a tool which can challenge the decisions made on other bases. In that way it can account for biases because it is an objective measure.
The tactical strategy allocation attractiveness scores can be a challenge to the strategy allocations that would be the natural consequence of the team’s macro view. This is an example of one of the strengths of FERI Trust. The staff are expected to be able to take challenges from others. There is some academic curiosity in the group, and an academic culture is prevalent, but it has some of the competitiveness that can also be a feature of academic life. This manifests in frequent friendly wagers between staff members.
The decisions made through this process in the last year or so have seen a reduction in allocations to CTAs and “mixed arbitrage”. The corollary has been an increase in allocations to distressed securities, special situations and to some equity strategies.
Having looked at the top-down part of the investment process of alternative investment advisory work at FERI Trust, we now turn to the bottom-up, manager selection.
Initially identifying hedge funds to follow is done the same everywhere – databases with some personal recommendations and conference/cap intro exposure to round out the universe. Those who invest in hedge funds as a business tend to use screening tools, and peer group analysis. In FERI’s case they use PerTrac, Riskdata FOFiX and EViews for these tasks and for performance analysis and risk aggregation.
In assessing a manager five sorts of risks are looked at: administrative risks, operational risks, management risks, portfolio risks and personal risks. To address the last of these, FERI Trust does thorough background checks of all fund managers including education, former employers and checks with former/existing clients. Where a manager has a PhD the checks include verifying the claimed theses and papers are attributable to the person claiming them, with the checker speaking directly to the supervising professor.
As part of the initial due diligence on a manager and fund FERI Trust has a structured analysis process and conducts systematic documentation and evaluation. The in-house database has 50 process fields and 400 information fields per manager. It is indicative that a customised questionnaire for missing information is generated and completed through further meetings and conference calls. The investment advisor conducts over 250 manager meetings and visits a year. To put this in context, FERI Trust has 96 approved managers and is invested in 62 at the moment. The firm has a bias to concentrated (fund of hedge funds) portfolios.
All the meeting notes, DDQs, quantitative measures and qualitative assessments are fused in the Rating Base System. There are five rating segments with different weightings (see Table 2). As discussed below, the in-house hedge fund ratings feed into the portfolio construction process at FERI Trust.
Expectations of loss
According to Storr it is possible to make good estimates of the downside risk of hedge funds. The losses that could appear as left-hand-side tail risk may be available from historical data or from estimates. The estimate of loss can be made from knowing what the manager trades, knowing what they do, and how they hedge. FERI records on its own system what the expected drawdown of the funds they analyse and invest in could be. Then, rather like a stop-loss on a single manager’s equity holding, when a fund is in drawdown there is a reference that can act as a review/action trigger.
When funds are in drawdown and approaching the expected limit FERI staff speak to the manager with increased frequency to find out whether the manager is following the portfolio-level risk controls. So a manager with a 25% drawdown forecast that is down 20% will get a call, and provided that the manager follows their own rules about cutting exposure and taking losses then it is possible for FERI to stick with them.
Marcus Storr says that not infrequently FERI allocates more capital to a manager in a drawdown, but with conditions. Those conditions are that the accumulated loss is within the range of expectations, the risk control procedures of the manager have been followed, and that the explanations for the loss are accepted. He says that FERI has just topped up a position with a manager in quantitative equity hedge that has had a 15% loss.
That FERI Trust can be responsive to market change is illustrated by the change in exposure to distressed company investments. Storr explains: “Certain sorts of distressed bonds in the US increased in attractiveness even as equity markets rose last year. This eventually led us to increase allocations to distressed securities. This was done through managers that work with companies that need funding. It is a sort of event-driven investing, and is nearly activism wherein small groups of hedge funds will approach a company with a financing package. It may be a convertible bond, it may be a corporate bond with an equity kicker, but in the negotiation it is a source of alpha.”
He expands: “This is a case where the hedge funds, and those that invest in the strategy through them, receive an illiquidity premium. We are fortunate that FERI has long-term investors (pension funds and insurance companies) and so is able to participate in illiquid strategies. Of course the strategy has to make sense from a risk premium perspective too.”
Educational and cultural aspects
It is much harder to differentiate between funds of funds than between single-manager hedge funds. In the latter processes differ and risk appetites are over a wide range. For PMs of multi-manager hedge fund portfolios there is much more common ground. In the case of FERI Trust there are distinguishing features in the processes and resulting from the client base.
Storr explains: “Transparency for us in Germany means education. We have to explain what these products are about. There is a gap to fill – the questions we ask managers and the quantitative measures we apply are in our presentation because we assume people in our market don’t know what we do at all. If you run around US pension plans you probably wouldn’t have a 70-page presentation like we have to.”
The footprint of FERI Trust is German-speaking Europe, Holland and Northern Italy. There is a natural (language) barrier to entry for non-Germans to address the market for hedge funds in Germany. However, that barrier to providers is not seen by FERI as the only one for their competitors. Certainly large providers, like funds of funds, sell into Germany. But Storr sees a flaw in the approach for the target client base, having worked for two investment banks himself. The indigenous potential clients often have a bias against how Anglo Saxons (for which read Brits and Americans) do business. “We can respond to a request for a face-to-face meeting tomorrow,” says Storr. This is in contrast to the investment banking model of parachuting in staff for a road trip to a territory. In the latter case “we will see you on our next quarterly visit, if the schedule permits” may be all a potential client is offered.
Storr expands: “The investment banking approach is to offer product A or B, and if they are not seen as appropriate, then off-the-shelf product C is quickly proffered.” The sales process of investment banks and those following that way of operating is short-term, according to Storr, whereas the German buying mentality is to consider buying something offered today next week or the month after.
So there at least three advantages for FERI Trust providing services to the markets they do. The firm is prepared to spend time on education of the potential clients, inculcating relationships. They are physically close and can respond quicker than a cross-border supplier, and third, in contrast to an investment bank, the cultural fit of a local supplier is good.
As a hedge fund advisor to investing institutions FERI has no bias to quantitative analysis. Storr says there is no competitive advantage in this element of fund selection. “Take the historical numbers and squeeze them through statistical methods – anybody can do that,” he asserts. “If we relied solely on those methods we would look the same as everybody else who selects hedge funds.” By inference FERI Trust make efforts to distinguish themselves on the non-quantitative parts of hedge fund analysis. However, it is worth stating that it is inevitable that a large part of the portfolio construction process has to be quantitative. So being distinctive in qualitative aspects does not diminish the largely quantitative aspects of portfolio construction. FERI Trust says that traditional portfolio optimisation is not suitable for hedge fund portfolios. What are suitable in portfolio construction at the Bad Homburg-based firm are four elements: the top-down allocation view, the quality of the fund, the value at risk, and systemic risk factors.
The top-down allocation view is a result of macro analysis, strategy view and opportunity set. The quality of the fund aspect results from the qualitative fund evaluation (conviction plus in-house rating). The value at risk is estimated for each fund, and the contribution to the portfolio VaR is calculated. The systemic risk factor element of portfolio construction looks at diversification based on factor exposures and specifically where there is clustering of risk as a result of including a fund. Also, in the portfolio construction phase FERI Trust tests the portfolio of hedge funds through scenario analysis and assesses tail risk.
FERI Trust GmbH has the largest hedge fund team in Germany, and that team is proportionate for the largest independent advisor on hedge funds there. FERI Trust advises on and manages over a $1 billion of hedge fund assets on behalf of pension funds and insurance companies. It is telling that $200 million of the AUM are in 20 UCITS hedge funds, including CTAs and credit long/short funds and UCITS hedge funds.
There are lots of tasks to perform to be a good selector of hedge funds, and to construct effective portfolios of hedge funds. It is easy to think that being thorough is most of the job. However, in the modern way of carrying out these tasks the funds selected have to be both representative of their strategy (to fulfil the portfolio-level role), but with superior outcomes (return, risk-adjusted returns or control of downside risk) compared to peer groups/indices. That is, fund selectors have to add value at the portfolio level and the manager level. The hedge fund advisory business of FERI Trust has a deep process of selection and monitoring, and the buy-in to the funds selected has to be significant because the portfolios constructed are concentrated.
But there is value added beyond manager selection. The FERI Trust uses tools to be systematically warned of opportunity, and this enables Marcus Storr and colleagues to allocate to strategies where potential profit is perceived to be abundant. So FERI Trust’s hedge fund advisory business is both structured and (perhaps surprisingly) responsive to opportunity, and in applying judgement is even-handed in its use of qualitative and quantitative inputs.