CSAM QIS manages about USD 2.1bn in assets across its LAB, managed futures, and multi-alternative strategies in a variety of vehicles on its investment platform. “The LAB UCITS fund has various share classes, which address investors’ preferences and needs in potential allocation sizes, currency exposure, and classification (institutional vs. non-institutional). And Credit Suisse’s Global Markets team has historically offered exchange-traded notes (ETNs) on LAB subsectors in the United States,” says Kung.

LAB is distinguished by a more nuanced philosophy and methodology than is seen in some “hedge fund replication” or “alternative risk premia” (ARP) products, though these descriptions are in any event not the most accurate labels. “Hedge fund tracking” or “hedge fund beta” is a better moniker than replication since the objective is not 100% replication and the strategies used have limited overlap with ARP. “LAB is both a superset and a subset of ARP, which can be both broader and narrower in different respects. LAB does not pursue all ARP strategies and it also does some things that are not usually classified as ARP,” says Kung. 

LAB’s fluid asset class, beta, style, and factor exposures are based on a mix of hedge fund industry asset weightings, factor exposures statistically inferred from asset-weighted indices, and importantly an a priori fundamental understanding as to what exposures hedge funds are seeking. Rather than attempting complete replication of hedge fund index returns, the approach delivers over 80% tracking accuracy, and its statistical modelling errs on the side of underfitting rather than overfitting. Through a full cycle, its net return profile has slightly outperformed broad hedge fund indices with slightly lower drawdowns. It has also avoided some systemic risks that have come from extreme concentration in certain corporate or other risks, such as a crowded stub trade in a German automaker. The implementation is flexible, using a variety of techniques and instruments to optimize costs.

Synthesising indices, fund selection, academic research, and derivatives banking

The LAB methodology builds on a heritage of more than 25 years of hedge fund industry performance monitoring and the approach has evolved over this period. “The Credit Suisse Hedge Fund Indices, published since the late 1990s, generated a lot of insight from gathering and cleaning the data and seeing commonalities and differences. In the early days, we ran an actual fund of funds attempting to track the broad Credit Suisse Hedge Fund Index (CS HFI). Though the product performed well versus funds of funds, pushback against the fund of funds model led us to explore academic and business school literature providing a conceptual basis for tracking hedge funds systematically using securities and derivatives rather than hedge funds,” explains Kung. Investable hedge fund indices have generally underperformed non-investable hedge fund indices. “We are seeking to track the broad CS HFI, which is an asset-weighted index that includes funds currently not accepting new subscriptions, to be aligned with best thinking from the smartest investors, rather than middle thinking,” says Kung.

A working group containing the LAB team and the fund of funds businesses inside Credit Suisse along with several academics outside the firm at London Business School spent 3 years of intense R&D conceptualizing and developing the strategy. 

A consistent methodology 

This led to the LAB strategy being launched in 2010. Since then, the core methodology and design has been only slightly modified. “The methodology has stood the test of time with a consistent 85% correlation with our own hedge fund indices. Investors have obtained exposure to hedge fund type returns without the same level of additional fees and without sacrificing liquidity,” says Kung. 

Top-down and bottom-up, statistical and fundamental analyses

There is a mix of top-down and bottom-up analyses. The weightings to the three broad LAB models, long/short equity, event driven, in tandem with global strategies, move relatively slowly with the asset composition of the hedge fund universe. These asset weightings also implicitly influence the inferred factor exposures since the indices tracked are asset weighted. “Equally weighting hedge fund indices has a different profile,” says Kung.

Yet LAB’s foundation is built around fundamental views on how hedge funds construct trades, and it is not only based on regressions or data mining. “Exposures are inferred partly from performance over various time horizons, but also from strong prior beliefs about what hedge funds do, which can be very different from the statistical inferences. These fundamental views are core to LAB and is why it is durable, whereas using machine learning algorithms could lead to quite erratic performance,” says Kung.

Not every possible factor is part of the mix, because they do not all move the needle of performance enough. Kung explains: “Linguistic differences do not always translate into statistical differences. For instance, the low beta factor might overlap with utilities or staples or even momentum. And our exposure to carry already has a liquidity provision dynamic in common with strategies such as statistical arbitrage, which provide liquidity and wager on reconvergence”.

Tactical and alternative beta

There is some exposure to tactical and alternative beta. There is tactical beta going long or short traditional asset classes and subdivisions of them, and alternative beta exploiting factors such as trend and carry across one or more asset classes. 

“Exposures to factors such as equity risk fluctuate, but net long exposure has typically ranged between 20% and 70% for the long/short equity model,” says Kung. “Credit exposure, which is primarily inferred from event driven managers, is frequently net long, but can range from relatively low levels of net exposure to much higher levels when event driven managers see opportunities in credit,” he adds. 

Beyond asset class betas, there are more granular and fluid exposures to geographies, sectors, and styles. In-house analytics are used to infer the exposures, with some different ways of fitting factors across models, to account for the differences in how risk is taken within the hedge funds. “For instance, in 2022 the long/short equity model identified a long developed markets/short emerging markets profile in terms of directionality and picked up biases to value, growth, momentum, and size in terms of style tilts. An additional filter picked up sector bets such as staples versus discretionary,” recalls Kung.

Capturing macro trading strategy exposures

The global strategies model aims to capture the return from trend following CTAs, discretionary macro funds, and additional strategies. To track these returns, CSAM QIS developed a diversified trend following strategy that follows market trends similarly to many managers in the category. 

The split between exposures such as trend, FX carry, and volatility carry varies over time. “The assumption is that a mix of trend and carry will capture some of the return profile from discretionary macro managers, who are usually more directional,” says Kung.

Some sources of hedge fund returns are not practical to capture, however, for both statistical and operational reasons: “If some discretionary macro managers are good at capturing turning points from convex trades, this is at the individual manager level, not the industry level. It would be hard to infer the positioning ex ante as it might only be visible after a large move. And it could require less liquid and possibly less transparent instruments to capture the convexity”. The fact that global macro managers were using convex instruments to trade the fixed income selloff in 2022 partly explains why LAB, unusually, underperformed the CS Hedge Fund Index in 2022. In 2023, LAB is up about 2.2% through the end of May, while the CS Hedge Fund Index gained about 0.3%.

Implementation and client adoption

From an ongoing implementation perspective, various instruments including stocks, futures, ETFs, and standardized or customized baskets or swaps are used, and the best execution committee regularly reviews transaction costs in line with fiduciary duty. “The CSAM QIS team has a dedicated trading function that considers how best to gain desired strategy exposures by considering factors such as best execution practices. Historically, the team has made frequent use of OTC contracts, thereby leaving a larger portion of the portfolio in government securities and cash,” says Kung.

Initially after launching the strategy, interest came mostly from investors with prior alternatives experience. Today, the strategy’s long track record and high correlation to the hedge fund index have attracted a number of additional allocator groups, including traditional index investors, that see the potential to gain index-like exposure in the alternatives space through LAB.