Hedge Fund Due Diligence And Its Limitations

Updated on

In a world where machines learning to create investment strategies is colliding with fundamental portfolio management, today’s algorithmically-minded investment managers go about their jobs in very different ways. One of the methods to mine alpha is to dig deep into niche markets. For a hedge fund allocator, such adventure can involve enhanced risk – particularly with smaller managers with exposure to exotic and sometimes illiquid markets. For many such investors, the reward can be meaningful when the focus is on risk management and ongoing monitoring in addition to manager selection.

Get The Full Seth Klarman Series in PDF

Get the entire 10-part series on Seth Klarman in PDF. Save it to your desktop, read it on your tablet, or email to your colleagues.

Position level transparency important, mainly when investing in smaller niche hedge funds

50 South Capital’s John  Schabilion, Analyst, Alternatives Investment Research, operates in a world where finding alpha involves finding hedge funds that run on the market’s edge. While the bulk of hedge fund industry allocations go to the most massive managers, finding opportunity is increasingly taking place beyond the well-worn path, such as is the case with 50 South, which focuses on strategies with $500 million or less in assets. The firm's approach is to identify “smaller, newer managers and investment strategies which tend to be more nimble and unique when compared to their larger peers” he said at a recent Opalesque Round Table event in Chicago.

50 South Capital is a wholly-owned subsidiary of Northern Trust Corporation, which has $1.2 trillion in assets as of the close of 2017.  50 South oversees $3 billion in hedge fund assets that Schabilion helps manage, and $6.9 billion in total assets, including private equity.

To analyze and monitor smaller funds, 50 South considers strategies based on several macro and micro factors and peels back the onion to a deep level to receive position-level transparency, something more significant funds are often shy about revealing.

“We receive position-level transparency for our managers through the RiskMetrics system, which allows us to model how a fund will fit within a particular portfolio and helps us make allocation adjustments based on changing market conditions or risk and return objectives,” Schabilion told ValueWalk.

Nathan Anderson, founder, and CEO of Clarity Spring Securities, a hedge fund due diligence and fraud detection firm, advocates that hedge fund investors utilize separately managed accounts (SMAs) whenever possible because they provide position level transparency.  "Institutional investors should demand it more," he told ValueWalk, noting that sometimes larger hedge funds are unwilling to provide such transparency. "Without it, investors run into pitfalls of such as exposure overlap and position duplicity."

Position level transparency is instructive from several levels, allowing financial professionals the ability to analyze, understand and model a strategy through different market environments. To deliver this type of risk/reward, Schabilion doesn’t rely on just a quantitative look, however. This means, in part, reviewing strategies and smaller managers often overlooked by large institutional peers.

“Our strategy evaluation process relies on both qualitative and quantitative diligence in order to determine if a strategy will be uncorrelated to both markets and hedge fund peers,” Schabilion said.  “Through this process, we are able to build a range of portfolios designed to meet different risk and return objectives, while avoiding unintended strategy, sector, factor, or asset class risks.”

NathanAnderson
Nathan Anderson, founder, and CEO of Clarity Spring.

Are those running AI strategies being given a pass regarding explaining why an approach works that systematic managers didn't get?

In allocating toward smaller hedge fund managers operating in niche markets, there are benefits and risks. The opportunity to find a unique source of alpha comes with risk not as predominate when investing with the world’s largest managers: individual business risk.

Depending on the due diligence process, specific allocators will go as far as obtaining a general sense of how much revenue a firm might be receiving and balance that with headcount to develop an overhead cost model. Further, some allocators monitor the trend in a hedge fund’s assets under management, as some statistical analysis of hedge fund failure points to a correlation between the two.[1] Other algorithmic processes monitor leverage to equity ratios and benchmark it relative to a strategy while still others consider exposure risk based on beta market environment.

Anderson evaluates fund managers by conducting a detailed background check as well as assessing them on several factors, including risk management, where he differentiates between harden-fast risk management rules and those hedge funds with a less concrete definition of their processes.

50 South Capital's process considers both mathematical and fundamental factors:

Our strategy evaluation process relies on both qualitative and quantitative diligence in order to determine if a strategy will be uncorrelated to both markets and hedge fund peers.  This includes an extensive review of the manager’s investment philosophy, research process, and risk management approach, coupled with a position-level analysis of the portfolio and a statistical review of the strategy’s historical track record.  We receive position-level transparency for our managers through the RiskMetrics system, which allows us to model how a fund will fit within a particular portfolio and helps us make allocation adjustments based on changing market conditions or risk and return objectives.  Through this process we are able to build a range of portfolios designed to meet different risk and return objectives, while avoiding unintended strategy, sector, factor, or asset class risks.

One of the hot topics is artificial intelligence (AI) and machine learning. While some systematic strategies, from large funds Man AHL and smaller Australian-based Goldsky Asset Management, have reported solid returns, questions remain on how to evaluate and recommend such strategies.

Certain allocators towards systematic strategies have required a fund manager to explain the strategy to the point they can model it relative to the larger market environment or have a fundamental logic as to why the approach is repeatable. They question AI hedge funds who can't or won't explain why a strategy is repeatable.

"Albert Einstein's theory of relativity might not make intuitive sense at first glance," Anderson noted. "But it can be explained on a fundamental, logical level." Correlation is not always causation, but if an investment manager cannot explain the difference what they are doing "just looks more like advanced data mining."

Some systematic analysts have maintained that strategy analysis should include the ability of the algorithmic fund manager to explain to a fundamental portfolio manager how the strategy operates and recognize why it generates repeatable performance. With the popularity of algorithmic investing just starting to expand, other analysts assume that correlation is in fact correlation and don’t require deep fundamental or economic justification for a strategy to work other than technical considerations.

Schabilion, for his part, takes a broad approach and then narrows down options:

The evaluation of an AI or quantitative strategy should start with a deep understanding of the investment team’s capabilities/expertise and current research process and then seek to review the current model within these parameters.  It is important that the team adhere to a strict development protocol in order to avoid adding misunderstood models or unintended risks to the portfolio.  With the case of AI and machine learning specifically, an investment team should have deep expertise in the space if they are employing adaptive techniques and have protocols in place to control/manage overall portfolio risk.  We have seen the greatest successes coming from managers who use AI techniques as part of a broader systematic tool kit and apply these methods to trading specific time periods or risk factors where they can add the greatest incremental value.

While algorithmic evaluation methods vary, Schabilion strives to understand each strategy, and when it deviates from expectations, such strategy drift is considered cause for re-evaluation of the investment:

Through the RiskMetrics system we are able to run a range of stress tests on a fund in order to determine its overall riskiness.  We combine this assessment with an understanding of a manager’s historical track record and risk management approach in order to frame the best and worst environment for a strategy along with an expected maximum drawdown level.  As part of our ongoing due diligence for current manager allocations, we closely monitor strategy performance and the overall risk levels in order to make sure that the strategy remains in-line with our expectations.  Any large-scale changes or departures in a strategy’s risk profile would prompt a deeper conversation with the manager and could result in an allocation reduction or termination.

 

[1] Study of the BarclayHedge “Graveyard” Database circa 2010.

Leave a Comment