The Secret Sauce to Source Multi-Manager Funds

Multi-manager funds have recently experienced remarkable growth in terms of asset acquisition. The rationale behind investors increasingly allocating their hedge fund investments towards these funds is understandable; they consistently delivered impressive returns with minimal risk in the past. However, the factors contributing to these desirable return profiles are somewhat nebulous. Hence, illuminating the key drivers becomes crucial for investors to refine their investment process and concentrate on elements that are pivotal while seeking new multi-manager funds. This consideration has become even more significant lately with the proliferation of new multi-manager funds, each varying in key aspects of their investment approaches. In this blog, we aim to demystify the risk-return profile of multi-manager funds from a quantitative perspective and show that contrarian to common believe, portfolio construction plays the key role.

Identifying Potential Return Drivers 

When asked about the essential aspects of their investment approach, multi-manager funds and their investors are likely to highlight the ability to identify and source talent. Given that multi-manager hedge funds consistently strive to recruit proficient portfolio managers (PM) and are willing to offer generous payouts to secure top talent, this response is unsurprising (for more, refer to this Bloomberg article). However, the talent within a multi-manager fund is challenging for investors to quantify when exploring new funds - unless they have insights into the performance of the multiple (sometimes over a hundred) PMs within such funds. Interestingly, contrary to this common belief, our findings from a previous blog post suggest that talent plays a less significant role than often assumed. Remarkably, even when we restricted our simulated multi-manager fund from investing in PMs ranking in the upper half, the fund still demonstrated impressive Sharpe Ratios. While talent undoubtedly influences a fund's performance, our results imply that the success of a multi-manager fund extends beyond merely sourcing the industry's best talent. Along with talent, leading multi-manager funds also demonstrate stringent risk management, swiftly cutting any PM's book as soon as certain risk limits are breached. Additionally, these funds are usually fast in allocating capital, quickly scaling individual PMs' books based on investment opportunities. Lastly, portfolio construction, particularly diversification across various strategies, could be another potential driver.

We will dissect each of these aspects individually, grounding our analysis in the simulation engine outlined in our previous blog post. Essentially, since January 2012, we select the top-performing PMs from our database of more than 10,000 hedge funds every month. We then conduct a portfolio optimization and rebalance our multi-manager portfolio monthly, subject to a rebalancing penalty to avoid excessive PM turnover. We terminate and blacklist any PM exceeding a 1.5 annualized sigma risk limit at the end of each month. This process results in a performance track record for our simulated multi-manager fund.

Risk Management

First, we examine the effect of risk management on the fund's risk-return profile. We do this by comparing the Sharpe Ratio of our core model (which includes individual PM level stop loss) with a model devoid of any PM-level stop losses.

The chart above reveals that risk management affects the Sharpe Ratio. While our model delivers an impressive Sharpe Ratio of 3.0, including a PM-level stop loss, the ratio modestly drops to 2.7 without any PM-level stop losses. Consequently, the annual PM turnover also reduces from 25% to 22%, understandably because the model will be less inclined to divest from entire PMs' books. While both Sharpe Ratios significantly still exceed the 0.6 average of traditional hedge funds, at first glance, risk management appears to have a rather low impact. However, it's crucial to consider that our simulation is confined to monthly data frequency. A real-world multi-manager fund with access to intraday data could react much faster than we do in our simulation. Therefore, we assume that the 0.3 deterioration in Sharpe Ratio is an underestimate, and the real-world impact of risk management would be considerably larger. Hence, investors should prioritize multi-manager funds with tighter risk management and risk limits over funds with lax risk management.

Speed of Capital Allocation

Next, we turn our attention to nimble capital allocation. To assess this, we limit how frequently our simulated multi-manager fund can rebalance its portfolio, gradually extending the interval from monthly to quarterly, semi-annually, and then annually.

The results of this approach, displayed in the chart above, reveal a decline in the original Sharpe Ratio of 3.0 to 2.4, and then oscillating between 2.2 and 2.3. This shows the significance of a multi-manager fund's responsiveness to new investment opportunities. Investors should favor multi-manager funds capable of quickly adjusting individual PMs' books, rather than those unwilling or unable to do so. Such funds typically employ their PMs directly, rather than investing in external managers. This approach avoids the risk of external investors over-utilizing a PM's capacity. Moreover, we expect that funds employing many rather than a few PMs also perform better in fast capital allocation. While this distinction wouldn't affect our simulation, in reality, distributing AuM among numerous PMs results in smaller dollar investments per PM, making adjustments to the scale of each book less impactful.

Portfolio Construction

Lastly, we examine portfolio construction. We start by exploring the influence of allocation to individual books.

The chart above contrasts a simulated multi-manager fund selecting PMs and optimizing its portfolio monthly with a fund selecting the same PMs but assigning them equal weights each month. A significant decrease of 0.5 in the Sharpe Ratio is observed when the fund does not optimize its portfolio. Generally, portfolio optimization can reduce risk by allocating to loosely correlated PMs, typically those employing diverse trading strategies. As such, the advantages of investing in a multi-strategy fund warrant consideration next.

The above chart showcases the performance of simulated multi-manager funds investing solely in Convergence, Divergence, or Long Biased strategies. Compared to our original multi-strategy fund, the specialized multi-manager funds exhibit lower Sharpe Ratios, ranging from 2.8 - 1.4. This clearly indicates that maintaining a diverse portfolio of strategies is probably the most critical factor contributing to the attractive risk-return profile of these platforms. With fast capital allocation and strict risk limits, multi-manager, multi-strategy funds can effortlessly adapt to an ever-evolving investment landscape.

Conclusion

The dynamic field of multi-manager funds, provides a fertile ground for investors. Our analysis underscores that the sterling performance of these funds is not solely predicated on talent acquisition. Rather, it is contingent upon three additional pivotal factors: risk management, speed in capital (re-)allocation, and portfolio construction.

The chart above highlights the potential impact of these factors on the multi-manager funds' performance. All factors have an impact and warrant consideration when sourcing new multi-manager funds. Yet, when an investor wants to focus on only one aspect, portfolio construction has clearly the largest impact. An investor should thus look for a multi-strategy approach among the multi-manager funds and if such funds are hard to source, multi-manager funds with Convergence focus are almost equally profitable.