Hedge fund overlap is an overlooked measure of an investor's diversification. In this article, we'll show you how to measure it.
The Novus Research Team dedicates many hours to examining how to isolate and identify manager skill sets. Most reports in our Research Library attempt to quantify these skills to help facilitate manager evaluation. An important step in solving this puzzle—putting aside our usual attribution obsession—is to inspect the underlying composition of a manager’s portfolio by analyzing the manager’s diversification and uniqueness against their peers. We call this hedge fund overlap. Ignoring such analysis and only focusing on returns creates unexpected risk from overlap between invested managers. In an environment where individual security crowdedness has reached an all-time high, it’s essential to understand how diversified your managers are at a portfolio level.
One of the longest-tenured and most-cited analysis tools at Novus is the hedge fund Overlap Model. This model sources positions from public filings to compare the portfolio overlap of peer groups. Such analysis has long been featured on the Novus research website for the Tiger family of managers. If you don’t have access to the Novus platform, you can follow along or investigate further using the Tiger Overlap Matrix. This publicly available tool, which is constructed similarly to the model used in this article, is updated post filings every quarter. For Novus Alpha users, log in to view the portfolio sampled in this report. Access to the peer group list is available upon request.
Investors and managers alike use the hedge fund overlap model for many different analyses, but its main function is to compare the sum of position sizes for securities owned by two managers. This calculation identifies what percentage of each manager’s portfolio is invested in the same names. For example, the graphic below shows that Pentwater has a 33% overlap with Point State, whereas Point State has a 40% overlap with Pentwater driven largely by their mutual put option on a S&P 500 ETF.
The below overlap matrix includes thirty of the largest institutions in the Novus Hedge Fund Universe to help us understand current overlap in the industry. Funds on the Y-axis are arranged in descending order of average overlap.
Point State, at 29%, has the highest average overlap compared to each of its peers. Point State and Pentwater are two of the three smallest funds by AUM to make the list, and have the highest average overlap. Both have a much larger number of invested positions than to the remaining top ten. To further dissect the overlap number, let’s examine the uniqueness metric. This measure indicates the percentage of each fund’s portfolio that’s comprised of totally unique positions—any not owned by another peer within this group. This can help detect the underlying cause of the overlap.
If a fund is in a few positions that are owned by many managers, they could end up with a relatively large average overlap. This doesn’t mean, however, that their portfolio isn’t differentiated. The manager’s diversification value begins to diminish when uniqueness becomes low. For example, if Point State’s overlap was exclusively driven by owning the SPY, they would still have 73% uniqueness. After scrutinizing Point State’s uniqueness, we find that while they own a large position in a popular name, their high average overlap isn’t fully explained.
The Novus hedge fund Overlap Matrix can also be used to check exposure to various indices. Increasingly popular is the option to compare a manager to the proprietary Novus 4 Cs—a constantly evolving collection of securities tied to respective movements within our Hedge Fund Universe.
Using hedge fund overlap as a proxy, you can see that Lone Pine has a 30% exposure to the Novus Conviction Index, our index of the biggest conviction bets in the HFU. This method can also isolate exposure to the most crowded names. Any fund with significant overlap to the Novus Crowdedness Index automatically assumes a substantial risk factor that’s otherwise hidden when focusing exclusively on attribution analysis.
Using hedge fund overlap as a holdings-based analysis alternative to attribution metrics offers a refreshing viewpoint on a manager’s offering, allowing a more precise measure of uniqueness. Holding or adding a manager with high levels of overlap contributes to risk factors and correlated return streams. It’s important to ensure that the diversification a manager markets actually exists. Measuring hedge fund overlap against your current portfolio or a hedge fund factor index is a surefire way to start.