Institutional investors, it turns out, are not unlike a herd of sheep grazing on the countryside.
A new analysis from Curalytics, a New York-based big data analytics company, shows which institutional investor portfolios are similarly constructed, aka “herd mentality,” and which are largely unique, aka “contrarians.”
“People know herding is out there,” Steffon Davis, head of product and founder at Curalytics, said in an interview with ThinkAdvisor. “We always kind of wonder who are the contrarians and who are the herds, but nobody wants to say they’re part of the herd. Nobody would ever claim that, it’s embarrassing. But what we can see is that they are out there — and the contrarians are out there, too.
Using Curalytics technology and Form 13F, which large investors must file with the Securities and Exchange Commission, the Curalytics team was able to examine more than 3,100 portfolios worth more than $100 million each.
Of those 3,100 portfolios, Curalytics found that 593 portfolios have more than 50% in common with each other.
From this data, Curalytics mapped out a graph that visualizes the similarities between portfolios. A similarity, in this case, is defined as two or more portfolios containing 50% or more of the same securities. Davis likened the graph to a Facebook social graph showing connections of friends, where every dot represents a person and the lines connecting the dots are their friendships.
In this case, the dots represent funds or portfolios and the lines identify what Davis calls the “thesis groups,” or a group of similarly constructed portfolios. Lines connect similar funds to “thesis groups” and clusters reveal concentrations of funds with similar theses.
“What it’s saying from a data point of view is that all of these funds have 50% or more in common with each other,” said Davis, during a phone call with ThinkAdvisor. “I mean it’s just a massive amount of overlap. It visually shows that there is herding behavior happening. It also shows that there are funds doing very different things.”
Davis has a background in curation technology and high-frequency trading. Before founding Curalytics, he worked in high-frequency foreign-exchange trading at MarketFactory and founded the content curation site Topiat.
Beyond confirming that there is a herding mentality among investors, Curalytics research also has some clear benefits for portfolio diversification analysis and management.
Davis gave an example.
“I want to diversify my portfolio, and I’m evaluating two different funds,” Davis said. “These funds may have marketing that says they have different strategies.”
By looking at Curalytics’ research and herding visualization, Davis can see that the two funds are actually within the same thesis group, so in reality they may actually be very similar, regardless of how the fund describes their strategy. “If my goal is to diversify my portfolio, I’m not actually going to achieve diversification by buying these two funds because they’re part of the same group,” Davis said. “You can actually visually achieve diversification by buying funds from different parts of the diagram or visualization because then you’re guaranteed that they’re different from each other.”
Curalytics’ research is also useful to evaluate portfolio management.
“The ability to take the portfolio that you’ve constructed, and the ability to run your own portfolio against the publicly available portfolio and see if what you’ve made [is different],” Davis said. Adding, “[You can] see where your thesis group is and depending on your investing goal you may want to rebalance your portfolio.”
Curalytics has started exploring applications of the technology to portfolio management, fund marketing and portfolio diversification analysis.
“What [Curalytics] offers as a company is access to this data programmatically,” Davis said. “We update it every day; you can connect to this data and run it in your system.”
— Check out Is Portfolio Diversification Overrated? on ThinkAdvisor.