The "ensemble" approach to finding highest-conviction stock picks from top-class managers, and then finding those stocks with the strongest consensus agreement, is an approach that has been known for years. What's new, however, is using AI tech to make this work.

(An earlier version of this article appeared in Family Wealth Report, sister news service to this one. While the firm in question is from the US, the lessons of its ideas are global.)

The approach known as ensemble active management – aka EAM – is proven to deliver higher results for actively managed US stock portfolios than any mutual fund firm in the US, its creators say in a new paper.

The EAM model, developed by Turing Technology (a firm based in Wilmington, Delaware), and launched into the market in 2018, has been detailed in a new paper for the CFA Institute, written by Rob Nestor, president of Turing and Vadim Fishman, the company’s co-founder.

The paper is entitled EAM: How and Why AI-Powered Active Management Will Dominate Passive. 

The EAM approach captures the highest conviction stock picks from a range of top-performing fund managers, and then identifies the stocks with the strongest consensus agreement. Modern technology such as AI can extract data from tens of thousands of managers’ stock picks to build what a particular consensus view is at any one point. (This news service interviewed Turing Technology here.)

According to Turing Technology, all nine lead EAM strategies have outperformed their benchmarks since inception, and the annual excess return vs its benchmark, for all nine style boxes, with averages of 5.2 per cent (516 basis points), the firm said in a statement.

(The investment industry labels equity portfolios based on size (large, mid, small) and style (value, blend, growth), with the 3x3 matrix creating nine “style boxes.”) 

Not a single mutual fund firm in the country, when their best nine funds are used and assessed over three-year or five-year periods ending year-end 2023, can make those two claims, Turing Technology said.   

“We know it defies conventional wisdom for EAM to elevate to these performance heights in just five years. There are over 400 fund companies with active US equity funds, and the largest of them (for example, BlackRock, Fidelity, Vanguard) have virtually unlimited resources, unlimited access to investment professionals, and decades to hone their craft. And not a single one of them can deliver more performance value than EAM,” Nestor said.

“There is a fascinating parallel to the launch of data science in Major League Baseball, highlighted in the iconic book Moneyball. In 2002, Billy Beane rolled out his first team built on ‘sabermetrics.’ With the league’s smallest payroll, the Oakland A's still broke a 96-year record for consecutive wins,” Nestor continued. “That…record-breaking run stunned the industry. But the Oakland A's data analytics created competitive advantages that tilted the competitive landscape. Today, there is not a professional sports league not impacted by Billy Beane’s insights. In the same vein, we expect EAM will become the guiding light for active management.” 

This news service asked Nestor how its EAM philosophy and approach was influencing wealth managers. 

“Firms are taking notice and momentum is building because the hard, verifiable validation of EAM keeps rolling in. The clarity in which the CFA paper explains how and why EAM works has been an accelerator for us, but real EAM portfolio returns, independent validation of results, and understanding how the machine learning’s ‘go to’ tool of ensemble methods creates alpha are what’s driving interest and buy-in," Nestor replied.

The CFA paper said EAM can generate excess returns over benchmarks of 4 to 5 per cent a year.