In 1996, The Onion published a satirical column entitled “Amish Give Up” that read:
“After centuries of enduring harsh, spare living conditions and voluntarily shunning modern amenities such as microwave ovens and red clothing, Amish leaders announced Monday that Amish across the U.S. will abandon their traditional ways and adapt to modern American life.”
“Said Amish Father Ezekiel Schmid at a Lancaster press conference, ‘This is pure bulls**t.’”
I reference this humorous column because financial advisors and their clients have endured decades of underperformance from their active managers, and it is time for them to follow Father Schmid’s lead.
Why now? Two reasons.
First, 2022 provided active equity managers with what S&P Dow Jones Indices characterized as a “favorable milieu” in which to thrive. In its year-end report assessing active managers’ performance, S&P Dow Jones added “a declining market, the underperformance of mega-cap stocks, record sectoral spreads, and above-average dispersion all militated in favor of active management.”
Yet 2022 was an annus horribilis. With few exceptions, across categories and geography, active equity managers failed to beat their benchmarks. Active managers even bombed in less-efficient categories, such as real estate and mid-cap growth, where they suffered from record annual underperformance rates of 88 percent and 91 percent, respectively, according to the S&P Dow Jones 2022 Year-End SPIVA Scorecard. The SPIVA research measures actively managed funds against their benchmarks.
While actively managed fixed-income funds fared a bit better, the SPIVA Scorecard shows “majority underperformance in 11 out of 17 fixed-income categories, topping out at 95 percent for actively managed government intermediate funds.”
Last year was not an outlier. An accumulating body of evidence shows that for the past two decades, actively managed public market strategies of all types failed to consistently provide promised returns. For example, over a 20-year period, almost 95 percent of U.S. large-cap equity funds underperformed their benchmarks on an absolute basis, while almost 97 percent of the same cohort failed to beat their benchmarks on a risk-adjusted basis. Fixed-income funds similarly underperformed their benchmarks over extended periods.
Additionally, there is little evidence that equity and fixed-income managers that beat their benchmarks will do so in the future. The S&P Dow Jones 2022 U.S. Persistence Scorecard concludes that “when success does occur, it tends not to persist,” the firm said in an email.
“Using ten years of data, managers with above-median results between 2013 and 2017 did not repeat their performance between 2018 and 2022; in fact, results were less good than we’d expect from flipping a coin,” Craig Lazzara, managing director of index investment strategy at S&P Dow Jones Indices, said in an email.
The steady flow of assets out of actively managed strategies and into passive alternative and private market strategies is evidence of investors’ continued frustration and disappointment with current active strategies. But for those advisors and their clients that remain committed to active management and are searching for a better path forward, now is the time.
And this takes us to the second point. The path to outperformance requires a persistent edge, and today that edge could be gained with new investment methods — specifically, certain types of artificial intelligence.
The recent release of ChatGPT has brought AI to the top of many investors’ minds. And while there are early indications that ChatGPT might help managers build better investment strategies, ChatGPT is just the tip of the AI iceberg. There are many other types of AI that can be applied to specific investment use cases. (I will leave aside the non-alpha-generating use cases, such as chatbots for client service, identity verification, and fraud detection.)
BlackRock summarizes possible applications in a recent report.
“Asset managers have developed AI and ML [machine learning] tools to compile, cleanse, and analyze the universe of data available, including analyst reports, macroeconomic data (e.g., GDP growth, unemployment) as well as newer ‘alternative’ data sources,” according to the asset manager. “Examples of alternative data include GPS and satellite imagery to see where consumers are going, internet searches and tweets to see what people are researching and talking about, and employee satisfaction data, all of which can be accessed online today. These data points can help portfolio managers better assess individual companies and sectoral trends.”
Many managers are already using a type of machine learning called natural language processing to parse audio related to business and finance, including product names, industry jargon, numbers, and currencies, giving managers a competitive edge in their decision-making, MIT Sloan School of Management finance lecturer Mikey Shulman said in 2020.
More advanced machine learning techniques that use deep neural networks — deep learning and deep reinforcement learning — have achieved superhuman results in other industries, and, as our experience at Rosetta shows, these systems can be used to build autonomous learning-based investment processes. (This source offers an overview of the various types of machine learning. Here’s another explainer.)
The main benefits of all of these machine learning models are that they can identify complex relationships in vast amounts of data that are undetectable to traditional, human-based investment techniques with programming; make incredibly accurate predictions in an entirely new way and without human biases; and, because they are self-learning, quickly adapt to changes in market conditions.
This self-learning, coupled with the idiosyncratic nature of each model, could result in a sustainable investment edge that persists over time — assuming the models are properly designed and tested. Building these models is as much art as it is science, meaning that two managers can use the same data and model type and build models that produce very different results.
It’s important that advisors understand and communicate to their clients that machine learning models are simply mathematical algorithms designed to be prediction engines. “It’s not a dark art — it’s math,” said Shulman.
Two recent polls of individual investors show they are ready for AI-based investing. According to a survey by The Motley Fool, “77 percent of high-income Americans have used ChatGPT for stock recommendations.” And a Morgan Stanley Wealth Management poll found that 72 percent of individual investors think AI will be a “game changer.”
This is not surprising. AI is deeply embedded in our daily lives — from navigation apps that predict the quickest route to recommendation systems used by streaming services and shopping sites.
However, AI is not a magic bullet that will instantly solve the problems of active management. Investment managers face significant barriers to adoption. For example, systems based on deep neural nets are difficult to design, develop, and commercially deploy, and they require special talent — talent that is difficult to attract and retain. Integrating this talent and the resulting models is also a challenge. And most importantly, there is no guarantee that even with the right talent, data, and models, the results will generate the desired alpha.
An investment manager’s fundamental duty is to provide clients with the best performance possible, even if it means a change in their view of investing.
And this takes us back to Father Schmid. Like him, financial advisors and their clients have endured decades of “harsh conditions.” Active managers using decades-old investment methods and processes will continue to struggle to find the edge necessary to outperform their benchmarks — on either an absolute or risk-adjusted basis. To quote the late Scott Minerd, CIO of Guggenheim Partners, “As an asset manager, I’ve come to view conventional wisdom as the surest path to investment underperformance.”
And like Father Schmid, it’s time for financial advisors and their clients to say, “This is pure bulls**t,” and start investing with managers that have abandoned traditional methods and processes and adapted to “modern life.”
