What if the very qualities that make factor investing so compelling are ultimately responsible for driving “smart beta” premia to extinction?
Figure 1: New investment concepts follow the technology adoption life cycle
Source: Smith House. For illustrative purposes only.
Over the last few years investors have been clamoring for so-called factor, or smart-beta strategies. These strategies emerged from academia, predicated on the work of Treynor, Sharpe, Haugen, Fama, French, and others in the late 1980s and early 1990s. Many early hedge-funds earned extraordinary profits from these strategies in the 70s, 80s and 90s by allocating to stocks with strong momentum or deep value characteristics, or markets with strong trends.
In some ways new investment concepts are like any new technology. The progenitors of any early technology typically earn extraordinary profits until competition heats up. Eventually competition drives down profit margins and the technology becomes commoditized.
But investment technology has a special quality that arises from the adaptive nature of markets. This property means that the profitability of investment concepts conforms to a unique trajectory, which most investors haven’t accounted for.
Figure 2: Trajectory of new investment concepts
Source: ReSolve Asset Management. For illustrative purposes only.
Factor investing is predicated on the idea that an investment opportunity exists because securities with certain characteristics are systematically mispriced by a cohort of investors. This may be because these investors are influenced by unique preferences or perceptions of risk.
Smart beta / factor strategies typically derive their credibility from academic credentials and peer reviewed journals. A paper is published which describes an investment strategy with an intuitive origin story. A backtest and comprehensive analysis is presented with strong economic and statistical significance.
Word spreads about the new investment concept. A few big institutions jump on board. Innovative managers launch funds, which do well for a few years and catch the eye of more adventurous investors and advisors. A few more years pass. Now most institutions are running the strategy internally. Index providers have launched a mosaic of takes on the concept, many of which go on to inform new index ETFs.
There comes a point when the arbitrage dollars start to crowd out the investors who were creating the opportunity in the first place. The securities that were under-priced become over-priced. The sign of the edge inverts – it is now a money-losing investment!
What do you think happens next? Investors eventually cry uncle and abandon the strategy in droves. At some point, the market finds a new equilibrium premium that is just large enough to keep the most disciplined investors engaged, but much smaller than the original pre-publication premium.
Figure 3: Mass adoption in action → Imminent abandonment?
How about a concrete example to help to crystallize the concept?
Many of you will recall that a few of the most sophisticated institutions and hedge funds started running systematic alternative premia strategies in the mid-2000s to harvest the size, value, low volatility, trend, momentum, carry and other so-called “factor” strategies.
These innovators and early adopters harvested rich premia for a few years before the ideas went mainstream.
By 2011-2012 the investment banks had launched alt premia indices and major investment managers launched funds, attracting tens of billions of arbitrage capital from the Early Majority investors. These billions were typically levered up 5x-10x.
By 2015 most major institutions had built or were building internal desks to harvest these premia and eliminate fund fees and retail investors were getting in on the action via an array of index funds.
Fast forward to 2016-2017. Alt premia strategies were broadly adopted and many of the largest Style Premia funds started to close to new investors. The Late Majority was “all-in”. Peak Alt Premia. What happened after 2017?
This chart plots the cumulative alpha from a benchmark combination of leading alternative premia and systematic multi-strategy funds from January 2018 through July 2020. We scaled the funds to express equal risk in the portfolio and called it the Alt Premia Benchmark.
Figure 4: Cumulative alpha of Alt Premia Benchmark*
Source: Data from Bloomberg. Analysis by ReSolve Asset Management. *Chart represents cumulative alpha of an Alt Premia Benchmark that combines an alternative risk premia fund and a systematic multi-strategy fund weighted for equal volatility. Check disclaimer for constituents.
We also overlayed a cone that charts the trajectory of a random walk with zero return and the same volatility as the Benchmark. Over the past 30+ months this composite – representing over fifteen different alternative premia sleeves – has produced a return trajectory that falls well below the lower threshold of the cone.
We’re reminded of a colorful anecdote from Nassim Taleb’s book, “The Black Swan” starring the characters Dr. Bob and Fat Tony. A third party asks them to assume that a coin is fair, i.e., has an equal probability of coming up heads or tails when flipped. I flip it ninety-nine times and get heads each time. What are the odds of my getting tails on my next throw?
Dr. John is dogmatic in his belief that the rules from a theoretical model of dice throwing must apply. He says that the odds are not affected by the previous outcomes so the odds must still be 50:50.
Fat Tony says that the odds of the coin coming up heads 99 times in a row are so low that the initial assumption that the coin is fair must be false. He figures, “The coin’s gotta be loaded!”
The chart above implies less than 1 chance in 1000 that the returns from our alt premia benchmark are drawn from a distribution with a positive mean. Dr. John says this is just an unfortunate coincidence. Fat Tony can’t help but conclude that something fundamental has changed.
So guess what – our conversations with institutions and consultants makes clear that investors are heading for the exits. Pension and endowment funds are dismantling factor desks and major wire-houses have started de-listing associated funds.
It’s ok – these premia were legitimate and the market will eventually find an equilibrium that compensates arbitrage investors for taking on unwanted risk from other classes of investors. The premium will be lower – probably on par with other major premia like the equity risk premium or the duration premium. But commoditization will drive costs down commensurate with lower long-term returns.
“The signals that we have been trading without interruption for
fifteen years make no sense.
Otherwise someone else would have found them.”
Robert Mercer – former co-CEO of
So what’s the lesson?
Investors seek comfort in economic intuition, expert opinions, peer reviewed academia, and recent performance. Sadly, these are the very qualities that destroy future returns. Alpha lives in the crevices and dark corners; lonely places where most investors don’t want to go.
If asset owners and investors want to earn excess returns then by definition, they must come to grips with the reflexive nature of markets. Where a strategy offers comfort in the form of published research and peer adoption, with simple mechanics, compelling backtests and intuitive stories, we should expect the market to quickly mediate this opportunity. There is no free lunch!
Figure 5: From naïve factors to bespoke factors
Source: ReSolve Asset Management. For illustrative purposes only.
So what works?
At ReSolve we are reformed factor investors. Where factors typically have intuitive explanations rooted in economic theory, we source edges directly from the empirical data. Where factors are predicated on common relationships across all assets, we seek patterns in the data that are unique to each market. Where factor relationships are simple in structure, we seek complex relationships that are difficult to spot. Where factor strategies rely on long-term average investor behaviour, we evaluate how markets respond under different conditions. And perhaps most importantly, where factor strategies are evaluated on typical in-sample backtesting methods, our strategies are validated by the advanced out-of-sample and hold-out methods used in machine learning.
Most importantly, investors must recognize that in adaptive markets the only sustainable edge is constant innovation. That means constantly seeking information sources that explain market returns from different angles. This involves many of the same features that factor investors use, as well as new data points sourced from, for example, the volatility surface, dealer gamma, dark index flows, cross-market information, and other alternative data sources.
But new information sources alone don’t produce alpha. You must have the infrastructure to constantly mine-for and select the best new strategies, weed-out antiquated edges, and pipe the constantly adapting alpha engine through to production.
While these concepts may be intuitive to some investors, we realize they can sound controversial and even uncomfortable to many. We have been debating these ideas internally for some time, and decided it was time to share our thinking. This is hopefully the start of a conversation that we intend to have with our clients, peers and the broader investment community. To that end, we have recently released a podcast on this theme, Resolve Riffs on a Post Factor World, (click here to watch or listen.) We certainly welcome your feedback.