Yet investors often misunderstand the extent to which they’re diversified and the strategies that can provide actual diversification, so they can end up disappointed by the performance of what they think of as diversifiers. “For instance, when volatility spikes, equities go down, but a diversifier could, in equal measure, go up, down or remain unmoved. Many don’t understand that being up when equities are down is a hedge rather than a diversifier,” he said.
Overestimation
“Equity outperformance over the last 10 to 15 years is a statistical fluke. There’s been a lot of support from monetary policies that have pushed asset prices up,” Seager said. “By its nature, that upward pressure is a fluctuation that is unlikely to continue, and we will see a reversion back to the mean, which is a level of risk-adjusted returns that is lower and more consistent with what we’ve seen historically.”
Also, in the rising-equity environment, many investors dived into private market strategies “that are really just leveraged versions of equities,” said Seager. “People have moved aggressively into the private markets space thinking they would be diversified, but those returns come from the equity premium. You also get some premium from holding an illiquid position, but there’s not much more to it than that.”
“There’s actually a lack of diversification, and portfolios are nowhere near as diversified as people think,” he said.
Poor correlation
Neither do fixed-income instruments always exhibit the diversifying features that they once did. While fixed income was steady for a period after the global financial crisis, mainly because inflation was low, in recent years the Federal Reserve has been tackling an inflationary regime.
“It might look like inflation is tamed, but fiscal policies are not helping to get it under control, and as long as we see inflation, real interest rates can be pushed lower, which isn’t good for fixed income, and it’s not good for equities either,” he said.
While the correlation between bonds and equities over the last 20 years has been mostly negative, according to CFM’s research, it is a statistical outlier. “The correlation between bonds and equities has actually been positive through most of history,”1 Seager said.
Therefore, “you do not get diversification or a hedge from bonds, and since the recent rise in inflation, that negative correlation has flipped and become positive. On a forward-looking basis, you get less diversification from a 60/40 portfolio than was previously the case,” he said.
Read: Bond-equity correlations: Are the times a-changin’?
Pragmatic process
CFM takes a practical approach to the use of quantitative strategies that deliver diversification. “We’re not guided by any sort of dogma about what should work. It’s based on trying to maximize risk-adjusted returns on a forward-looking basis by forecasting markets,” Seager said. In certain situations, one data set could forecast the market on a relative-value basis, another could forecast on a directional basis, and some could forecast both.
“We look at a wide set of markets and asset classes, including equity indices, fixed-income instruments and commodities. We do it empirically, and we use what works best in each environment,” he said.
The starting point is forecasting an instrument — including futures, options, credit or a single stock — which presents many different strategies for the firm to pursue. “The more strategies you have, the better to broaden diversification,” he said.
The key to delivering diversification is the ability to understand the correlations among all the instruments. “We’ve been developing and fine-tuning this process for decades,” he said. CFM has two research groups that work closely together, one for forecasting and the other for portfolio construction. We “combine all of these price forecasts and construct a portfolio that’s as robustly diversified as possible and trades with the lowest execution cost.”
The ‘black box’
One concern that CFM hears from prospective clients is a general mistrust of algorithms versus human input in the investment process. “Algorithms have become ubiquitous in daily life, and advances in data-driven techniques have dramatically improved people’s health, safety and quality of life. Yet in many situations, people fear they will underperform” as an investment process, said Seager.
“Many fear that a systematic process is too ‘black box.’ On the contrary. There is nothing more black box than the human brain, which is prone to biases and demonstrates less repeatability,” he said.
The best investment outcomes, according to Seager, depend on a repeatable, systematic process that reacts in a rules-based fashion in response to inputs from the market. “Our research process finds systematic patterns that occur and recur, and those patterns are then implemented in an unbiased fashion — even to exploit those inefficiencies in markets that are caused by human biases.”
“Data is becoming ever-more voluminous and available for consumption, and it gives quants a significant advantage over discretionary investors, because their investment decisions are better informed,” he said. Another advantage for a quantitative investor is Large Language Models that bridge the gap between quantitative and discretionary approaches. “We now have algorithms that trawl through mountains of text data to evaluate market sentiment and detect investment signals,” he said.
That said, CFM’s biggest asset is people, Seager said. “We recruit people with science backgrounds alongside those who have financial market experience so we can build algorithms with the right data and get the best outcomes.”
“We’ve been building strategies that are truly uncorrelated with traditional markets and provide genuine diversification since the early ’90s. It’s always the right time to diversify, but today it’s even more important than before.” ■
