A pandemic like we haven’t seen in many decades, a global policy response unprecedented in living memory, and a dramatic decline in economic activity leading to a big impact on asset prices, all make this environment exceptional.
That being said, I’m a big believer in “history doesn’t necessarily repeat, but it does rhyme.” That means we can look at the past and see useful lessons for how best to navigate this environment and continue to produce risk-adjusted alpha for our clients.
As a systematic firm, generating alpha is the driving force behind everything we do – building quantitative frameworks to predict future returns, based on past relationships.
There are several lessons I believe apply in navigating the current landscape: Stick to your knitting, adapt to the environment, and manage risk.
Sounds like motherhood and apple pie, hard to argue against but not very actionable. What does this mean more specifically for a quantitative manager in today’s world?
Stick to your knitting
Our process selects securities based on numerous fundamental attributes and other data.
Our experience is that building predictive models based on a short, unusual period of time leads to lower predictive efficacy than using long-term data.
We wouldn’t want to build a model based only on the COVID-19 crisis period to help us pick securities. Instead, we want to use a much longer time period, of which March through May 2020 is only a small part. We are confident in the efficacy of this longer-term data set, because while the payoffs to our process are only modestly predictive in any single period, we gain more robust information by repeatedly making those predictions over time. As a result, we believe it makes sense to continue to apply our investment approach, despite the unusual environment.
This also helps us avoid behavioral errors induced by a myopic focus on recent events. For example, the COVID-19 virus originated in China, and before it visibly spread into Europe and North America, the Chinese market was hit hardest. At that point, a more myopic approach would have avoided China and invested more heavily into the apparently virus-free US equity market. However, since mid-February, the Chinese equity market has far outperformed developed markets.
Similarly, those who were concerned about a pandemic might have thought that companies that own and operate hospitals would do well relative to the broad market, yet HCA Healthcare, which operates hospitals, has dramatically underperformed the broader US market since the start of the pandemic, as have many of its peers.
Adapting to the environment
During the strong market response to COVID-19 that we saw in the second half of March, bid-ask spreads and transaction costs rose significantly. As a result, we reduced our trading activity during this time. Although price dislocations increased opportunity, in many cases it wasn’t worth paying three or more times the normal cost to transact. As spreads and costs have come down, our trading activity has picked up.
We believe one of the benefits of a systematic process is that it provides a continual and methodical trade-off between return expectations and the cost of acting on those expectations.
Another example of adaptation builds on my earlier comment about consistently applying our investment approach. While long-horizon historical information is the primary driver of how we weight different company attributes in building return expectations, our process also considers short-horizon signals, and the data we look at suggests we lean into the attributes that we are now emphasising. For example, the dispersion in valuation signals across companies is especially high right now relative to the post-GFC environment, suggesting future payoffs to value will be greater than normal, despite the poor returns to many value signals over the past several years.
Adapting to the current environment has us leaning into value somewhat relative to the past.
Managing risk
Finally, it is critical to manage risk. A strategy that on average provides a very attractive return with some variability can blow up if enough leverage is applied, wiped out by a poor draw from an otherwise attractive return distribution.
Similarly, betting only on one signal type – like value – or being overly concentrated in your portfolio positions can result in an outcome over a specific period that can’t be tolerated, even if the underlying investment thesis turns out to be correct in the longer-term.
Managing risk effectively in terms of concentration in individual securities, markets, industries, and across other risk factors is critical in producing returns at an acceptable level of volatility. Part of the skill in quantitative investment management is applying effective risk management with the minimum necessary impact on alpha.
Market outlook
Many of our clients want to know where we think markets are going to go.
The vast majority of systematic long-only equity strategies are run with an equity market beta of one, meaning that they don’t engage in market timing.
However, a number of systematic multi-asset strategies do. In this context, the directional equity models that Acadian employs in our multi-asset strategies are currently neutral in their outlook – while ongoing fiscal and central bank stimulus continue to be a positive for equities, and interest-rate relative valuations are currently attractive, there is high uncertainty about future earnings, a poor near-term growth outlook, and material credit risk.
A key point to emphasise as we consider this forecast is that we face an environment with a great deal of uncertainty, including a lack of clarity on future fundamentals, as well as great instability around the catalysts that may affect the economy and capital markets going forward. This is no time to be a hero.
Instead, what is prudent is to continue to apply a consistent and disciplined investment process, adapt it to the environment where appropriate, and manage risk carefully.
John Chisholm, co-CEO, Acadian Asset Management
