‘Huge role’ for quants in Covid-19 response – MIT’s Lo

Policy-maker actions or missteps will drive markets, academic says

Clark Kent was a mild-mannered reporter; Peter Parker, a lonely high school student. They were the perfect disguise for their alternate selves, Superman and Spider-Man, allowing them to evade attention when they weren’t saving the world.

So, who will save the financial services industry from Covid-19? Andrew Lo, the MIT finance professor, suggests another unlikely hero – quants. Their superpower, he argues, is the ability to act as a bridge between the world of science and the world of finance.

“There are few people in the financial industry that have the tools to be able to understand the epidemiological consequences of this pandemic,” he says.

“All of the statistics and data science that we use routinely in quantitative strategies can be applied towards understanding and building models of how the epidemic spreads, and what that’s going to do to financial markets and economies.

“There’s also an ability to communicate with epidemiologists that I think is really critical.”

That power could help the industry head off what Lo sees as a genuine threat to wider human society. 

“This is an all-out emergency,” he says. “Make no mistake, this crisis, in my opinion, has a far bigger set of consequences than the financial crisis of 2007 and 2008.”

“By the time we realise it’s an emergency, it’ll be too late to do much about it. We need to take the opportunity now while we can. And I think then we’ll have a fighting chance of dealing with this.”

A “giant petri dish”

Right now, Lo, who has written extensively about how markets react to stresses, sees uncertainty stalking markets for months or years.

It’s possible, for example, that successful containment in one country will fall prey to weaker efforts elsewhere, he warns. “In a few months, the rest of the world will be a giant petri dish that may very well reinfect China,” Lo says.

China then would struggle to contain further outbreaks without locking down immigration, which would be difficult given its global trade links, Lo thinks. Such complexity as it plays out, will continue to roil markets. “It’s going to be a very tough 12 to 18 months,” he says.

“The productive capacity of the global economy is still there. Eventually we will be able to contain the virus. We will be able to deal with it. It may impose a tremendous social, political and economic cost, but it will be contained. At that point, markets will recover.”

But recovery could take “five to seven years”, he says.

This is where quants and risk managers come in. In the short run, markets will continue to see elevated volatility, a flight to quality and pressure on credit spreads across the board. In the medium term, risk managers will need to look beyond their usual array of value-at-risk-type models for ways to quantify what is likely to happen.

“They need to start reaching out to epidemiologists and public health officials to understand exactly what the public health risks are and what the likely government responses to those risks will be,” Lo says.

“For example, if China really does try to curtail immigration to reduce the risk of reinfection, that’s going to cause major disruption to all sorts of supply relationships that occur across the board and we need to start calculating the implications for revenues, earnings and monthly credit.”

Waiting for decisive action

For now, those calculations are being made on the fly.

Fear of the Covid-19 virus that has so far infected more than 110,000 people caused an 11% selloff in global equities during the week of February 24 and triggered the worst post-crisis trading day for UK and US markets on March 9. Yields on 10-year US Treasuries have reached their lowest-ever levels as investors seek safe assets amid fears the epidemic will crush global growth.

What happens next will be driven by how policy-makers respond to the epidemic, Lo says. “Markets are going to anticipate what the consequences are of various different kinds of government action or inaction,” he says. “The fact yields are so low for US Treasury securities and volatility is high and that, even in the face of Fed interest rate cuts, we still see market declines: that is telling you that people are freaking out.”

Lo, whose adaptive markets theory explains market behaviour as the product of the often-competing rational and instinctive behaviour of investors, thinks the flight to safety is likely to continue.

“Until markets see decisive action that will reduce the kind of damage we are going to have from this pandemic, markets are going to continue to fall. My view is the worst is still to come, and it’s going to get quite a bit worse before it starts to get better.”

Lo believes that, apart from China, the response from governments so far is nowhere near decisive or co-ordinated enough. “That’s what markets are reacting to,” he says. Containment in other countries to the extent achieved in China would be the “best-case scenario”, he believes.

Funding a cure

Quants could also help combat the epidemic itself, says Lo, possibly structuring deals to secure private-sector funding for research into treatments or a vaccine. Lo has long advocated using financial engineering to encourage investors to back exploratory medical research.

This could be done quickly, he says. He draws a comparison with the investment of private institutions in structured deals to take on toxic assets during the financial crisis.

“We could actually create portfolios of drug development projects that could be funded and securitised to the point where tens of billions of dollars can be deployed relatively quickly… and where investors can earn a reasonable rate of return on their investment.”

“By itself, the private sector will not do this because the rate of return is not sufficiently high to warrant private sector investors to go into this market,” he says.

Distributing a vaccine in the US, assuming one can be developed, would require “tens of billions of dollars”, far more than the $8.3 billion so far assigned by US president Donald Trump, he adds. Lo was speaking before Trump trailed a new stimulus package, which is expected to be detailed today (March 10).

Lo is also calling for changes to statistical tests in drug trials to help get treatments to market more quickly. “My proposal is to change the approval threshold – the so-called p-value of statistical significance that one needs in order to allow a drug to be marketed,” he says.

Tough existing standards may be cutting off lines of drug development. “We can approve a drug that doesn’t really work but may have side effects. That’s a mistake. Or we can turn down a drug that actually does work. Those two types of mistakes have different costs, particularly in a situation like this where you’re faced with a global pandemic,” Lo says.

Changing test requirements would allow more potential therapies through, he argues, better balancing the rejection of ineffective drugs versus missing out on drugs that might work.

“Some may not work, so your false positive rate will go up without a doubt. But at this point, false positive rates going up a little bit is not nearly as problematic as missing out on a drug that could work and allowing tens of thousands of people to die.”

The Bayesian decision analysis approach Lo recommends for clinical trials is commonly used in investing to decide whether or not to shut down a losing trade, he notes – to decide whether losses are a temporary glitch or a permanent sign of declining profits.

Lo has been in conversation informally with policy-makers and biotech companies that are actively developing vaccines.