Over the past month, JPMorgan’s Marko Kolanovic penned nearly a half-dozen notes on COVID-19.
In those notes, Kolanovic modeled a variety of things, including, but by no means limited to, hospitalization rates, herd immunity and mortality rates. He also sketched out the broad contours of a national plan for pandemics, which would entail harnessing big data and machine learning in order to help humanity stay ahead of an outbreak.
I feel like I should remind readers – because sometimes this gets lost amid the media’s obsession with his calls on equities – Kolanovic isn’t just “the man who moves markets”, as CNBC famously dubbed him. He’s also a PhD in theoretical high-energy physics, and boasts top-cited research papers in quantum gravity (you can look those up on Google Scholar, if you’re so inclined).
The point is, this is an accomplished individual, who, one certainly imagines, takes his recent work on COVID-19 pretty seriously. In fact, I’d wager Kolanovic considers the notes excerpted in the linked posts above to be very important and worth the public’s attention.
Regular readers will note that Marko’s projections for the day of the New York coronavirus apex and the number of hospital beds weren’t just prescient, they were as close to 100% accurate as you can get under these circumstances, something he underscores in his latest note, out Tuesday.
“In our recent work, we used big data and a multidisciplinary scientific approach to forecast the date of the apex in hospitalizations and peak resource use”, he writes, adding that his work “suggested significantly lower mortality but higher attack rates”.
All (or most) of those forecasts are “now confirmed by actual data”, Marko says. “New studies indicate our mortality and attack rate analyses are on the right track as well”. He cites a Stanford paper on Santa Clara and a USC study on LA County. I make no claims to evaluating the veracity of those linked studies. The links are there for reason. You can take a look for yourself.
Whatever you want to say about the likelihood of a “second wave” of infections in the US or about the relative merits of reopening the economy (and regular readers know I’ve written breathlessly on both topics, not from a scientific perspective, of course, but from the perspective of asset prices and investor psychology), it’s now pretty clear the worst-case scenario for the first wave of COVID-19 isn’t going to play out in the United States.
That’s obviously not to suggest this hasn’t been a terrible human tragedy (it has), it’s just to say that if you look at some of the models being circulated last month of the worst-case scenario (in terms of fatalities, infections and other key virus metrics), it does not appear likely that the US will hit those dubious marks, or at least not in the first wave. (And thankfully so.)
Markets are responding to good news on the virus front and also to any and all headlines which suggest the economy may be back up and running ahead of “schedule” (whatever “schedule” means in this context). It helps tremendously that the Fed has committed to supporting the credit market and has pulled out all the stops to tamp down funding stress and ensure the underlying market “plumbing” remains functional.
Kolanovic on Tuesday reminds folks that in light of his assessment of the epidemic, and especially in consideration of the Fed’s support, he expects US equities to rebound. To wit, from his Tuesday missive:
After the Fed introduced the first wave of monetary easing, we expressed our view that the S&P 500 will reach its previous highs in the second half of 2021. This was based on our analysis of the pandemic and market structure (flows, positioning, and liquidity). Following the second wave of monetary measures, which included a backstop for certain credit assets, we shifted our forecast that the previous highs are likely to happen in the first half of 2021.
You might have been incredulous towards those calls, but probably not so much after the last two weeks, during which the S&P surged 15% in nine sessions despite some of the worst data in the history of modern economic statistics.
Never mind whether it makes “sense” – for now, it just is what it is. (Stocks have given some back in the new week amid the carnage in crude, but that’s another story entirely.)
Next, Kolanovic moves to address one of the most vexing questions about stocks in the current environment – namely, how does one go about valuing equities when earnings are zero?
As I put it on Monday, the problem is the distinct possibility that if we knew what earnings really were, we would discover that equities are trading not just at elevated levels versus temporarily depressed profits, but at a ludicrous multiple that can’t be justified by any otherwise plausible notions about “looking through” an anomalous quarter.
Marko doesn’t dodge this. In fact, he tackles it in a straightforward manner befitting of a mathematician. That is, he just states the facts as they are without even a tinge of emotion (which is a good thing – you don’t want emotion creeping into your investment decisions).
“What is the right multiple that should be applied given the unprecedented earnings hit, but also unprecedented monetary measures?”, he asks, before writing the following:
In a natural disaster such as a pandemic, earnings multiples should temporarily soar and then gradually decline as earnings recover. As an illustration, consider a business that needs to be temporarily closed due to an epidemic (e.g., hotel, restaurant, etc.). If the business is closed for 3 months, quarterly revenues will drop to zero. With the earnings near-term negative, if the business is to be worth anything (e.g., $1), P/E will effectively be infinite. Once the business opens, and earnings come back, the P/E will drop from infinity and normalize to some longterm sustainable level.
That’s correct. And the bit about buying stocks trading at a multiple of “infinity” is what has some market participants worried.
But Marko isn’t worried. Instead, he values the S&P using “various scenarios for the earnings decline and recovery due to the pandemic” on the way to quantifying the impact of the Fed’s actions on rates and credit spreads – so, the discount rate.
“The combined suppression of the risk free rate and credit spreads by the Fed likely has a bigger positive impact on equity valuation, compared with the negative impact of the temporary earnings loss”, Kolanovic says. Specifically, between rate cuts and the Fed’s overt support for credit markets (i.e., tightening spreads on the back of the central bank’s corporate bond buying programs which include IG, fallen angels and ETFs, both high grade and high yield), the discount rate has been pushed 100bps lower, from ~225bps to ~125bps.
“Given the massive suppression of the discounting rate, the present value of future earnings in all three earnings impact scenarios is above the pre-crisis level”, Kolanovic writes. “This indicates that the S&P 500 should attain previous all-time highs if the monetary measures are sustained”.
Below, find the scenarios rolled up into one table, along with Marko’s explainer on how to interpret it:
If the Fed only cut interest rates to zero, but did not reduce credit spreads, the discounting rate would likely be ~225bps (same as pre-crisis, i.e., lower interest rates offset by wider credit). In this case the fair value for the S&P 500 calculated as the PV of earnings should be down by ~10% compared with pre-crisis levels. Had the Fed not acted at all to cut interest rates or to reduce credit spreads, the discounting rate would be ~325bps and PV of S&P 500 earnings would be down ~20%. For countries that don’t have an ability to effectively reduce interest rates or credit spreads (e.g., non-reserve currencies, foreign denominated debt, no political agreement to backstop individual regions, or widening sovereign spreads), the impact of the COVID-19 shock would result in an increased discounting rate via both sovereign and corporate spreads (e.g., to 500bps) and PV of earnings would be down by ~30%. Our model also indicates, that if there were no COVID-19 pandemic, and the Fed cut rates to zero while credit spreads were constant, the market should have rallied 10%
As noted in the excerpt, this comes with the caveat that it doesn’t apply outside of the US. If you don’t control your own destiny (e.g., if you don’t print a reserve currency, if you’ve issued lots of foreign currency debt, if you’re overly-reliant on external funding and if you can’t simply treat the rate you pay on your own bonds as a policy variable), then the situation is largely out of your hands. That’s one of the reasons the IMF is stepping in to help countries without access to the Fed’s swap lines and foreign repo facility (see here and here). Also, the model represents an input for IG corporates and would need to be modified if applied to high yield borrowers or ex-US companies.
So, are there any risks to this? Asked differently: Is this set in stone? Can we now just buy equities indiscriminately, comfortable in the notion that stocks will infallibly be back to record highs by Q1 of 2021?
Well, of course not. Nothing is certain. And Kolanovic obviously acknowledges that:
A risk to this approach is that earnings disruptions are much longer than 18 months, that other events lead to rise of corporate credit spreads, or that the monetary support is withdrawn before a full earnings recovery.
Some of that simply cannot be known ahead of time. Or at least not by any mortal. Remember, he’s only “half-man, half-god” (to use another chyron from CNBC).
Or, if you’re a fan of classic American cinema, you might simply quote Bill Murray’s Phil Connors: “I’m a god. I’m not the God.”