Monday, March 9, 2020

A Viral Market Meltdown Part II: Clues in the Debris!

Update on 3.9/20: In a sign of how volatile times are, over the weekend, oil prices plummeted to close to $30, the treasury bond rate to less than 0.4% and the market looks set to drop substantially. A work in progress indeed....

I wrote a post on how the Corona Virus was playing out in markets on February 26, two days into the market going into convulsions, and while I tried to make an assessment of the value effect, I also said that this analysis was a work in progress, that I would revisit as we learned more about the virus and its economic consequences. Eleven days later, we still don't have clarity on the health or economic effects of the virus, but we do have substantially more data on what the market reaction has been. In this post, I will begin by doing a quick update on the viral spread across the world, but spend more time on the market damage, looking at where it has been greatest, seeking clues for the future. 

A Virus Update
In the last week and a half, the virus has clearly expanded its global footprint, with Italy and South Korea now in the front lines, in terms of exposure, but with the numbers climbing rapidly across the rest of the world, it is clearly now on its way to becoming a global pandemic. 
NY Times, as of March 6, 2020

While the word "pandemic" alone is often enough to drive us to panic, it is not the first, nor will it be the last, and it helps to gain perspective to compare it to pandemics in the past, both in terms of contagion and health consequences. This chart from the New York Times reflects what we know about the virus as of February 28, 2020:
Note that the large band of uncertainty around the fatality rate related to the virus, reflecting how little we know about its potential consequences and how it measures up against other viruses in terms of contagion. Put simply, this is not just the common flu with side effects, as some have argued, but it is perhaps not the deadly killer that others at the other extreme has painted it as. The X factor that makes this virus potentially more difficult to contain and more likely to have global consequences is globalization, one more argument that populists will undoubtedly use to argue against it. The reality is that travel, especially across borders and continents, is not only easier than ever before but also more affordable, as income levels rise in the developing world.  Over the next few weeks, it is likely that we will see the case numbers rise dramatically in countries which have been hitherto exposed only lightly to the virus, the fatality numbers will rise among those affected, and health systems around the world will come under pressure. 

A Market Update
Over the last three weeks, we have had a glimpse of how quickly market moods can shift. Looking at the major US equity indices, you can see the euphoria that resulted in the market peaking on February 12, 2020, not only faded quickly but has been replaced with panic and desperation:
If there is one thing that can be said about markets during this tumultuous period, they were not playing favorites, since all of the indices registered double git drops, with the NASDAQ showing the smallest drop.

a. Melting Away - Dollar Value Lost
The focus on the indices can obscure the staggering decline in market values that occurred in a three-week period and in the table below, I chronicle the loss in market value globally, broken down by region.
Download spreadsheet
The first four columns look at total market capitalization and the change in both dollar and percentage terms between February 14, 2020 and March 6, 2020. Globally, equity markets lost $7.3 trillion in value over this three-week period, and it is ironic that China, the starting point for the Corona Virus, is the only part of the world where stocks have collectively seen an increase in market capitalization. That can be explained perhaps by the fact that Chinese stocks had already registered drops in the weeks leading into February 14, and that the rest of the world is playing catch up. The last five columns look at the percentage change in individual stocks to illustrate how widely the pain was felt. In ten of the twelve regions, with China and Africa being the exceptions, less than 25% of stocks went up during the three week period. In most of the markets, the percentage change in overall market capitalization is similar to the percentage change in the median stock, indicating that this is not a decline being caused by a subset of stocks being hit with extreme price movements.

b. The Sector/Industry Breakdown
There is no question that the virus not only has the potential to hurt the global economy, but the hurt will be felt disproportionately by companies in different businesses. To assess how the market has repriced different sectors, I look at the market capitalization lost, in both dollar and percent terms, by sector, for global companies:
Download spreadsheet
The biggest losers were energy and financial service companies, and the sectors that performed the best were utilities, health care, real estate and consumer staples. Breaking down the sectors into more detail, I looked at US stocks, by industry, and the following is the list of the five worst and five best performing industries between February 14 and March 6:
Download spreadsheet

The full list is available for download by clicking here. For anyone who has been following the news stories of airlines scrambling to cancel flights and mollify passengers and hotels dealing with cancellations, it should come as no surprise that aviation and hotel stocks were the worst performing industry groupings, followed by oil, broadcasting and life insurance. The best performing industry grouping also carries no surprises, with precious metal companies benefiting from the rise in gold prices, grocery retailers and tobacco drawing on their strengths as non-discretionary products and biotech companies benefiting from the focus on a solution for the virus.

c. Size Classes
The conventional wisdom, when there is a market crisis, is that investors move their money to safety. While that has clearly happened with money shifting into US treasuries, the question is whether investors are abandoning smaller companies for larger ones, presumably driven by the perception that smaller companies are riskier than larger ones. To answer this question, I looked at all global companies, broken down by market capitalization into ten classes:
Download spreadsheet
The results don't line up with expectations, as small companies saw a small increase in overall market capitalization and large cap stocks registered the largest decline. It is worth noting that even among the smallest stocks, the median stock lost 7.73%, suggesting that the increase in value is coming from a small percentage of stocks in the group. (Looking at just US stocks, you get very similar results.) 

d. Value and Momentum Classes
The drop in the market has provided some measure of vindication to those who have long been arguing that the market is over priced, but while the fact that the market was priced so richly set it up for a larger fall, breaking down the decline in market cap into classes can provide us some insight into whether the stocks that had gone up the most were the ones that saw the biggest drop off in value between February 14 and March 6. In the table, we break global stocks down into ten classes based upon the price change in the year prior to February 14 and look at the change in market capitalization, by class:
Download spreadsheet
In keeping with the story that what goes up the most must come down the most, you find that stocks that had performed the worst in the year leading into February 14 had an increase in market capitalization, though the median stock was still down, within this group. Using another proxy for rich pricing, I also broke stocks down by PE ratio classes from lowest to highest, based upon market capitalization on February 14, 2020, and looked at the change in market value between February 14 and March 6:
Download spreadsheet
Here, the evidence contradicts the market correction hypothesis, since there is no discernible relationship between PE ratios and market value change. In fact, the best performing stocks are in the top two deciles of PE ratios.

e. The Rest of the Story
One of the perils of getting focused on equity markets is that you can miss all of the action in other markets, and the changes in those markets can not only help augment the story that equities are telling us, but they can yield insight into other facets.

I. US Treasury rates
If the drop in stock prices over the last three weeks took your breath away, the shifts in the treasury market were even larger and more unsettling:

The 10-year US T.Bond dropped below 1% for the first time in history on March 3 and continued trending down to settle at 0.74% on March 6. In tandem, the other treasuries also dropped, bring the US dollar risk free rates closer to the Euro and Yen risk free rates. While some of the decline in rates can be attributed to a flight to safety, there is also a much depressing read of the same drop. To the extent that long term risk free rates are proxies for nominal economic growth, the treasury bond market seems to be signaling not just a shock to near-term economic growth from the Corona virus, but a long term decline. We will get a better sense of what the bond market is expecting, once equities settle in, but if the 10-year rate stays below 1%, it is not a good sign for the economy.

II. Gold and Bitcoin (Millennial Gold)
The other asset class that always attracts attention and money during crisis is gold, and for good measure, I will also look at Bitcoin, which some have suggested is the millennial equivalent of gold:

It is perhaps a little unfair to draw a conclusion from just contest, but the fact that Bitcoin has behaved more like stocks than like gold suggests that millennials who have held on to it, as their asset of refuge, may want to rethink their positions. 

3. Oil and Commodities
The final piece of the market puzzle comes from the commodity markets, with oil as its front runner. In the three weeks which have taken equity markets on a ride and caused US treasuries to hit new lows, oil prices have been on a journey of their own:

Not only have oil prices dropped 20% during the three weeks, they are plumbing depths seldom seen in this century. The decline in oil prices not only reflect an expectation of global economic slowdown but also how dependent oil and other commodity prices have become on China's continued growth and prosperity. The smaller decline in natural gas prices, much less tied to the Chinese market, reinforces this argument.

Revisiting the Viral Value
With this long lead in, you might have lost interest already, but if you are still reading, it is time to turn to specifics and look at how what I have learned in the last 12 days has or has not changed my views on the market.

Recapping the Drivers
Very quickly recapping what I argued were the drivers of the value of stocks, I argued that there were three components to value:
  1. Earnings Growth: In my 2/26/20 valuation of the S&P 500 index, I argued that the corona virus is now almost certain to cause earnings effects for companies, and estimated the drop to be 5% (a significant revision down from the 5.52% growth that had been predicted in the index. In the last few days, analysts have started adjusting earnings expectations down for companies, and this snapshot from Zacks today captures some of the adjustment:
    Note that the expected earnings on the index for 2021 has dropped from 172 for next year, two weeks ago, to 163 this week, matching the earnings generated in 2019. That is still better than the 5% drop that I was projecting, but my guess is that I am still undershooting the actual earnings decline and I have increased the expected earnings drop in 2020 to 10%. To complete the assessment of growth, I also need to estimate how much of the earnings drop in 2020 will be recouped in future years. In my valuation on February 26, I had estimated that half of the earnings drop in 2020 would be recouped but that the rest would be lost for the long term. I will continue to hold on to that assumption In addition, since my long term growth rate converges on the US T.Bond rate, the precipitous drop in that rate has lowered my growth rate in perpetuity to 0.74% (to match the T. Bond rate).
  2. Cash flow Payout: The second component of value is the cash that companies can return, in dividends and buybacks. I assumed that companies, driven by uncertainty, would scale the percent of the earnings that they return to stockholders from the 92.33% that they were returning prior to the crisis to 85%, more in line with the ten-year average. In the days since, there have been no announcements of dividend cuts or scaling back of already announced buybacks, but I would not be surprised to see that change in the next few weeks. 
  3. Discount Rate Dynamics: The discount rate dynamics are the trickiest. On the one hand, the lower T.Bond rate will create a lower base from which to build up, but the increase in volatility (actual and expected, as captured in the rise in the VIX over the last three weeks) has pushed equity risk premiums up. I will scale up my ERP to 5.69% to match my implied premium at the start of March 2020.
With that combination of assumptions (10% drop in earnings, 50% recoupment between 2022-25, 85% cash return and a 5.69% premium), the value that I derive fo the index Is 2889, and much of the reason for the drop from the value that I estimated on February 26, 2020, can be attributed to the the lower growth rate that I am estimating in the near term and in the long term.

Value Dynamics
In the days to come, there will be more information that comes out about not only how the virus is spreading across the globe but also on its consequences for businesses and economies. To provide a measure of how this will affect stock values, I computed the value of the S&P 500 (which stood at 2972.37 on March 6, 2020) as a function of what I believe are the two big uncertainties; the effect that the virus will have on earnings in 2020 and how much it will affect long term earnings growth:

Download valuation spreadsheet
Note that the big concern, if you are an investor focused on value, is not how much the Corona virus will affect earnings this year, but how much of that earnings drop is permanent. If you are in the camp that believes that there will be an earning drop, but that it will be fully recouped, stocks look cheap even if earnings drop by 20% in 2020. Conversely, if you believe that this earnings drop is likely to be permanent, with none of the drop being recouped, the value drop will be more closely linked to the earnings drop and suggests that there is more pain ahead for the market.

What to watch for..
Needless to say, there will be plenty of distractions in the coming weeks, but my suggestion is that you stay honed in on the value determinants, screening news stories for consequences for these determinants. In particular, I plan to watch the following developments:
I know that my view that T.Bond rates staying low and getting lower is not a positive but a negative for stocks puts me in the opposite camp from those who believe that the Fed will be the savior. When rates are as low as they are, central banks are more helpless bystanders than powerful trend setters, and the message about future growth that is imputed in low rates more than drowns any short term positive effects. 

The Big Things in Life
As I write this analysis of how the virus can affect stock market values and portfolio returns, I am aware that there is a human toll that it is taking that makes any market effects seem trivial. If I were given the choice, I would trade a large market drop for a small loss of lives and a quick passing of the virus. At times like these, I am reminded again of the fragility of life and the importance of good health and family. Be well, Godspeed and please wash your hands!

YouTube Video


Datasets
  1. Market Damage, by region, sector, industry, size and momentum

Spreadsheets
  1. An Updated S&P Valuation Spreadsheet: March 6, 2020

7 comments:

Anonymous said...

Hi Aswath,

Thank you or another great post. In your post Part 1 you had a table where you analyzed where you said you feel the market is still overvaluing the S&P 500. Could you publish a revised version of the same or any comments around it given the new data you have?

Anonymous said...

Another great post. Thanks for the update. I looked at the tab in your model that has earnings data for the SP500. There are numerous years when earnings dropped more than 10%, most notably 2008, when they dropped a whopping 40%. I think using a 10% drop for 2020 may end up being too conservative. Wall Street analyst don’t do any investigative work other than accept guidance from company managements so their estimates are a moving target. The good news that I saw in the data, which goes back to 1960, is that in just about after every drop, earning recover back to pre-drop level within three years.

Anonymous said...

Professor: Does it make sense to use the current Long Bond rate as a proxy for "Expected growth rate in the long term?". One, it is at extremely low levels and I mean compared to ever. Two, it is too volatile at the moment due to extreme fear. Would it not make sense to use a long term average or something that represents a median view instead of this current environment?
Thanks.

Yoon-Seri (Seri's Choice) said...

Dear Prof:

Great post, as usual! :)

I believe the decline in oil prices is not just due to the dependency on China. But a significant portion of this decline is due to the ongoing OPEC+ supply disagreement between Saudi Arabia & Russia.

Manu Bansal said...

Hello professor. I had one question. While I understand the logic behind listing cash flow payout as a value driver, wouldn't companies be better off retaining and reinvesting their internally generated cash, especially if they are in the growth phase?

PB said...

Hello Sir,

Thank you for a great summary of what is going on in the market.

Could you comment on how the current situation is affecting the derivatives markets? I reckon they might be benefitting from the volatility we are seeing in the markets.

PB said...

Hello Sir,

Thank you for a great summary of what is going on in the market.

Could you comment on how the current situation is affecting the derivatives markets? I reckon they might be benefitting from the volatility we are seeing in the markets.