In parsing the latest developments in the ongoing pandemic and its related effects on market volatility, it may be useful to consider an often unfortunate characteristic in human behavior.
There are roughly seven million girls and boys that play high school sports in the United States. Ask any one of them and, on average, 53.7 percent believe they are at least “somewhat likely” to receive a college scholarship for their given sport.1 That’s right…take your local high school soccer team of 20 boys, and a little more than 10 believe they are at least “somewhat likely” to make it to the next level and receive financial support for doing so.2
Unfortunately, the reality is something very different. There are only 126,000 student-athletes that receive either a partial or full athletic scholarship.3 That’s 1.8 percent of the high school athlete population.
How can that be? Less than 2 percent of athletes receive scholarships in college, but more than 50 percent believe they will?
Those crazy kids. Yes, well, let’s look at adults. Ninety-six percent of American drivers believe they are better drivers than the median driver.4 For the astute reader, yes, that is statistically impossible because, well, the definition of the median is the middle, and it’s impossible for everyone to be better than the driver with median skill.
Yes, let’s meet the bane of our existence as investors: Overconfidence. Our friends at Wikipedia define it as:
“A well-established bias in which a person's subjective confidence in his or her judgements is reliably greater than the objective accuracy of those judgements, especially when confidence is relatively high. Overconfidence is one example of a miscalibration of subjective probabilities. Throughout the research literature, overconfidence has been defined in three distinct ways:
Overestimation of one's actual performance;
Over-placement of one's performance relative to others; and
Over-precision in expressing unwarranted certainty in the accuracy of one's beliefs.”5
Among other things, overconfidence has been blamed for the sinking of the Titanic, the Deep Horizon spill, the nuclear accident at Chernobyl, and the list goes on and on.6
So, if it’s not just kids, and it’s not just adults … then maybe it’s just humans?
It turns out we humans have a remarkable ability to over-predict our skill and, unfortunately, we tend to make poorer decisions as a result. “Sure, I can make that yellow light”, “no, this ship is unsinkable”, and “I better sell now because this market is surely going down further” are some of the more notable examples.
I was reminded of this recently when reviewing investor behavior during the market sell-off. In February, investors withdrew roughly $20 billion from U.S. equity funds, while placing roughly $31 billion into money market funds.7 From the low in March, the Standard & Poor’s 500 (S&P 500) has thus far returned +45.3 percent.8 Said another way, investors that sold as the pain was at its apex missed one of the greatest equity market rallies in history.
The more interesting question, perhaps, is was it right to sell? And by extension, should we be selling today?
To explore those questions, what follows are two sections. The first section reviews the latest data on the pandemic, and the second discusses what we know and explores our decisions.
With that, let us begin.
The status of the pandemic
As of the morning of July 219:
- The novel coronavirus (COVID-19) has infected more than 14 million people and killed more than 611,00010 people worldwide.
- In the United States, there are now more than 3.9 million confirmed cases along with more than 141,000 deaths.11
- Globally the countries with the highest growth rates continue to change order, with the more developing countries now growing the quickest (Brazil, India, and South Africa of recent concern).
Chart 1 demonstrates how these numbers have changed over time.
The blue lines demonstrate the total number of cases around the world and correspond with the left axis. The grey line demonstrates how the number of cases is growing and corresponds with the right axis. On March 22, growth rates were roughly 20 percent per day.8 Today, those growth rates are consistently between 1 percent and 2 percent, and have averaged 1.2 percent since the beginning of June.13
Perhaps more intuitively, we can use those growth rates to calculate how long it takes for global case numbers to double. In the middle of March, for example, global cases were doubling every four days. At that growth rate, if you start with 100,000 cases today, in four days, we would have had 200,000 cases. Today, given the low growth rate, we are now doubling the global number of cases every 50 days—clearly a tremendous improvement.
Within the United States, in aggregate, the picture is similar, although we have had higher growth rates more recently (more on that in a moment). On March 21, the U.S. growth rate was 40 percent. As of July 20, it is now hovering consistently between 1.7 percent and 2 percent.14 One month ago, the number of cases doubled every 1.8 days; today, the number of cases is doubling every 38 days.
Chart 2 puts those growth numbers in context by showing how the number of days to double has changed over the past several months.
This then brings us to the bad news. The growth rate in the United States has, after falling precipitously for three months straight, begun to rise again. On June 15, we were doubling the number of cases every 64 days, and today it is 38 days, where it has hovered for the last four weeks.
Chart 3 zooms in on the most recent COVID-19 growth rate in the United States and demonstrates the materiality of that change.
Clearly, the United States made incredible progress, but that progress has stalled and begun to revert. As such, as with any confusing time, positive and negative anecdotes abound. New York went from averaging 11,000 cases per day in mid-April to now averaging 600 cases per day. Connecticut went from averaging more than 1,000 cases per day in mid-April to now less than 100 cases per day. On the negative side, Texas went from averaging 1,000 cases per day in mid-April to now averaging more than 9,000 cases per day. Florida went from a similar 1,000 cases per day in mid-April to now averaging more than 10,000 cases per day.17
Thus, netting all states together, we see a massive decline in the growth rates in April and May, and then a slight increase in growth rates since the bottom in mid-June.
Does this foretell a reversion to where we were?
It’s not clear, but it doesn’t appear likely. We, in aggregate, have learned much about this virus and have adapted in remarkable ways. Unless we completely throw caution to the wind (which, to be clear, has happened in several notable states), in aggregate I remain optimistic that we will be able to contain its growth and mirror some of the more successful policies already adapted in other countries.
The market question
With that, let us now turn to the question we started with. Were the investors that sold in February and March correct? I strongly believe the outcome is not what dictates the quality of the decision-making as most decisions are made under uncertainty. If true, then we should evaluate what we knew at the time, and if we would make that decision again.
Therefore, let’s try a mental exercise.
Imagine we dropped into a sophisticated and developed country with one of the world’s most successful economic systems. At the time of this hypothetical analysis, the stock market is relatively expensive by most measures, and the bond market is similarly dear. All at once, everything changes. People become ill through some new illness and roughly 140,000 citizens are tragically lost. Schools close, businesses have their employees working at home, many restaurants and bars are closed, airlines are operating at 10-20 percent of capacity, and international travel is almost non-existent. Unemployment is at record highs, economic activity has contracted at a pace never seen, and the government has pumped trillions of dollars into the economy.
Where, therefore, do we expect the stock market to be?
- Down 50 percent?
- Down 25 percent?
- Down 10 percent?
Businesses and consumers have clearly changed their behavior, and economic activity is clearly dramatically different. In the long run, by definition, market valuations only represent the expectation of future cash flows … so haven’t those changed?
Well, in this instance, no, they haven’t. That synopsis is roughly what has occurred in the United States and, in case you haven’t noticed, the market is basically flat for the year. That’s right, had you invested $100 on Dec. 31, 2019 in the S&P 500, today, you would have roughly $100. Chaos has ensued, volatility has abounded, hundreds of thousands have died worldwide, and the market is largely acting as if this is now over.
Is that right? Is the market priced appropriately? Were the investors that sold in February correct, or were the investors that purchased in March incorrect?
In short, I have no idea.
What I do know is that historically, markets are remarkable generators of capital over the long term, and volatile and unpredictable in the short term.
While there are moments when I am sure that the market is overpriced (as I was in ‘96 or ‘06), I have more often discovered that the market was stronger at generating reliable returns than I am at predicting turning points. In 2006, I was right, but two painful years early, and had I sold or changed my positioning, would have missed remarkable returns.
I leave us then with a chart I have referenced in times past. Chart 5 is the result of research we performed analyzing the worst sell-offs in market history. We went back to 1920 and retrieved the periods where the market fell the most (denoted by the orange bars). We then looked at how the market had performed over the subsequent one-year, three-year, and five-year periods (denoted by the various shades of blue bars).
There are several takeaways:
- First and foremost is that every sell-off in this chart has resulted in subsequent returns that were positive over a reasonable time period.
- Second, the returns were quite significant.
Will this market be the same? Should we sell? Or should we buy? I would point you to the anecdotes about over-confidence. We, as humans, often believe we know more or have more skill than we do, and we often make poorer decisions than we could otherwise. With regards to markets, investors often earn less than the index precisely because of this reason. We sell when the pain is high, and we buy when the pain is low.
Right now, I would contend there is more risk to the downside than the upside, and I believe we are likely to see more volatility in the short term. Therefore, please keep the following in mind as we continue through this uncertain period:
- Planning trumps timing — Never count on timing the market. Even the most seasoned professionals don’t do it well, and that leaves everyone else with a negative expected value.
- Asset allocation matters — Maintaining a thoughtful and diversified portfolio can help you reach your goals. Diversification is designed to help dampen the extremes and can help you stay on track during the most difficult times.
- Control what can be controlled — Consider keeping costs down, minimizing taxes, and being thoughtful about risk tolerance. We cannot control the current crisis, and we won’t be able to control the next crisis. As such, focus on what can be controlled, and rely on capitalism to help generate higher returns than inflation over the long term. Stay focused on your long-term objectives rather the short term … the short term is both confusing and perilous.
In closing, stay safe, stay calm, and please turn off the financial news channels.
Learn more from MassMutual …
4 Ola Svenson, ‘Are We Less Risky and More Skillful than Our Fellow Drivers?’, Acta Psychologica, 47 (1981)
6 Ashraf Labib and Martin Read, ‘Not Just Rearranging the Deckchairs on the Titanic: Learning from Failures through Risk and Reliability Analysis’, Safety Science, 51.1 (2013)
8 Sources: Bloomberg, as of July 20, 2020
9 https://worldometers.info/coronavirus, as of July 21, 2020
10 Johns Hopkins University, https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6, as of July 21, 2020
11 https://www.worldometers.info/coronavirus/; as of July 21, 2020
12 Sources: Bloomberg, World Health Organization
13 Source: Johns Hopkins University, https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6
14 https://www.worldometers.info/coronavirus/country/us/, as of July 21, 2020
15 Sources: https://www.worldometers.info/coronavirus/country/us/, as of July 20, 2020
16 Sources: https://www.worldometers.info/coronavirus/country/us/, as of July 20, 2020
18 Sources: Bloomberg, as of July 20, 2020
19 Sources: Bloomberg, as of July 20, 2020