I recently read this post by the Brooklyn Investor about the normative impact of COVID-19 on equities. The post itself was excellent, and elucidated in a mathematical way the intuition that I and a lot of other investors have had about COVID-19’s impact, but there were nuances that I wanted to build on here.
I’ll summarize the Brooklyn Investor’s post briefly here for context (although I do recommend reading it in its entirety). Let’s start with the basic principle that any financial asset is worth the present value of all its future cash flows. In the case of common stock, this means the value of all future cash flows to common equity (this means earnings after interest payments, preferred dividends, taxes, etc.) generated by the company, discounted back to the present day at a discount rate. The value of the company is therefore affected by two factors: its future earnings trajectory, and the discount rate that you use. I’m not explaining anything new here; this is all corporate finance 101.
We can also value groups of companies in aggregate in this way. Say, hypothetically, that you have a group of companies that comprise a large swath of the American economy. These companies collectively earn exactly $10/year for the next 10 years (purposely not using real S&P 500 values to avoid distracting from the point here). At a discount rate of 4%, this group of companies is worth $243.24 today (a terminal multiple of 25x is used in year 10, since a stream of earnings that does not grow is worth 1/0.04 = 25x with a 4% discount rate). This is represented by the leftmost of the three scenarios below.
Let’s say, then, that COVID-19 strikes and this group of companies’ earnings are collectively wiped out for two whole years. That is to say, American corporate profits are essentially zero for two years (bear in mind, this did not happen in 2008 or in the Great Depression and would be a far worse scenario than analysts are expecting). What’s our index of companies worth then? $224.38. That’s a whopping 7.8% decrease in value (middle scenario above). On the other hand, the S&P 500 index was down more than 30% at its trough in late March due to COVID.
We can do the same with 4 years of earnings wiped out (rightmost scenario above). This gets us a 15% decrease in PV of future cash flows - far less than 30%.
All these scenarios look pretty rosy, and that is the point of the original post. COVID-19, due to its temporary nature, should theoretically not merit an outsize impact on equity valuations.
But let’s look deeper. There’s another factor we’re not considering, which is the discount rate. The discount rate is the rate of return investors require to hold equities. The discount rate for an equity index is the risk-free rate (we can use 10-year Treasuries) plus a market equity risk premium. Again, this is corporate finance 101. We used such a 4% discount rate in all previous scenarios.
But, right off the bat, that discount rate seems low. It seems low for a few reasons:
We would hope to earn more than 4% on our equities. S&P returns since 1871 have been 9.15% CAGR.
The equity risk premium has averaged somewhere in the range of 4-5% in the U.S. since 1871. This is calculated as follows: 9.15% nominal S&P CAGR - 4.6% nominal 10-year Treasury yield = 4.55%. If the equity risk premium is 4-5%, the discount rate must be greater than or equal since Treasury yields are not currently negative.
Interest rates are at historic lows. As I’m writing this, Treasuries yield 0.73%. At these levels, a reasonable discount rate could be 0.73% + 4.55% = 5.28%. However, if interest rates go any higher, the discount rate will go up.
All of this means 4% is probably a low-ball estimate for the discount rate going forward. But does this matter? The answer is yes, because low discount rates make future earnings more valuable, which lessens the impact of near-term crises on valuations. To demonstrate this, I conducted a sensitivity analysis across discount rates for the above scenarios.
Each column represents a different number of years with no earnings. The base case (left-most column) is no impact from COVID-19, with more of a near-term impact as you go progressively right. Different rows show different discount rates, ranging from 2% to 10%.
For the 2 years of no earnings scenario, we can see that the valuation impact of 7.8% in the table matches up with our 7.8% impact with a 4% discount rate from above. However, when we take the discount rate up to 8% (this could happen with Treasuries at a very reasonable 3.5%), the impact increases to 15%. For 4 years of no earnings, this increases to 28%. However, it is important to note that only in the most dire scenario (4 years of no earnings @ 10% discount rate) does the impact equal the magnitude of the market decline at 33%.
Regardless of discount rate, this shows that the broader point still stands; assuming the economy makes a full recovery from COVID-19 in the next 1-4 years, the market reaction was likely overblown at its trough from a purely theoretical perspective (many individual companies will go bankrupt, but others will fill their places and that’s why we are looking at the aggregate here). However, I wanted to point out how the choice of discount rate makes a significant difference in the magnitude of the market impact. Thus, any analysis of the impact of COVID-19 on valuations must make reasonable discount rate assumptions. To me, it seems such assumptions should be higher than 4%. Exactly how high, I don’t know.
In general, this is a microcosm of the larger point that a lot of investors fail to see how dependent equity values are on discount rates/interest rates over the short-term (by this, I mean 10 years or so, which is short in the context of equity investing).