Earnings Models: Has the Rally Outpaced Reality? By: Gabriel PotterMBA, AIFA® 2020.05.18

The economy and the markets seem disconnected

We have repeatedly asserted that, in the short-term, there is limited correlation between the current economy and the stock market, because economic data is generated weeks (or even months) after real world events occurred and the stock market looks forward to what’s likely to happen next.  The market action over the past few weeks have demonstrated this axiom quite clearly.  The news cycle tells us that the economy is clearly suppressed.  Many service industries have almost completely shut down and unemployment is reaching Great Depression levels.  Meanwhile the stock market, even at its March 23 nadir, only fell 33% from its recent all-time high in February.  That sort of stock market pullback is typical to a standard recession, not the cataclysm we’ve experienced.  Moreover, the market has meaningfully recovered over the past six weeks, up approximately 30%, which reduces total losses of this pullback to about half of what it was.  The dissonance is so high, it is as if these events are happening on different planets.

We have also asserted that, in the long-term, there is a meaningful correlation between the real world economy and the stock markets because investors are ultimately owed the benefits which companies generate (directly through dividends or indirectly through capital appreciation) as they act inside of an economic system.  This raises the question:  do current long-term economic forecasts support the recent stock market rally?  What does the rally suggest about investors expectations of the future?  We can begin to understand the answer to that question by creating a basic forecast of company earnings, seeing how much that potential has been damaged by the economic slowdown, and then determining if the market pullback is commensurate with the presumed losses of corporate earnings.

Creating a model:  earnings and value over time

There are classic measures of stock valuation which analysts use to evaluate stocks and the stock market in general.  The most common is the P/E ratio – price over earnings.  In the long run, the US stock market tends to have a P/E ratio of about 19X.  In other words, the price of a stock tends to be about 19 times bigger than the current earnings of the company.  If the P/E ratio is lower, then it means investors are paying less today to get ownership of a company’s earnings.  If the P/E ratio is higher than 19, it means investors have to pay more to purchase a stock and receive the benefits (dividend payments and capital appreciation) for owning it.  The P/E ratio is a fine measurement, but its usefulness is limited during times of stress because both price and earnings will radically change, but at different points in time.  Changes in price happen instantly; changes in reported earnings happens over quarterly reporting cycles.  So, the P/E ratio gets out of alignment quickly.

To account for these periods of stress, when prices and earnings are both in flux, it might be more useful to valuate stocks (or the stock market) over a longer period of time to coax the long-term ratios back into alignment.  For instance, a widely used method for determining a fair price of a company (or any financial venture) is to make estimates about the totality of its cash payments over time, next discount these cash flows by interest rates, and finally sum these discounted cash flows to  a single number.  This is known as net present value analysis, and we can use its principal approach.

A full net present value analysis of the underlying US stock market is, perhaps, too ambitious at the moment since so many earnings estimates are still unknown.  However, if readers are interested, our annual market prediction roundup, which we last published as our January 2020 monthly article, presented S&P fair value predictions from more than a dozen Wall Street firms.  Instead, we are trying to determine if the market recovery is commensurate with the immediate loss of earnings brought upon the government suppression of economic activity due to coronavirus containment efforts.  To make this determination, we will revisit the assertion that the long-term value of a stock is, at heart, a function of its earnings over time.  This earnings-based examination creates a stock valuation estimate based on the top line of an income statement:  how much is the company making and how much is that number growing?  This simplified analysis avoids worrying about costs, profitability, taxes, interest, debt, competitors, and everything else that a real-world corporate officer would consider. 

We could compare the total expected earnings for the US stock market today to the collective expectations of the market during a sunnier time, say the end of 2019.  Thus, we will create two net-present value models of US stock market earnings. First, we will model of total future earnings based upon what we knew as of December 31, 2019 and compare it to the stock price at the time.  Second, we’ll run the same model as of today to see if the investors are getting the same implied valuations for the general stock market. 

Model 1:  earnings and the stock market at year end 2019

For our base case model, we should have an expectation for earnings.  The S&P 500’s operating earnings at the end of 2019 was $157.12 per share, but that number isn’t constant over time.  In general, earnings numbers grow with the economy.  The last few years of earnings saw high increases (up 17.2% in 2017 and up 21.8% in 2018), driven by the changes in the tax code but that effect finally slowed in 2019, leading to a modest 3.6% increase year over year.  We will use 3.6% in our net present value calculation as our expected earnings growth target.  So, in the first year, we’ll assume 3.6% additional earnings over $157.12, which is $162.78.  Assuming a constant growth rate, the following year sees earnings hit $168.64, and so on. 

Another key input in our net present value calculation regards the discount rate.  After all, getting a dollar today is better than getting a dollar a year from now.  We have to adjust the value of future earnings based upon how long we have to wait to get them.  On the plus side, given our low interest rate environment, the interest rates we must use to discount future cash flows are quite modest from a historical perspective.  In fact, given the scarcity of inflation, we are going to presume a 2.0% interest rate for our model – matching the implicit target set across by the Federal Reserve.  We will compound the effect each year, so future cash flows become less and less as interest rates soften their present value. So, for instance, this first year’s expected earnings of $162.78 has a discounted present value of $159.58.  The following year earnings of $168.64 has a discounted present value of $162.09, and so on.

The baseline considers market conditions as of December 31, 2019.  Starting our model at Year 0 and modeling out to Year 30, the net present value of the future earnings generated by the S&P 500 works out to $6211 generated per share.  At the end of 2019, the price of a share of the S&P 500 was $3,257, thus creating a Price / NPV of Earnings ratio of 0.52.

Model 2:  earnings and the stock market at May 2020

After the coronavirus hit, the suppression of earnings through key service industries could be felt.  Nobody knows how long the government mandates and consumer driven reluctance will depress earnings, but – for simplicity’s sake – let us presume that the affect will be focused on the first year of the model.  After a year, it is likely vaccines or other treatment options may mitigate the effect, however changes to business operations will reset the starting point of the economy to a lower baseline. According to the most recent data from FactSet, the most recent combined earnings decline was 13.7%.  The annual 2020 earnings decline will assuredly change from that initial estimate.  The damage is likely to bottom out in Q2 and Q3, and we may experience partial recovery by Q4 2020.  As a reasonable starting point for this basic model, we can start to discount our earnings estimate for 2020 by the 13.7%. 

Everything else in the model will remain the same.  After year 1, however, the long-term growth rate will allow earnings to appreciate annually.  The long-term growth rate of 3.6% and interest rate estimate of 2.0% are constant across both models.  Starting our model at Year 0 and modeling out to Year 30, the net present value of the future earnings generated by the S&P 500 works out to $5200 per share.  At market close, May-14-2020, the price of a share of the S&P 500 was $2832, thus creating a Price / NPV of Earnings ratio of 0.55.


We know this model is a crude facsimile, but even the most detailed analyses are merely estimating real world outcomes.  Professional models made with infinite complexity and thousands of inputs still cannot account for the unpredictability of real-world events and the irrationality of collective investors.

With that caveat out of the way, here’s what our model suggests: current investors in the stock market are actually presuming a more optimistic outcome based on the limited data than they were at year end.  This is significant given the untested complacency of the investment environment after the 11-year bull run. 

Based on the simplified forecast, investors are potentially paying more today for future earnings than they were at the year’s end, even after an incredible market advance throughout 2019.  We always knew snapshot comparisons of valuation (i.e. – Price / Earnings ratios) was going to look make today’s market look unfavorable because current earnings data is weak.  Our analysis suggests that the stretched overvaluations persist even once we combine potential earnings out over decades, thus removing the myopic limits of today’s strained economic conditions. 

Today’s stock rally seems very optimistic given the near-term collapse of earnings; this suggests investors believe the pullback in earnings is temporary and that earnings will rebound quickly to 2019 levels.  More specifically, our model suggests that investors expect the earnings level to reach the previous 2019 peak faster than the four years which our second model anticipates.  In other words, today’s investors may be expecting something like a “V” or “U” type recovery of earnings, wherein economic activity rapidly returns to the levels seen before the crash.  Only time will tell how justified that optimism is.


The information contained herein has been obtained from sources that we believe to be reliable, but its accuracy and completeness are not guaranteed.  Westminster Consulting, LLC reserves the right at any time and without notice to change, amend, or cease publishing the information.  It has been prepared solely for informative purposes.  It is made available on an "as is" basis.  Westminster Consulting, LLC does not make any warranty or representation regarding the information.  Without prior written permission from Westminster Consulting, LLC, it may not be reproduced, in whole or in part, in any form. The information in this document is confidential and proprietary to Westminster Consulting, LLC including its business units and may be legally privileged. Any unauthorized review, printing, copying, use or distribution of this document by anyone else is prohibited and may be a criminal offense. Indices mentioned are unmanaged and cannot be invested into directly.  Past Performance does not guarantee future results.





Gabriel Potter

Gabriel is a Senior Investment Research Associate at Westminster Consulting, where he is responsible for designing strategic asset allocations and conducts proprietary market research.

An avid writer, Gabriel manages the firm’s blog and has been published in the Journal of Compensation and Benefits,...

More about Gabriel Potter
Sign up for our Newsletter
Sign up for our Newsletter