Yardeni on S&P 500 Earnings, Valuation, and the Pandemic: Part I

This three-part series is excerpted from my 2020 book S&P 500 Earnings, Valuation, and the Pandemic (co-authored with Joseph Abbott). Part I (Earnings): Introduction. Discounting Forward Earnings. Lots of S&P 500 Earnings Measures. Revenues, Earnings & Profit Margins. Part II (Valuation): Flying with the Blue Angels. In the Eyes of the Beholder. Reversion to the Mean. Fundamentals Matter. Discipline of Dividends. Part III (Pandemic): Very Useful Indicators. The GFC and the GVC. Fed-Led Valuation Meltup. Epilogue. 

Part I: Earnings

Introduction

I started my career on Wall Street in 1978. I spent the prior year at the Federal Reserve Bank of New York in the economics research department after receiving my undergraduate degree in economics and government from Cornell University in 1972 and my PhD in economics from Yale University in 1976. Over the past 40-plus years, I’ve worked as both the chief economist and the chief investment strategist at several firms on Wall Street. Since January 2007, I’ve been the president of my own consulting firm, Yardeni Research, Inc.

My job continues to be to predict the financial markets, particularly the major stock, bond, commodity, and foreign exchange markets around the world. I’ve learned a lot about these markets over the years. I started sharing what I’ve learned recently in a series of books and studies.

In this study, I will focus on the S&P 500 stock price index, examining how it is determined by the earnings of the 500 companies that are included in the index and the valuation of those earnings by the stock market.

Why pick the S&P 500?

The S&P 500 is a stock market index that measures the stock price performance of 500 large companies listed on stock exchanges in the United States. It is one of the most widely followed equity indexes. The stocks in this index are a representative sample of leading companies in leading industries. Many equity managers benchmark the performance of their portfolios to the S&P 500. Among the largest exchange-traded funds are those that track the S&P 500. The S&P 500 represents more than 83% of the total domestic US equity market capitalization.[1]

The widely followed Dow Jones Industrials Average (DJIA) has only 30 companies. It was launched in 1896 and was a spin-off of the Dow Jones Transportation Average, which was first compiled in 1884 by Charles Dow, the co-founder of Dow Jones & Company. The S&P 500 dates back to 1923. That year, the Standard Statistics Company (founded in 1906 as the “Standard Statistics Bureau”) developed its first stock market index, consisting of the stocks of 233 US companies and 26 industries, computed weekly. (The company also began rating mortgage bonds in 1923.) In 1926, it developed a 90-stock index, computed daily. In 1941, Poor's Publishing merged with Standard Statistics Company to form Standard & Poor's. On March 4, 1957, the index was expanded to its current 500 companies and was renamed the “S&P 500 Stock Composite Index.”

The components of the S&P 500 index and other S&P indexes are selected by the firm’s US Index Committee, which meets monthly. All committee members are full-time professional members of the firm’s Indices staff. At each meeting, committee members review pending corporate actions that may affect the indexes’ constituent companies, statistics comparing the indexes’ composition to the broad stock market, candidate companies under consideration for addition to an index, and the bearing of any significant market events on the indexes.

The committee identifies important industries within the US equity market, approximates the relative weight of these industries in terms of market capitalization, and then allocates a representative sample of stocks within each industry of the S&P 500. There are 11 sectors according to the Global Industry Classification Standard (GICS): Communication Services, Consumer Discretionary, Consumer Staples, Energy, Financials, Health Care, Industrials, Information Technology, Materials, Real Estate, and Utilities.[2] These sectors are further divided into 24 industry groups, 69 industries, and 158 sub-industries.

Candidates for inclusion in the S&P 500 index must meet specific criteria in eight areas: market capitalization, liquidity, domicile, public float, GICS, financial viability, length of time publicly traded, and stock exchange listing. The index is reconstituted quarterly, though changes are made infrequently.

The S&P 500 index is a free-float, capitalization-weighted index. That means that companies are weighted in the index in proportion to their market capitalizations. To determine the market capitalization weight of a company, only the number of shares available for public trading (“public float”) is used. Shares held by insiders or controlling shareholders that are not publicly traded are excluded from the calculation. The largest companies (based on market capitalization) in the S&P 500 account for a substantial portion of its total market capitalization. Since the index is market capitalization-weighted, these companies have the greatest influence on the index’s price performance.

Notwithstanding occasional bear markets, the S&P 500 has been a great investment over the years—so much so that “S&P” could stand for “Success & Profit.” Since January 1, 1955 through September 2, 2020, the index has been down in bear markets during 3,029 of the 16,535 trading days, i.e., just 18.3% of the time. It has risen at a compounded annual rate of 6%, a rate that doubles the value of this portfolio every 12 years. And that doesn’t include the dividend return provided by many of the S&P 500 companies.

Part I in our study covers the various measures of earnings for the S&P 500 and why we favor forward earnings among them. Part II discusses various models of valuation, again focusing on the S&P 500. Part III uses the resulting analytical framework to review how it has worked in good times and bad, focusing on the Great Financial Crisis (GFC) and the Great Virus Crisis (GVC).

Discounting Forward Earnings

This primer for investors develops a simple framework for analyzing and forecasting the widely followed S&P 500 stock price index (P). Doing so should be simple enough. One only needs to forecast two numbers, i.e., earnings per share (E) and the price-to-earnings valuation ratio (P/E) in the stock market equation:

P = P/E · E

Forecasting these two variables is easy; getting them right is the hard part. Most investment strategists use their own deductive “top-down” earnings forecasts for the current year and coming year and multiply them by their forecasts of the P/E for the current year and the next year. I modified this approach during 2001 with my “Earnings Squiggles” analytical framework. I start with the “bottom-up” data on earnings expectations of industry analysts, benchmarking my outlook to theirs.

Why incorporate analysts’ expectations into my thinking about the prospects for earnings? The stock market discounts future expected earnings. Past and current earnings are relevant, but only to the extent that they influence the outlook for future earnings.

Whose earnings expectations does the market discount, and how far into the future?

The market doesn’t discount the earnings expectations of individual investment strategists or even the consensus expectations of top-down strategists. It discounts the bottom-up consensus earnings expectations of industry analysts. It’s those expectations that I want to quantify and use in the stock market equation as a benchmark for my own forecasts.

Therefore, the bottom-up earnings expectations data that I use are an aggregation of the estimates for all the stocks in the S&P 500 covered by industry analysts. As the saying goes, the stock market is a market of stocks. Therefore, using bottom-up earnings estimates makes more sense than using top-down forecasts. Since the stock market is forward-looking with stock prices discounting future earnings prospects, I don’t use so-called “trailing earnings” in the stock market equation.

More specifically, I believe that stock investors are basing their decisions on the outlook for earnings over the year ahead, i.e., the next 12 months or 52 weeks. Experienced investors recognize that anything beyond that is too far off to forecast with any degree of accuracy. Investors obviously rely on industry analysts for their insights about earnings. Consequently, I view analysts’ consensus forecasts as a treasure trove of valuable information on earnings for the stock market equation. However, analysts don’t provide rolling earnings forecasts for the coming 12 months. Like company managements, they focus on quarterly estimates for the current year and the coming one.

Fortunately, I/B/E/S data by Refinitiv fills this void. I/B/E/S--which stands for "Institutional Brokers' Estimate System"--compiles analysts’ consensus earnings-per-share expectations for each of the S&P 500 corporations and combines them to calculate the consensus expected earnings per share of the overall S&P 500 for each of the quarters of the current year and the coming year. I/B/E/S provides a useful proxy called “12-month forward consensus expected earnings” for the S&P 500. It is a time-weighted average of the analysts’ consensus earnings estimates for the current year and the coming year, which starts during September 1978. This series supplies the forward-looking earnings (E) and the forward-looking valuation (P/E) I need to assess the stock market.

In 2000, I hired Joe Abbott from I/B/E/S to help me develop an in-house database and analytical tools to monitor and analyze the consensus data so that we could track forward earnings for the S&P 500, its 11 sectors, and the more than 100 industries that compose them. Joe had been a senior equity strategist at I/B/E/S for 14 years, before it was acquired by Thomson Financial during 2000, so he was exceptionally well qualified for the job. Together, we developed a simple graphical framework for visualizing the I/B/E/S consensus data.

We dubbed the framework “Earnings Squiggles” because the time series for each calendar-year forecast, which we update monthly, tend to look like squiggles (Fig. 1 and Fig. 2). Forward earnings is simply a time-weighted average of the analysts’ latest consensus estimate of earnings for the current year squiggle and for the coming year’s squiggle. At the start of a year (i.e., in January), forward earnings is identical to the current year’s consensus forecast. One month later (i.e., in February), forward earnings is the weighted average of 11/12 of the current year’s estimate and 1/12 of the coming year’s estimate. So as any given year progresses, forward earnings gradually converges with the estimate for the coming year, and by January it is once again identical to next year’s consensus outlook. Of course, the next year’s earnings estimate is a moving target because it changes as analysts revise their earnings estimates, as does the current year’s estimate. (See Appendix 1, Deriving 12-Month Forward Earnings.)

In the monthly charts, we show every year’s squiggle spanning 25 months from February to February, i.e., 11 months before a given year begins and two months after it ends. That’s because for calculating forward earnings, the next year only enters the calculation once the current year is one month old, as just noted above. Then the squiggles run through the 12 months of the actual year and another two months after it ends. That’s because the fourth quarter’s results for each year are reported in the earnings season during January, with some stragglers during February. Each annual squiggle starts one year early since it needs to be time-weighted with the current year. Each annual squiggle ends after earnings are released for the final quarter of the year it covers, though it only matters through the end of its year.

This may be a bit confusing, so a more specific example should help to make sense of it all. During January 2018, forward earnings was identical to the consensus estimate for all of 2018. No weight was given to 2019’s estimate. During February 2018, we started to track the 2019 squiggle because forward earnings represented 11/12 of the latest 2018 estimate and 1/12 of the latest 2019 consensus estimate. By January 2019, the 2018 squiggle was no longer relevant, but we plotted the squiggle through February 2019 (when data for the final quarter of 2018 were available) to show that it had converged to closely match the actual result for 2018.

One of the biggest advantages of the I/B/E/S forward earnings consensus is that the data are available much more frequently than the measures of actual profits that are provided quarterly, with a lag of a few weeks for the S&P 500 by Standard & Poor’s and for the corporate sector broadly in the National Income and Product Accounts (NIPA), compiled by the Bureau of Economic Analysis (BEA). The Earnings Squiggles and forward earnings for the S&P 500 are available not only monthly from September 1978 onwards but also weekly from March 1994 on (Fig. 3 and Fig. 4).

We have 40 years of complete annual squiggles from 1980 through 2019, with 25 months of data for each one of them. From the beginning to the end of each annual squiggle, estimates fell for 31 of those years and rose for nine of them. The squiggles tend to decline over time because analysts tend to be overly optimistic about the outlook for their companies’ earnings the further in the future they are. The up-year exceptions were 1980, 1988, 1995, 2004, 2005, 2006, 2010, 2011, and 2018 (Fig. 5). Of course, the steepest downward slopes have occurred during recessions, when analysts are scrambling to cut their estimates. The few years when they raised their estimates in the past tended to be during economic recoveries, especially following bad recessions during which analysts had become too pessimistic. The overall average decline for the 25 months of the 40 years was –11.9%, with the 31 down years averaging –17.5% and the nine up years averaging 7.0%.

Joe and I track Net Earnings Revisions Indexes (NERIs) for the S&P 500 as well as its 11 sectors and 100-plus industries. NERIs show the percentage of analysts’ forward earnings estimates that have been revised higher minus the percentage of them that have been revised lower, divided by the total number of forward earnings estimates. The resulting indexes are extremely volatile on a weekly and monthly basis and tend to be most active around earnings reporting seasons, when analysts are more likely to adjust their forecasts. We’ve found that the three-month average of NERIs provides the most useful information, since it encompasses the entire quarterly earnings cycle. Our data start during January 1985 on a monthly basis and mid-January 2006 on a weekly basis.

In the past, S&P 500’s NERI always turned negative during recessions and tended to be positive during recoveries (Fig. 6). During expansions, it has shown mixed performances. Given that most Earnings Squiggles have downward slopes, NERIs tend to have a negative bias. So during expansions, we are not overly concerned to see negative NERIs and give more weight to the positive ones.

It’s human nature for industry analysts to be biased toward optimism about the prospects of the companies they follow. Most have a strong professional interest in their designated industries and companies. Most prefer to give buy ratings rather than sell ratings, which is why they have more of the former than the latter. Analysts don’t want to follow companies that are likely to go out of business. If they are following a dying industry, they can score points by being bearish ahead of the curve. However, eventually they’ll have to start all over again covering a different industry. Also, analysts are loath to get too negative on companies that have investment banking relationships with their firm. Given analysts’ inherent optimism, savvy investors know that when the analysts downgrade their recommendation on a stock from “buy” to “hold,” they probably mean “sell.”

Now, imagine the following stylized conversation between a “sell-side” industry analyst and a “buy-side” portfolio manager:

Jan: Jim, thanks for visiting us today and sharing your earnings outlook for your industry, especially for ABC Corp., which we own. We always value your insights and analysis. However, you’re always too optimistic on earnings and invariably lower them. When might you be cutting your estimates yet again for this year?

Jim: Jan, thanks for taking the meeting. Look, this year is half over. Let’s not dwell on it too much. Let’s talk about next year. It’s going to be a great one for the company.

That in a nutshell is how earnings are discounted by the stock market, in our opinion, and why we are fans of forward earnings as the “right” earnings measure to use in the stock market equation to forecast the market. The average portfolio manager is relying on the earnings expectations of the average analyst over the next 12 months. The P/E earnings valuation multiple in the stock market equation, however, is determined by investors, not analysts. It’s up to investors to decide how much they are willing to pay for the time-weighted average of analysts’ consensus earnings expectations for this year and next year.

We need to stress a very important point about the Earnings Squiggles and forward earnings: the stock market can go up when analysts are reducing their earnings estimates for the current year and the coming year. We are often asked how this can be. The answer is that if next year’s estimate exceeds the time-weighted average, forward earnings will rise as next year gets more weight while this year gets less. Arithmetically, forward earnings converges to next year’s consensus estimate. While it is possible for forward earnings to rise if the current-year consensus plunges, it’s hard to imagine that next year’s estimate won’t take a dive too. Let’s continue the above conversation:

Jan: Okay, Jim, we can talk about next year now that it’s fast approaching. However, I see in your spreadsheet that you’ve already started lowering your estimate for next year!

Jim: That’s true, Jan, but my forecast for this year is still better than last year’s result, and next year still exceeds my number for this year. Things are continuing to get better for the company, though not quite as great as I had been predicting.

Institutional investors (on the buy side) clearly value the opinions of industry analysts (on the sell side). Why else would Wall Street hire them and pay them so well? Few investors have the time or the in-house resources to do their own industry-specific research and channel checks. Those who do have their own analysts sometimes hire them from Wall Street. In-house analysts are deluged by research provided by Street analysts and often develop a close professional relationship with the ones they respect the most. Indeed, many buy-side money management firms have a voting system whereby their internal analysts collectively allocate commission dollars to the sell-side firms whose analysts have helped them the most.

To assess its accuracy, forward earnings can be pushed ahead by a year and compared to actual quarterly S&P 500 operating earnings on a four-quarter-trailing basis, i.e., the moving sum over the past four quarters (Fig. 7 and Fig. 8). The former turns out to be a very good leading indicator of the latter, with one rather important exception.

Collectively, industry analysts generally don’t do a very good job of anticipating recessions, which causes them to slash their earnings estimates for both the current and the coming years. Conversely, during economic expansions they do a very good job of forecasting earnings over the year ahead using the forward earnings proxy. Fortunately, expansions tend to last much longer than recessions. Since 1945, there have been 12 recessions that lasted 130 months in total, just 15% of the time since then through the end of 2019.

Analysts aren’t economists. It isn’t their job to see a recession coming. Besides, investors would probably ignore such warnings coming from an analyst unless he or she had insights from a company that was especially well positioned to see a recession coming. That may happen occasionally, but there is certainly no evidence that analysts collectively provide any early warnings of a coming recession.

Predicting recessions is what economists are supposed to do, and we don’t do a very good job of it. Indeed, it seems that every recession produces a superstar economist who was alone in anticipating the latest downturn. Of course, at the end of the day, it is up to investors to anticipate recessions. Some rely on their favorite economists to assess this risk. Most simply react to the news headlines. If the economic news is bad, many will sell stocks and raise their cash position even if industry analysts remain upbeat on earnings. If the news turns good, then stocks will rebound, and the analysts’ forecasts will have more credibility.

Joe and I also track analysts’ consensus earnings expectations for each of the quarters of the current year and coming year on a weekly basis. That provides us with a window into a strange tendency of the “too-high” analysts to lower their estimates as earnings seasons approach and the “low or just right” analysts to hold their forecasts steady. Often, that sets the market up for a positive earnings surprise, which looks like an upside earnings hook, when actual reported results turn out to be better than consensus estimates.

Furthermore, company managements generally deliver bad news and warnings during the “preannouncements” that precede earnings reporting seasons. The good news is typically held back by company managements, often causing their stock price to pop when they release the better-than-expected results. They will also get a bigger positive surprise score in the consensus database services. The size of the surprise is an oft-used screening criteria for investors who rely on quantitative analysis. This game causes the aggregate forecast often to fall ahead of actual results because downward revisions are dominating the analysts’ community then, often setting the stage for the upside hook.

Analysts’ quarterly consensus earnings expectations for the S&P 500 are available on a weekly basis from late March 1994. Joe and I track them in our chart publication titled S&P 500 Earnings Squiggles Annually & Quarterly. From the first quarter of 1994 through the second quarter of 2020, there were 106 quarterly squiggles. Of these, 87 ended with earnings hooks, where the actual results were better than analysts predicted at the start of the earnings season by at least 0.1%. Squiggles reflecting estimates that were “beat” by 3.0% or more totaled 54, while only 20 squiggles reflected big positive surprises exceeding 5.0%. The longest streak of positive surprises occurred during every quarter from the first quarter of 2009 through the second quarter of 2020.[3] Such upward surprises don’t happen during recessions, when actual results often turn out to be worse than the rapidly falling estimates, so the earnings hook is much smaller or nonexistent.

Of course, most of the information in companies’ quarterly earnings reports is old news, having happened during the previous quarter. However, the information does provide insights into the likely future course of a company’s earnings. From the perspective of our forward earnings analytical approach, the fourth quarter of each year is the least important. That’s because by the time the results are reported during January and February of the next year, the previous year (including the fourth quarter, of course) is irrelevant to forward earnings, which no longer gives any weight to it. However, each quarter’s results can significantly impact earnings revisions for coming quarters. This will be an important consideration in Part III when we discuss the fact that the Lehman calamity occurred late in 2008, while the Covid-19 pandemic occurred in early 2020.

Lots of S&P 500 Earnings Measures

Above we examined the relationship between forward earnings and the actual operating earnings of the S&P 500, both calculated using I/B/E/S data. The former tends to be a good leading indicator of the latter when the economy is growing but not when it is falling into a recession. There are other measures of S&P 500 profits. They aren’t forward-looking or available weekly; only forward earnings has these advantages. The rest represent actual quarterly results compiled by other private-sector data vendors. What many of these series offer are more historical data than are available for forward earnings, which starts in September 1978. So, they can provide a longer-term perspective on the trends and cyclical performance of profits.

Here is our brief survey of these other earnings measures available for the S&P 500 on a quarterly basis:

? Reported (GAAP) earnings (S&P data since 1935). Standard & Poor’s has compiled S&P 500 quarterly earnings on a reported basis since the first quarter of 1935 (Fig. 9). The Securities and Exchange Commission (SEC) requires that publicly traded companies include financial statements in their (unaudited) 10-Q and (audited) 10-K reports, including earnings figures based on Generally Accepted Accounting Principles (GAAP).

 Interestingly, the long-term annual growth rate of reported earnings has mostly been around 6.0% and ranged between 5.0% and 7.0% since the start of the data.

? Operating earnings (Standard & Poor’s data since 1988). Since the first quarter of 1988, Standard & Poor’s has provided an operating version of the quarterly earnings of the S&P 500 companies (Fig. 10). Unlike the reported earnings series, it excludes one-time write-offs, charges, and gains. Both are available on a per-share basis as well as on a total-dollars aggregate basis. 

The operating measure almost always exceeds the reported one because one-time unusual costs and losses tend to occur more frequently than one-time windfall gains. S&P’s in-house analysts determine the “one-offs” that are excluded from each of the 500 companies’ reported earnings when compiling the operating earnings series for the S&P 500.

? Operating earnings (I/B/E/S data since 1993). To complicate matters, other widely respected data vendors calculate S&P 500 operating earnings somewhat differently than does Standard & Poor’s. The most widely used numbers are compiled by Bloomberg, FactSet, I/B/E/S, and Zacks. Joe and I prefer the longer history of the I/B/E/S series.

The data in both the I/B/E/S and Standard & Poor’s series are on a pro forma basis, so they reflect the composition of the S&P 500’s portfolio as it was in each period. As a result, changes in the value of the index over time is an exercise in apples-to-oranges comparison: The value of today’s index is being compared to the values of past versions of the index, even though its composition changes over time as companies are added, subtracted, merged, and acquired through the years.

The I/B/E/S measure tends to be the same as the comparable Standard & Poor’s series but has diverged at times. Particularly during recessions, I/B/E/S operating earnings tends to well exceed Standard & Poor’s measure because it treats more of the losses incurred during bad times as one off (Fig. 11).

For example, according to I/B/E/S, the S&P 500 Energy sector had operating earnings of $3.04 per share during the first quarter of 2020, while Standard & Poor’s calculated a loss of $9.16. Oil prices dropped sharply during the quarter. Standard & Poor’s included the revaluation (or write-down) of the oil reserves. Neither I/B/E/S nor industry analysts did so in either their estimates or actual results. A similar plunge in oil prices caused a divergence between the Standard & Poor’s and I/B/E/S operating earnings calculations for the Energy sector from the first quarter of 2015 through the second quarter of 2016.

So, the big difference between the Standard & Poor’s and I/B/E/S measures of operating earnings per share is that the former determines which one-time items to exclude, while the latter is based on “majority rule.” In other words, it is based on the industry analysts’ consensus on operating earnings, which tends to be the same as the operating numbers reported by the companies in their quarterly filings.

Obviously, company managements prefer to determine their own non-GAAP measure of operating earnings and are more favorably disposed toward industry analysts who follow their guidance. The SEC has warned some companies not to hype up their operating results by excluding bad stuff that shouldn’t be excluded. Importantly, industry analysts and investors who are after the unvarnished truth can always analyze the results based on GAAP, which must be reported in the quarterly filings and reconciled back to any non-GAAP measures presented. Not surprisingly, the net write-offs tend to be greater for the I/B/E/S measure of operating earnings than for the one compiled by S&P (Fig. 12).

Some strategists disparage the concept of operating earnings, calling it EBBS, or “earnings before bad stuff.” They insist that reported earnings is the only correct measure. We prefer following both measures, since reported earnings tend to diverge the most from operating earnings during downturns and bounce back when operations are back to normal.

Nevertheless, even during normal times, the I/B/E/S measure of operating earnings tends to exceed the S&P measure of operating earnings. One major reason is that Standard & Poor’s doesn’t concur with the relatively widespread practice, especially among technology companies and the analysts who cover them, of excluding stock option compensation as an expense when calculating operating earnings.

We believe that the stock market reflects the I/B/E/S measure of operating earnings since most industry analysts follow the guidance provided by company managements for what is considered to be one-time bad stuff. That’s why all our work on the S&P 500’s Earnings Squiggles, forward earnings, and the earnings valuation multiple are based on I/B/E/S data.

Revenues, Earnings & Profit Margins

We started this primer with the stock market equation. Now let’s examine the earnings equation, which simply states that earnings per share (E) equals revenues per share (R) multiplied by the profit margin (E/R):

E = E/R · R

The actual quarterly data for S&P 500 revenues per share are reported by Standard & Poor’s several weeks after the end of each quarter. Unlike for earnings, this is the only series for tracking actual revenues.

We use the same approach to calculate forward revenues as we use for forward earnings, i.e., taking the time-weighted average of analysts’ expectations during the current year and the coming one. With both forward earnings and forward revenues in hand, we can also derive the implied S&P 500 profit margin. Dividing analysts’ expectations for earnings by their expectations for revenues allows us to impute a series for the forward profit margin of the S&P 500. The monthly data for forward revenues start during January 2004, while the weekly series is available since mid-January 2006.

As we noted above, forward earnings per share tends to be a very good leading indicator of the four-quarter sum of actual earnings per share during economic expansions. Similarly, we’ve found that the monthly and weekly forward-revenues-per-share series are excellent coincident indicators of quarterly actual revenues per share, annualized simply by multiplying the series by 4.0 (Fig. 13). Not surprisingly, the implied weekly forward profit margin is also a very good coincident indicator of the actual quarterly profit margin (Fig. 14).

While the consensus annual estimated earnings squiggles tend to decline, as noted above, the slopes of the annual estimated revenues squiggles tend to be less predictable. We have 15 years of complete annual revenues squiggles from 2005 through 2019, with 25 months of data for each one of them (Fig. 15). We also have weekly revenues squiggles starting in January 19, 2006 (Fig. 16).

From the beginning to the end of each annual squiggle, using the monthly data, revenues estimates fell for six of the years and rose for nine (2005, 2006, 2007, 2008, 2011, 2012, 2017, 2018, and 2019) (Fig. 17). As with earnings squiggles, the steepest downward slopes in revenues squiggles can be found during recessions, when analysts can’t seem to cut estimates fast enough, while the steepest upward slopes are characteristic of economic recoveries—especially recoveries following recessions that were so bad that analysts became overly pessimistic.

The overall average revenues decline for the 25 months of the 15 years was –0.5%, with the six down years averaging –7.4% and the nine up years averaging 4.1%. For comparison purposes, earnings squiggles over the same the 2005-2019 period averaged -7.2%, with the up five years averaging a gain of 6.8% and the 10 down years a decline of 14.1%.

Again, we observe that S&P 500 forward revenues per share is an excellent coincident indicator of actual quarterly S&P 500 revenues per share. It has been a very useful economic indicator for us. That’s because both the actual quarterly data and the forward data are highly correlated with manufacturing and trade sales, as well as numerous other cyclical economic indicators including both the Index of Coincident Economic Indicators (CEI) and the Index of Leading Economic Indicators (LEI) (Fig. 18). The same can be said about S&P 500 forward earnings per share: it too is highly correlated with both the CEI and LEI (Fig. 19 and Fig. 20).

(For a handy table listing the various S&P 500 measures of earnings and revenues that we have discussed so far, along with their start dates, see Appendix 2: S&P 500 Price Index, Revenues & Earnings Data Series.)

___________________________

[1] See S&P Global, S&P 500: The Gauge of the Market Economy and S&P U.S. Indices Methodology, August 2020. See S&P Dow Jones Indices.

[2] The Global Industry Classification Standard is jointly developed and maintained by S&P Dow Jones Indices and MSCI.

[3] Despite the Covid-19 pandemic, earnings during the first and second quarters of 2020 turned out to be better than the downwardly revised consensus estimates, particularly for the second quarter.

___________________________

Institutional investors may sign-up for a complimentary trial of Dr. Ed's research service at https://lnkd.in/eYNwch8. Books and studies in his "Predicting the Markets" series can be found on his Amazon author's page.

Sign up to follow his LinkedIn Newsletter at https://lnkd.in/eyW46Dn

Don't miss each edition of Dr. Ed's Newsletter: Click Subscribe above or below to be notified.  

___________________________

Copyright ? 2020 Edward Yardeni. All rights reserved. No part of this publication may be reproduced in any form or by any electronic or mechanical means, including information storage and retrieval systems, without permission in writing from the publisher, except by reviewers, who may quote brief passages in a review. ISBN: 978-1-948025-08-9 (paperback) ISBN: 978-1-948025-09-6 (eBook)

Part I: Earnings .......... Part II: Valuation .......... Part III: Pandemic