Fact: legacy data providers like S&P Global (SPGI) and Refinitiv miss material unusual gains/losses in footnotes that our superior fundamental data and earnings models catch.
Proof: Core Earnings: New Data & Evidence, forthcoming in The Journal of Financial Economics.
Core Earnings (as featured on CNBC Squawk Box) measure the normalized operating profitability of a business. Accordingly, when calculating core earnings, analysts should strip out any gains/losses that are non-core, non-operating, or unusual.
The Problem
Identifying unusual items that distort reported and consensus earnings has become increasingly difficult. Companies hide this information in the footnotes in ever more complicated ways. It’s gotten so difficult to unravel, that most analysts and data providers don’t do it. As a result, markets inefficiently measure earnings – as proven in The Journal of Financial Economics (JFE).
The Answer
Our Robo-Analyst technology enables us to provide the best database of unusual items in the world and, as a result, provide the best measure of Earnings Distortion – as proven by the paper Core Earnings: New Data and Evidence, forthcoming in The Journal of Financial Economics.
The paper empirically demonstrates that our adjusted Core Earnings are meaningfully superior to the best offerings from Refinitiv’s IBES and S&P Global (SPGI), and “increasingly so over time” to quote the paper.
Figure 1: How to Calculate Core Earnings
Total Revenue
+ Total Operating Income
– Total Reported Operating Expenses, Net
+ Total Hidden Non-Operating Expense, Net
– Reported Interest Expense/(Income), Net
– Reported Losses/(Income) from Unconsolidated Subsidiaries, Net
– Unreported Employee Stock Option (ESO) Expense
– Cash Taxes on Core Earnings
– Reported Minority Interest Expense, Net
– Reported Preferred Dividends, Net
– Reported Dividends on Redeemable Preferred Stock, Net
= Core Earnings
Sources: New Constructs, LLC and company filings
For more details on our calculation of Core Earnings, including the hidden and reported items we collect, we provide case studies here. For details on the difference between the “Reported” and “Hidden” items, click here. Note that data feeds with Earnings Distortion details are typically sold only to Insitutional members.
Earnings Distortion = Net Income Minus Core Earnings
Earnings Distortion is the non-core, non-operating, and unusual gains/losses that must be stripped out when calculating Core Earnings. It equals reported earnings minus Core Earnings. Figure 2 shows the key components of Core Earnings distortion. The higher the Earnings Distortion, the more overstated reported earnings are. The lower the Earnings Distortion, the more understated reported earnings are.
Figure 2: Components of Earnings Distortion
+ Earnings Distortion from 13 Categories of Hidden Items, Net
+ Earnings Distortion from 13 Categories of Reported Items Pre-Tax, Net
+ Income Tax Distortion
+ Earnings Distortion from 2 Categories of Reported Items After-Tax, Net
= Total Core Earnings Distortion, Net
Sources: New Constructs, LLC and company filings
Figure 3 shows the “Hidden” items we display on the Core Earnings pages in our company valuation models on over 5,000 stocks.
Figure 3: “Hidden” Items Details from our Models
Sources: New Constructs, LLC and company filings
Figure 4 shows the “Reported” items we display on the Core Earnings pages in our company valuation models on over 5,000 stocks.
Figure 4: “Reported” Items Details from our Models
Sources: New Constructs, LLC and company filings
Clients who subscribe to our valuation models get access to all the data behind our calculations. Our models also provide audit-ability of all data back to the original 10-Ks and 10-Qs upon which our models are built.
Analyzing Core Earnings & Earnings Distortion for the Overall Market
We leverage our database of footnotes adjustments and calculation of Core Earnings to spot trends across the market. Core Earnings, over the trailing twelve months (TTM) period through 1Q21, fell below S&P Global’s (SPGI) Operating Earnings for the first time since April 2020. Investors only looking at reported earnings would get a misleading picture of profitability across the market. Per Figure 5, SPGI Operating Earnings have been more volatile over the last ~18 months than our Core Earnings. More details are in our report on this topic.
Figure 5: SPGI Operating Earnings vs. Core Earnings: Dec 2019-1Q21
Sources: New Constructs, LLC and company filings, and S&P Global (SPGI). Note: the most recent period’s data for SPGI’s Operating Earnings are based on consensus.
Our Core Earnings analysis is based on aggregated TTM data through 6/30/13, and aggregated quarterly data thereafter for the S&P 500 constituents in each measurement period.
Figure 6 shows the level of Earnings Distortion from the unusual gains and losses since 2000. Note the rapid rise in the distortion from gains buried in footnotes over the last few years.
Figure 6: Core Earnings Distortion: Worst Since 2000 – Data Through November 18, 2019
Sources: New Constructs, LLC and company filings.
Most Under/Overstated GAAP Net Income Vs. Core Earnings
Figure 7 shows the companies with the most under/overstated TTM Street Earnings per share (another flawed legacy earnings measure) as compared to Core Earnings per share as of 1Q21.
Figure 7: Companies with Most Understated/Overstated TTM Earnings as of 1Q21
Sources: New Constructs, LLC and company filings.
Per Figure 7, Broadcom Ltd’s (AVGO) Core Earnings are $15.90 per share less than Street Earnings per share over the TTM period. Below, we explicitly reconcile Core Earnings to GAAP Earnings, but we cannot do the same for Street Earnings because analysts do not publicly disclose what is captured in Street Earnings. However, we know Street Earnings often miss material unusual items included in GAAP Earnings.
The large disconnect between GAAP earnings and Core Earnings stems from over $250 million in gains from reported other non-operating income. These non-operating items include $116 million from lapse of indemnification, $56 million in other income, $53 million in interest income, and $31 million in gains on investments in 10-K 2020.
Ameriprise Financial (AMP), Alphabet, Inc. (GOOGL), Biogen, Inc. (BIIB), and Prudential Financial (PRU) also have significantly overstated Street Earnings after adjusting for unusual items.
On the other hand, NVR Inc.’s (NVR) Street Earnings per share (and GAAP Earnings for that matter) greatly understate its Core Earnings. NVR Inc. (NVR) took a $25.6 million impairment of deposits in its 2020 10-K. When combined with other non-core expenses, such as $3 million in interest hidden in cost of sales in the 2020 10-K, we see that NVR’s Core Earnings of $261.74/share are much higher than the $247.75/share Ii Street Earnings.
Allstate Corp (ALL), Chubb Limited (CB), Humana Inc. (HUM), and Wynn Resorts, Ltd’s (WYNN) Street Earnings understate the Core Earnings of their business operations in the TTM period. Although understated earnings alone don’t mean a stock is worthy of investment, four of the five stocks with the most understated earnings in Figure 7 receive an Attractive-or-better rating.
Get Earnings Distortion Scores for All ~3000 Stocks Under Coverage
We leverage our Earnings Distortion measure to provide Earnings Distortion Scores for all stocks under coverage. These scores indicate how likely companies are to beat or miss expectations.
All members get our Earnings Distortion Scores for the stocks in their Portfolios by expanding the rating details using the “+” button in the first column of the Portfolio page, per Figure 8.
Figure 8: Earnings Distortion Scores Available By Expanding Rating Details
Sources: New Constructs, LLC and company filings
Platinum and higher members also get our Earnings Distortion Scores on the Ratings page. Per Figure 9, the Earnings Distortion Score will display above Analyst Notes on the Ratings page.
Figure 9: Earnings Distortion Scores on the Ratings Page
Sources: New Constructs, LLC and company filings
Institutional members get access to our Earnings Distortion tab on the Ratings page (and available via the Portfolios page). Per Figure 10, the Earnings Distortion tab displays more details about Earnings Distortion, such as historical Earnings Distortion, Earnings Distortion as a percent of different metrics, and more.
Figure 10: Earnings Distortion Tab for Institutional Members
Sources: New Constructs, LLC and company filings
This article originally published on January 13, 2020.
Disclosure: David Trainer, Kyle Guske II, Alex Sword, and Matt Shuler receive no compensation to write about any specific stock, style, or theme.
Follow us on Twitter, Facebook, LinkedIn, and StockTwits for real-time alerts on all our research.