In theory, Wall Street analysts adjust EPS for unusual gains and losses that distort reported EPS. In reality, they do not. Street EPS estimates are materially misleading more often than investors realize[1].
The same is true for other well-known “adjusted” measures of earnings as we detail in S&P’s “Operating Earnings” Remain Overstated in 2Q21.
This report shows:
- Five S&P 500 companies with understated Street estimates likely to beat 3Q21 earnings
- Why Street Earnings, GAAP Earnings, and consensus estimates are flawed
- How our Core Earnings[2] and Earnings Distortion factor generate novel alpha
Get our report on the S&P 500 companies more likely to miss 3Q21 Street EPS estimates here.
Street Understates EPS for 122 S&P 500 Companies
Per Figure 1, Street EPS understate Core EPS for 25% of S&P 500 companies for the trailing twelve months ended calendar 2Q21[3]. For 10% of S&P 500 companies, Street EPS are understated by more than 10% (examples are below).
Figure 1: Street Earnings Are Understated for a Large Amount of S&P 500 Companies
Sources: New Constructs, LLC and company filings.
When the Street understates EPS, it does so by an average of -23% per company, see Figure 2.
Figure 2: Street Earnings Understate by -23% on Average
Sources: New Constructs, LLC, and company filings
Five S&P 500 Companies Likely to Beat Calendar 3Q21 Earnings
Figure 3 shows five S&P 500 companies likely to beat calendar 3Q21 earnings based on understated Street EPS estimates. Below we detail the hidden and reported unusual items that have created Street Distortion, and understated Street Earnings, over the TTM for T-Mobile U.S. Inc (TMUS). Contact us for the same details on the other companies.
Figure 3: Five S&P 500 Companies Likely to Beat 3Q21 EPS Estimates
Sources: New Constructs, LLC, company filings, and Zacks
*Assumes Street Distortion as a percent of Core EPS equals the same percent in 3Q21 as TTM ended 2Q21
T-Mobile: The Street Understates Earnings for 3Q21 by $0.45/share
The Street’s 3Q21 EPS estimate of $0.55 for T-Mobile is understated by $0.45/share due, at least in part, to large restructuring costs in historical EPS. Our Core EPS estimate is $1.00, which makes T-Mobile one of the companies most likely to beat Wall Street analysts’ expectations in its calendar 3Q21 earnings report.
Unusual expenses, which we detail below, materially reduced T-Mobile’s 2Q21 TTM Street and GAAP Earnings and make profits look worse than Core EPS. When we adjust for all unusual items, we find that T-Mobile’s 2Q21 TTM Core EPS are $5.68, which is better than the 2Q21 TTM Street and GAAP EPS of $3.12.
Figure 4: Comparing T-Mobile’s Core, Street, and GAAP Earnings: TTM as of 2Q21
Sources: New Constructs, LLC, company filings
Below, we reconcile the differences between T-Mobile’s 2Q21 TTM Core EPS and GAAP EPS so readers can audit our work. Usually, we cannot reconcile Core EPS to Street EPS because we do not have the details as to exactly what makes Street EPS differ from GAAP EPS. However, given that Street and GAAP EPS are the same, Street EPS are likely missing the same unusual gains/losses as GAAP EPS.
Figure 5: T-Mobile’s GAAP Earnings to Core Earnings Reconciliation: TTM as of 2Q21
Sources: New Constructs, LLC and company filings.
More details[4]:
Hidden Unusual Expenses, Net = -$3.22/per share, which equals -$4.0 billion and is comprised of:
- -1.0 billion in merger costs recorded in selling, general, and administrative in the TTM period, based on
- -$632 million in the TTM based on -$1.3 billion in 2020
- -$145 million in 1Q21
- -$251 million in 2Q21
- -$750 million in merger-related costs in the TTM based on -$1.5 billion in merger-related costs in 2020
- -$732 million in merger costs recorded in cost of services in the TTM period, based on
- -$323 million in the TTM based on -$646 million in 2020
- -$136 million in 1Q21
- -$273 million in 2Q21
- -$612 million in restructuring plan expenses in the TTM period, based on
- -$530 million in the TTM based on -$1.1 billion in 2020
- -$34 million in 1Q21
- -$48 million in 2Q21
- -$461 million in accelerating amortization expenses in the TTM, based on
- -$77 million in the TTM based on -$153 million in 2020
- -$123 million in 1Q21
- -$261 million in 2Q21
- -$229 million in COVID-19 costs in the TTM based on -$458 million in COVID-19 costs in 2020
- -$118 million in transaction costs in the TTM period, based on
- -$8 million in 3Q20
- -$97 million in the TTM based on -$201 million in 2020
- -$13 million in 1Q21
- -$107 million in merger costs recorded in cost of equipment sales in the TTM period, based on
- -$3 million in the TTM based on -$6 million in 2020
- -$17 million in 1Q21
- -$87 million in 2Q21
- -$35 million non-cash impairment in 4Q20
- $30 million gain on receivables in the TTM period, based on
- $18 million in 1Q21
- $12 million in 2Q21
Reported Unusual Expenses, Net = -$0.26/per share, which equals $326 million and is comprised of:
- -$326 million in “other expenses” in the TTM period, based on
- -$99 million in 3Q20
- -$101 million in 4Q20
- -$125 million in 1Q21
- -$1 million in 2Q21
Tax Distortion = $0.91/per share, which equals $1.1 billion
- We remove the tax impact of unusual items on reported taxes when we calculate Core Earnings. It is important that taxes get adjusted so they are appropriate for adjusted pre-tax earnings.
Conclusion: Generate Novel Alpha with More Reliable Core Earnings
As demonstrated above, Core Earnings do a better job of excluding unusual gains and losses and provide a more reliable earnings measure. All Core Earnings adjustments to GAAP Earnings are 100% transparent; so users can audit and trust the research.
To enable you to easily monetize our novel factor, Earnings Distortion, we present multiple trading strategies that drive alpha. Learn more below.
- ExtractAlpha presents a long/short market-neutral strategy that generates 9.3% annualized return net of Fama-French 5 factors, momentum, short-term reversal, and 12 sectors. More details here.
- CloudQuant presents two strategies to monetize the alpha in Earnings Distortion:
- the long-only portfolio outperformed the S&P 500 by an average of 4% per annum over 10 years
- the dollar-neutral long-short portfolio returned 60% over 10 years with a Sharpe Ratio of ~1 over the last five years. More details here.
- AltHub presents three strategies to monetize the alpha in Earnings Distortion:
- Earnings Distortion S&P 500 Smart Beta Portfolio: 10-yr annualized return of 13.9% vs 12.1% for the S&P 500 with a Sharpe Ratio of 0.97.
- Earnings Distortion ML Model S&P 500 Smart Beta Portfolio: 3-yr annualized return of 18.2% vs 13.7% for the S&P 500 with a Sharpe Ratio of 0.82.
- Truth Stocks (Companies without Earnings Distortion) Portfolios:
- S&P 500 Universe: 10-yr annualized return of 16.8% vs 12.8% for the S&P 500 with a Sharpe Ratio of 0.7.
- Russell 3000 Universe: 9.5-yr annualized return of 28.4% vs 12.6% for the S&P 500 with a Sharpe Ratio of 1.
- More details here.
This article originally published on October 8, 2021.
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.
[1] The same is true for other well-known “adjusted” measures of earnings: S&P’s “Operating Earnings” Remain Overstated in 2Q21.
[2]The Journal of Financial Economics proves that only Core Earnings enable investors to overcome the flaws in legacy fundamental data.
[3] The most recent Core Earnings and Street Earnings values are based on the latest audited financial data from calendar 2Q21 10-Qs.
[4] For unusual items found only in the latest 10-K, we show the amount applied to TTM calculation and link to the disclosure in the 10-K.