The Difference Between GAAP And Non-GAAP Q3 EPS For The Dow Jones Was 16%

The last time we looked at the near-record difference between GAAP and non-GAAP Dow Jones earnings, we found that it had crept to a (virtually) unprecedented 25%. To be sure, that was exactly one year ago, when the economy was perceived as being in worse shape than it is now, thanks to the narrative of a "global coordinated recovery" which is really just record central bank liquidity injections, and Chinese credit creation, both of which have recently hit the brakes.

That said, going back to the question of GAAP vs non-GAAP divergence, one would assume that in light of the so-called global recovery of 2017, company earnings would be more real and not the "pro forma, one-time, non-recurring" fabrication that US corporations are so fond of. Alas, one would be wrong.

As Factset's John Butters writes in a recent blog post, as of today, all of the companies in the Dow Jones Industrial Average (DJIA) have reported actual EPS for Q3 2017, which brings up several questions: what percentage of these companies reported non-GAAP EPS for Q3 2017? What was the average difference and median difference between non-GAAP EPS and GAAP EPS for companies in the DJIA for Q3 2017? How did these differences compare to recent quarters?

Here are the answers:

For Q3 2017, 21 (or 70%) of the 30 companies in the DJIA reported non-GAAP EPS in addition to GAAP EPS for the third quarter. Of these 21 companies, 16 (or 76%) reported non-GAAP EPS that exceeded GAAP EPS. Over the past six quarters (Q1 2016 – Q2 2017) 68% of the companies in the DJIA reported non-GAAP EPS in addition to GAAP EPS and 80% of these companies reported non-GAAP EPS that exceeded GAAP EPS.

Thus, slightly more companies in the DJIA reported non-GAAP EPS in Q3 2017 relative to the average of the past six quarters, while slightly fewer companies in the DJIA reported non-GAAP EPS above GAAP EPS in Q3 2017 relative to the average over the past six quarters.

For Q3 2017, the average difference between non-GAAP EPS and GAAP EPS for all 21 companies was 284.1%, while the median difference between non-GAAP EPS and GAAP EPS for all 21 companies was 10.1%. The average difference between non-GAAP EPS and GAAP EPS for the DJIA was unusually large in the third quarter because of Merck. The company reported non-GAAP EPS of $1.11 and GAAP EPS of -$0.02 for the quarter. Thus, the percentage difference between non-GAAP EPS and GAAP EPS for Merck for Q3 exceeded 5000% (on an absolute basis).

So let's normalize: excluding Merck, the average difference between non-GAAP EPS and GAAP EPS for the remaining 20 DJIA companies was 15.8%. How does that number look in context: Over the past six quarters, the average difference between non-GAAP EPS and GAAP EPS for companies in the DJIA was 72.8%, while the median difference between non-GAAP EPS and GAAP EPS was 13.4%.

Finally, if one takes the average of the median DJIA median differences for the past 4 quarters (LTM), one gets just over 14% (and 15.8% if "normalizing" the latest quarter's data).

This means that while the forward non-GAAP P/E multiple may be 18x based on a 33.4 (non-GAAP) S&P EPS, if one assumes that roughly 14% of the latest earnings, and those projected for the next 5 quarters, are "fluff" then applying a 14% haircut to the forward consensus EPS of 143... 

... which amounts to 123 in EPS for the S&P500 - then the market's forward GAAP PE multiple is 21x. With the exception of the pre-dot com burst, the market's forward P/E multiple has never been that high.


philipat Giant Meteor Mon, 11/20/2017 - 00:38 Permalink

Thanks ZH for highlighting this nefarious practice. I have been taking every opportunity possible to call out articles which use non-GAAP Earnings data. Of course, even GAAP EPS data is still subject to the distortions caused by the enormous share buyback programs that Companies have borrowred to finance thus inflating EPS to juice bonuses and options exercises of executives. Going into highly leveraged debt to juice EPS is a despicable practice which should be discouraged wherever possible.

In reply to by Giant Meteor

turing philipat Mon, 11/20/2017 - 06:46 Permalink

You are correct about using debt to increase the EPS makes no financial sense but there is more which people are overlooking. We know what the rules for the GAAP game are. But what are the rules for the  non GAAP games? They are variable in a way that always allows for a greater EPS. How convenient. Why can I not apply that to my own financial situation. For example, in math it is well founded that squaring two negative numbers yields a positve real number. Fine, my debts are signed as negative. If I square them I no longer have any debt and instead my assets have increased in proportion to the square of my debt. This makes mathematical sense and is as valid as including the debt in that part of the debt financed GNP. Therefore the higher the debt the greater the GNP. 

In reply to by philipat

buzzsaw99 Turin Turambar Sun, 11/19/2017 - 19:27 Permalink

the algos have a short attention span of about a microsecond.  they only hear "beat" or "miss" and then they're on to the next bit of news.  When you get your two minutes of corporate quarterly fame you'd better get it right the first time.Now, we're not ones to go 'round spreadin' rumors, Why, really we're just not the gossipy kind, No, you'll never hear one of us repeating gossip, So you'd better be sure and listen close the first time! [/hee haw]

In reply to by Turin Turambar

Chris88 Sun, 11/19/2017 - 19:36 Permalink

Could list hundreds of instances that GAAP is retarded, not that it matters to the folks who couldn't read a balance sheet.  Another FSA inflammatory article.

buzzsaw99 Chris88 Sun, 11/19/2017 - 19:52 Permalink

it sucks to be a corporate accountant or cfo a lot of times.  you get told to do all kinds of skeezy kitchen sink shit to try to beat already reduced estimates so you don't get cornholed by the algos.  you can say gaap is retarded, the tax code is retarded, whatever, but you can't refute that non gaap reported earnings by these fuckers always look better than gaap and that ain't no accident.

In reply to by Chris88

Chris88 MrSteve Sun, 11/19/2017 - 22:58 Permalink

Which are - non-GAAP.  Read my profile for my credentials.  Mortgage servicers - back out changes in MSRs FV for core earnings - totally makes sense.  CLO payouts better measured by cash payouts as GAAP disallows real cash flows coming in above GAAP "effective yield" method.  The list goes on and on.  

In reply to by MrSteve

Phillyguy Sun, 11/19/2017 - 20:13 Permalink

In 2008, the US faced the biggest financial collapse since the Great Depression- a direct consequences of decades of attacks on labor, tax cuts for the wealthy, financial deregulation, job outsourcing and taxpayer funded $ multi-trillion wars in Afghanistan, Iraq, Libya, Syria which have developed into strategic debacles. Since 2008 the stock market and trendy real estate have been inflated with over $4 trillion of ultracheap money from the US FED for low interest loans, share buybacks and MA deals in what David Stockman has described as an “orgy” of corporate debt. Bogus corporate earnings statements notwithstanding, none of the problems giving rise to the 2008 financial implosion have been resolved. This is going to end very badly.

Yen Cross Sun, 11/19/2017 - 21:38 Permalink

  This is why it's okay for the scumbag banksters to be forced to charge for "prospectus" advice offered to potential, current and future clients.

buzzsaw99 Yen Cross Mon, 11/20/2017 - 15:05 Permalink

Chris88 probably works for the firm that audited Merck.  lulz...According to the information and explanations given to us, no fraud on or by the Company has been noticed or reported during the course of our audit. For B S R ft Co. LLP Chartered AccountantsFirm's Registration No: 101248W/W–100022Vikas R Kasat PartnerMembership No: 105317Date : 26 February 2016  Place : Mumbai

In reply to by Yen Cross