In a recent write-up from Barron’s, author Jack Hough argues that Google (GOOG) shares are undervalued. In the article (here), he makes the following statement:
“Shares trade at 18 times projected profits for the next four quarters, versus 16 times for the broad market. Google’s earnings are forecast to grow 16% to 18% yearly over the next five years. In other words, investors pay a slight premium for much faster growth potential.”
A quick look shows the stock trades for ~$525 per share; the sidebar to the article shows an estimated $25.88 per share of earnings for fiscal 2014 – roughly 20x the 2014 results (difference from multiple discussed above is next four quarters vs current fiscal year). The sidebar states that this estimated figure is 16.5% higher than the result for the prior year.
There’s only one problem: Google didn’t earn $22.21 per share in FY13 (the amount needed to make the math work); the 10-K for the prior year shows diluted EPS of $19.07, adjusted for the recent stock split. Using the same expected increase (+16.5%), we would expect diluted EPS for FY14 to be slightly above $22 per share – nearly 15% lower than the figures cited by Barron’s.
Let’s be clear about something: the Barron’s write-up makes no mention of this discrepancy (online version); the reason why we’re so far apart is their use of non-GAAP earnings. The adjustments are due to the accounting for stock-based compensation expense.
Apparently, Barron’s felt there was no need to address the material adjustments made. In my experience, the analyst community largely agrees – they accept the results as presented as well.
For years, Google’s management team has maintained that this is appropriate:
“Our management believes that these non-GAAP financial measures provide meaningful supplemental information regarding our performance and liquidity by excluding certain expenses and expenditures that may not be indicative of our recurring core business operating results…”
Whether or not outside investors accept that explanation is another question (as I noted in an article a few months back, I don’t); as the EPS changes show, this materially impacts results.
One would assume that Barron’s has reviewed this disclosure, and accepts the rationale provided by management; as a result, they’ve decided that they will not include stock based compensation charges in their calculation of EPS. It’s not clear why they decided to do so without telling their readers (it seems logical that the starting assumption is GAAP, unless otherwise stated).
As a result, more than $4 billion of expense simply evaporated from the income statement (for the FY14 estimate); on a pre-tax basis, were talking about more than $50 billion in market value – all of which has been accepted without a single mention by a leading investment publication.
The odd thing is that this doesn’t appear to be a conscious decision by the Barron’s staff. As an example, they wrote a favorable article about Intel (INTC) a few weeks back (here), saying the following: “Earnings per share are expected to climb 19% this year to $2.25…”
Unlike Google, Intel’s management team doesn’t back out stock based compensation as a non-GAAP expense when reporting results to investors. As a result, Barron’s includes the ~$1.1 billion the company will spend on stock based compensation in their earnings estimate for 2014 (backed into from 2013 GAAP results); if they made the same adjustment as they’re willing to accept for Google, Intel’s pre-tax EPS would be higher by more than $0.20 per share.
The same can be said for plenty of other companies, especially in tech: in their most recent fiscal years, Microsoft (MSFT) and Apple (AAPL) spent $2.45 billion and $2.85 billion on stock based compensation expense, respectively. All of these companies feel the impact of a rising share count (which Barron’s called out in the end of their GOOG write-up); the difference is that Intel, Microsoft and Apple all show the impact on the income statement as well.
Why the differing treatment? The answer is quite simple: Google is the only company among its large peers that breaks out stock-based comp as a non-GAAP expense. It seems that Barron’s has concluded that rather than make the adjustment to whatever they consider to be the proper convention (either back out stock-based comp for everyone or present GAAP results for Google), they’ve instead concluded that they’ll simply report whatever management decides to give them. This seemingly innocuous conclusion results in differing price estimates with market valuation swings in the tens of billions of dollars (assuming multiples are justified as presented – the argument made in the first quote of this article suggests that’s the case).
I think it's pretty stunning that a leading financial publication would present non-GAAP results that amount to such significant adjustments to the P&L without batting an eye (or even making a footnote disclosure it to their readers – again, for the online edition).
A few years ago, Barron’s addressed the issue of reporting for stock-based compensation head on (here), saying the following:
“Many highflying tech companies encourage investors to ignore significant stock-based compensation expense when calculating earnings. Investors willingly oblige.”
Later in that article, they made a sound suggestion to readers:
“Investors probably should re-evaluate how they stack up [those that back out stock based compensation] relative to companies inside and outside tech that use standard earnings calculations.”
It might be time for Barron’s to consider their own advice and start reporting Google’s GAAP EPS – or at least let readers know when they “willingly oblige” and accept non-GAAP numbers.
