A number of articles have been published recently warning investors to disregard non-GAAP earnings and focus instead on GAAP earnings. For example, in yesterday's New York Times, Gretchen Morgenson, published a piece called "Earnings, But Without the Bad Stuff," that highlighted the opinions of Jack Ciesielski, publisher of the Analyst's Accounting Observer, who makes the case for GAAP.
This issue has received much more attention because of the strong recovery in corporate profits since the 2008 recession. Many market analysts have pointed out that corporate profits are at record levels. Some say that profits probably cannot go much higher unless revenues begin to rise at a quicker pace. (I agree.)
Source: US Bureau of Economic Analysis and Lark Research calculations
According to the Bureau of Economic Analysis (BEA), the after-tax profits of U.S. corporations grew at a compounded annual rate of 8.5% from 1987 to 2012. From 2000 to 2012, the compounded annual growth rate was 11.4%. As a result, corporate profits as a percent of nominal GDP increased from 4.1% in 1987 to 10.8% in 2012. Profits have recovered sharply since 2008, even though GDP growth has been lackluster.
A different picture emerges when we look at the after-tax reported (i.e. GAAP) profits of the S&P 500. Earnings growth for these large corporations has not been nearly as great. This is illustrated in the chart below, which compares S&P 500 as-reported profits (which exclude discontinued operations, cumulative accounting changes and extraordinary items, as defined in GAAP.) to the BEA's measure of corporate profits using 1988 as a base:
Source: BEA, S&P Dow Jones Indices and Lark Research calculations
The comparison here is striking and raises more questions. According to S&P, as-reported profits of the S&P 500 fell much more sharply in 2008 and have just recently exceeded their 2006 peak. The decline in U.S. corporate profits, on the other hand, was not nearly as severe and profits are now well above the 2006 peak. Although their definitions of profits differ, they are hopefully similar enough to make a comparison relevant. If so and since the profits of S&P 500 companies are presumably included in the national figures, most of the difference between the two should represent profits from publicly-traded small- and mid-cap corporations and closely-held businesses. It suggests that nearly all of the growth in U.S. corporate profits have come from non-large cap businesses, which is a little surprising, since smaller businesses are believed to be still struggling in the recession's aftermath.
I have not looked at the GAAP vs. non-GAAP issue as closely as others (like Mr. Ciesielski), but the following chart highlights the difference between operating earnings, S&P's proxy for non-GAAP earnings, and as-reported earnings for the S&P 500:
Source: S&P Dow Jones Indices and Lark Research calculations
The chart above plots the difference between operating and as-reported earnings as a percentage of as-reported earnings for the S&P 500. (The definitions of operating and as-reported earnings are given in this 2002 summary from accountingweb.com, but S&P Dow Jones Indices should publish up-to-date definitions on its web site.) The spikes in the chart typically occur during recessions, such as 1991-1993, 2001-2003 and 2007-2009. (The peak, which is not shown here, was in the 2009 first quarter at 527%.) Thus, the spikes occur primarily as a result of sharp declines in as-reported earnings. Operating earnings, which exclude corporate and unusual items, have much narrower peaks and valleys. That said, it does appear - according to my imperfect eyeballs - that the difference between operating and as-reported earnings, excluding the peaks, has been rising for the past 25 years.
Although S&P's operating earnings is clearly a non-GAAP measure, it employs a uniform approach that excludes corporate costs and unusual items that are not seen as recurring. In many cases, therefore, it may very well differ from each individual company's definition of adjusted or non-GAAP earnings.
Even using S&P's definition of "as-reported" earnings, it is difficult to conclude that the market is overvalued. As of October 31, the S&P 500 was trading at 18.5 times as-reported earnings, below the 25-year average of 24.7 times and median of 19.6 times. The as-reported P/E for the S&P 500 has ranged between 11.7 times and 122.4 times over that period. There may be other reasons to think that the market is richly valued, but the S&P is well within its historical ranges and could conceivably go higher in the quarters and years ahead.
Although the profits of publicly-traded corporations may not be quite as high as suggested by the National Income Accounts, the overall level of profits and especially the differences between GAAP and non-GAAP earnings should not be taken lightly. Such comparisons are probably best made on a case-by-case basis or perhaps by industry sector, however, and not in a macro analysis across the universe of publicly-traded companies.
In my opinion, it is not that difficult for an experienced security analyst to identify red flags in a company's financial reporting. It is also fair to say that most companies have them. However, it is much more difficult to know whether and when these potential problems will make a difference to the market.
For example, I believe that any competent housing analyst could have seen problems developing in the housing market by the year 2000; but it was not until 2006 that those problems became manifest. In between 2000 and 2006, many housing-related stocks appreciated at a 40% compounded annual rate.
It is also difficult to say whether red flags will become headwinds to earnings or lead to an outright collapse. Services such as those offered by Mr. Ciesielski can help portfolio managers find turning points and avoid problems, but it is also important to monitor investments closely.