Is There a Perfect Way to Measure Earnings?

Written by: David Lebovitz Investing 101 teaches us that when buying a share of stock, investors are paying for a share of future profits. What the textbooks do not tell us, however, is which measure of profits to look at. In the U.S., there are three widely accepted measures of earnings – earnings based on generally accepted accounting principles (GAAP), operating earnings, and pro forma earnings. From time to time, however, these various measures of earnings tell very different stories about what is happening with respect to a company’s profitability. GAAP earnings are the foundation of any earnings calculation, and are calculated according to rules set by the Financial Accounting Standards Board. GAAP earnings are defined as after-tax net income from continuing operations, excluding discontinued operations and extraordinary items. From here, things start getting added back. Operating earnings adjust GAAP earnings for unusual items such as restructuring charges and discontinued operations; these are determined at the company level but then reviewed by Standard & Poor’s to ensure compatibility. Finally, there are pro forma earnings, which add back things like gains/losses on asset sales, asset or goodwill impairments, M&A costs, non-cash compensation, and pension fund gains or losses. Pro forma earnings are calculated by the companies themselves based on the methodology used by the majority of sell-side analysts – so while they are widely used, they also tend to be quite subjective. Over time, these adjustments matter – for example, since 2000, cumulative pro forma S&P 500 earnings have been $365 higher than GAAP earnings, and $130 higher than operating earnings. So in a sense, by looking at pro forma earnings, you have gotten about a year's worth of extra earnings in adjustments over the past 20 years. The question for investors, however, is which measure of earnings has the highest correlation with stock market returns. Based on data back to 2001, pro forma earnings have the highest correlation; that said, the data is extremely skewed by the fact that operating earnings turned negative in 4Q08. If this data is excluded, operating earnings exhibit the highest correlation. This dynamic, coupled with the fact that there is a far longer time series available, suggests operating earnings are the best measure for long-term investors to use when attempting to gauge corporate profitability. Which earnings should I look at?