Those days are in the past.
A positive outcome of the Enron scandal was the realization among investors that corporations don’t always exhibit 100-percent integrity. Making this worse, auditing firms were operating with a potentially glaring conflict of interest. Even today, about five years since the passage of Sarbanes-Oxley, auditing firms continue to gain substantial revenue from nonaudit work, often for the same companies for which they perform annual audits.
The larger the revenue number, the bigger the possible conflict of interest, but there’s no end in sight for this problem. Even if it were possible to completely trust companies and their audit results, the bigger question is whether the numbers themselves are accurate in the first place.S&P’s Core Earnings
To address this question, Standard & Poor’s developed the concept of “core earnings.” It originally was a way for S&P to adjust its credit rating of corporate debt instruments. The idea was to make adjustments to reported profits by taking out noncore earnings and adding any core expenses that had been excluded.
Examples of noncore earnings include capital gains, profits from exchange rate fluctuations and pro forma income on investments of pension plans. Expenses left out by some companies included stock options (which weren’t booked as expenses but served as a form of compensation) and restructuring charges.
In its first full year providing core earnings, S&P reported average earnings of $26.74 per share for companies in the S&P 500 index. But when adjusting for core earnings, the number fell to $18.48, a reduction of more than 30 percent.
Dozens of listed companies have core earnings adjustments each year. The difference between their reported net earnings and S&P-calculated core earnings often run into the billions of dollars.
In 2002, one of the biggest adjustments was made to earnings of E.I. du Pont de Nemours (ticker DD). The table on this page summarizes reported and core outcomes for DuPont’s past five fiscal years.
The adjustments are enormous. In 2002, the first reported year, the difference between reported net earnings and calculated core earnings was $1.4 billion, and an analyst relying on that $1.84 per share would need to adjust to a core-based EPS of only 40 cents.
DuPont’s rather large adjustment in 2002 wasn’t even the most extreme case. AT&T (T) and IBM (IBM) both reported net earnings that were more than $5 billion above core earnings. In addition, Ford Motor Co. (F), General Electric (GE), General Motors (GM) and Verizon (VZ) all were subject to core adjustments above $3 billion in 2002.
The 2002 adjustment for IBM, the Dow Jones industrial average’s largest component stock, was one of the highest core adjustments in 2002. Its five-year history is in the table on this page.
The adjustments in this case were even more severe than DuPont’s, especially in fiscal 2002, when core earnings were 1/38 of reported EPS. That’s quite a variance. But even 2005 core EPS was down nearly $1 per share, or about 20 percent, from reported earnings. These are big adjustments, and they distort the entire range of fundamental analysis.Critical Judgment About Core EPS
Because the core adjustments are made only to calculate ratios and not to change actual reported results of companies, you might well question whether this is a serious problem. When you consider the cascading effect of these adjustments on virtually all forms of fundamental analysis, however, the importance becomes more apparent. A change in EPS is only the surface adjustment. The net return, price-earnings ratio and return on investment are all affected.
If you recognize the big differences for some companies between reported and core earnings, the problem goes even further. Company-to-company comparisons can also be inaccurate. A company with huge core adjustments cannot be accurately compared with another with no core adjustments.
Despite the efforts of many companies to portray their current business results truthfully, financial results include a fair amount of guesswork. Worse, companies have ways of bending results within “the rules” of generally accepted accounting principles (GAAP). Think about what this actually means. If you were allowed to report your earnings in filling out loan applications on the same basis that corporations report noncore earnings, it would cause great chaos.
Let’s say you earned $75,000 last year but also made a net profit of $135,000 from selling your residence. Is it realistic for you to apply for a loan by reporting to the bank that last year’s “net earnings” totaled $210,000? Following the same standards companies use in reporting, the answer would be yes. Although no lender would allow this to go through without removing the capital gain, this is exactly what corporations do when they don’t adjust their noncore revenue and report it below the net earnings line. Managing Earnings
Some companies have had very little or no core earnings adjustments, such as Wal-Mart (WMT) and Microsoft (MSFT). And some adjustments actually place core earnings above reported net earnings.
Why are some companies prone to large core earnings adjustments while others aren’t? Some of these problems may be related to the industry a company is in. For example, in GM’s case S&P made large adjustments for the automaker’s very large pension assets and liabilities, whereas there were no such adjustments for Wal-Mart.
But the reasons for these large adjustments also have a lot to do with year-to-year volatility. When financial results are inconsistent, it worries investors — and it should. One of the troubling things you may discover when checking core earnings is how that volatility is disguised. A company’s standard revenue and earnings might appear quite predictable, but when core earnings adjustments are made, the numbers may change quite a bit. This is a troubling aspect of what accounting rules allow.
Investors don’t like to see revenues and net earnings jumping around wildly, up one year and down the next. It’s unsettling when the trend is impossible to establish and when forecasts are unreliable, but most people continue to restrict their analysis to the imperfect GAAP numbers. It’s equally disturbing to witness large core earnings adjustments every year, especially when they may be up or down without any predictability.
There’s an advantage to tracking the annual differences between core and reported net earnings. You’ll find a tendency for highly volatile fundamentals, including core earnings adjustments, to correspond to more volatile trading ranges in stocks. The opposite is generally true as well: Companies reporting little or no annual adjustments also tend to have less volatile stock prices, with a relatively narrow trading range and predictable gradual growth in price levels over time.
In other words, well-managed, profitable companies usually have low core earnings adjustments and safer, less volatile trading ranges. (This article is for educational purposes only. No investment recommendation is intended.)Another Tool for Analysis
This insight is valuable because it provides an extra fundamental indicator worth watching. Like all indicators, use it as a confirming piece of information along with other tests you apply, and study the relationship between core and reported earnings over several years. Just because a company shows a spike in its core earnings adjustments in one year doesn’t mean it’s no longer a quality investment.
For example, most investors consider Altria Group (MO) a high-quality company. When it sold its Miller Beer segment a few years ago, however, the event caused a one-time core earnings adjustment of $3.6 billion. This is the kind of adjustment everyone could accept as “business as unusual.”
Unfortunately, when you see big adjustments every year, it raises more questions than it answers. Highly volatile differences in the reported and core earnings might be something worth checking out.
This is why you need to keep an eye on core earnings adjustments. It can point out companies with potential reporting problems and lead you to companies with exceptionally stable reporting policies.