Nardelli's fake bogey: earnings per share

NEW YORK (Fortune) -- After the buckets of ink - and oh, the buckets of money! - that have been spilled on the departure of former Home Depot CEO Bob Nardelli, you might think there would be nothing left to say. Wrong! While everyone (everyone but The Wall Street Journal editorial page, that is) may know that Nardelli's ouster may mark the height of well-deserved outrage about CEO pay, it may also mark the death of something else: earnings per share.

During Nardelli's reign, when analysts and investors would complain about performance, he would respond by saying that Home Depot had delivered earnings per share growth in excess of 20 percent for four consecutive years. Home Depot (Charts) was one of only two companies in the Dow Jones industrial average to accomplish that feat.

Robert Nardelli stepped down as CEO of Home Depot in January - with a $210 million severance package.

More from Fortune

Small banks, big problems Let us praise the venture capitalists The yin and yang of cybersecurity

FORTUNE 500 Current Issue Subscribe to Fortune

And yet, the stock price lagged: It sank roughly 12 percent over Nardelli's time at the helm. (Total return, which includes dividends, was a negative 7 percent.) But isn't earnings growth supposed to drive stock prices?

Maybe not. It turns out that there could be an answer to the puzzle, courtesy of a research firm called Matrix Investment Research, which uses a methodology called "economic value added" to measure the performance of companies. (For more on Matrix, see this column.)

Their research reports don't even mention earnings per share, which shows you how relevant they think that measure is. What matters is return on capital. "The biggest impact on stock price is the rate of change of return on capital, or the slope of the line," says research head Ivan Feinseth.

Under Nardelli, Home Depot's return on capital remained high, but it bounced around - from 14.2 percent in 2001 to 17.4 percent in 2002 to a high of 18.6 percent in 2005 and back down to 16.7 percent in 2006, according to Matrix's numbers. (These measurements are taken not at the end of Home Depot's fiscal year, but rather for the trailing 12 months as of the end of the third quarter.) Over time, Home Depot's return on capital didn't improve much. And neither did the stock.

As for Home Depot's earnings-per-share growth, although Nardelli was monomaniacal about it, he achieved his bogey at least in part "in artificial ways," says Feinseth. Those include share buybacks - since 2002, Home Depot has bought back some 450 million shares, or 19 percent of its outstanding shares - and cost cutting. You may not be surprised to learn that during 2004, Home Depot changed the executive long-term incentive plan so that in future years, payouts would be linked not to stock performance, as they were for past grants, but instead to ... EPS growth!

(In fairness to Home Depot, it's easy to criticize just about any compensation methodology. For instance, plans that are linked to stock price may also cause executives to attempt to juice the price short-term.)

Now, just for the fun of it, contrast Home Depot to arch rival Lowe's (Charts), which has also reported steady growth in earnings per share. But unlike Home Depot, Lowe's return on capital, according to Matrix, has risen steadily, from 10.2 percent in 2001 to 13.5 percent in 2003 to 16 percent in 2006. (Lowe's numbers are calculated on the same basis as Home Depot's.) While Lowe's still has a lower return on capital than Home Depot, it's the growth, and the steadiness of that growth, that matters. And - surprise! - Lowe's stock price has risen 180 percent over the last six years.

Home Depot does not agree with this analysis. A spokesman says that their calculations show that return on capital has increased from 18.3 percent in 2001 to 22.4 percent in 2005. Home Depot also says that return on capital was dragged down by a big acquisition in 2006, and that if it really were the determinant of their stock price, then the stock should have climbed through 2005 - which it didn't.

Feinseth responds that indeed, "no company is ever going to agree" with Matrix's numbers, because Matrix makes a number of adjustments. Most importantly for a retailer, it capitalizes leases. Feinseth says that this process not only gets to the "real numbers," but also allows Matrix to make apples-to-apples comparisons between companies, such as Home Depot and Lowe's. Otherwise, because every company calculates return on capital in its own way, a comparison is useless. "An apples-to-apples comparison shows that Lowe's has kicked Home Depot's butt," says Feinseth.

As a not-so-small aside, for any of those who still believe that pay has any relationship to performance, check out Nardelli's pay relative to that of Robert Niblock's at Lowe's.

Over the last three years, Nardelli's base salary, bonus, and other compensation (loan forgiveness) totaled $37.4 million - more than quadruple the $8.1 million that Niblock got! The picture gets a lot worse when you add in grants of restricted stock and options. Over the last three years, Nardelli got almost $40 million in restricted stock, and over 1.7 million options. Niblock got $9.4 million in restricted stock and 272,000 options.

As for severance pay...oops! Only one of these guys lost his job.

It's also worth noting that new Home Depot CEO Frank Blake has a radically different compensation structure than his predecessor. He may earn up to $8.9 million this year, but a hefty portion of that is at risk, and he has no severance package.

End note: Matrix does currently rate Home Depot a "strong buy." Although its performance has been mediocre, says Feinseth, it is very low risk.

__________________

Home Depot ends spat with investor group

Judge won't block Nardelli's huge severance