SYNOPSIS: Please note that Paul Krugman has written a correction to the last paragraph of this article

For the last few months there has been a running debate about the U.S. economy, more or less like this:

American families: "We're not doing very well."

Washington officials: "You're wrong - you're doing great. Here, look at these statistics!"

The administration and some political commentators seem genuinely puzzled by polls showing that Americans are unhappy about the economy. After all, they point out, numbers like the growth rate of G.D.P. look pretty good. So why aren't people cheering?

Some blame the negative halo effect of the Iraq debacle. Others complain that the news media aren't properly reporting good economic news. But when your numbers tell you that people should be feeling good, but they aren't, that means you're looking at the wrong numbers.

American families don't care about G.D.P. They care about whether jobs are available, how much those jobs pay and how that pay compares with the cost of living. And recent G.D.P. growth has failed to produce exceptional gains in employment, while wages for most workers haven't kept up with inflation.

About employment: it's true that the economy finally started adding jobs two years ago. But although many people say "four million jobs in the last two years" reverently, as if it were an amazing achievement, it's actually a rise of about 3 percent, not much faster than the growth of the working-age population over the same period. And recent job growth would have been considered subpar in the past: employment grew more slowly during the best two years of the Bush administration than in any two years during the Clinton administration.

It's also true that the unemployment rate looks fairly low by historical standards. But other measures of the job situation, like the average of weekly hours worked (which remains low), and the average duration of unemployment (which remains high), suggest that the demand for labor is still weak compared with the supply.

Employers certainly aren't having trouble finding workers. When Wal-Mart announced that it was hiring at a new store in Northern California, where the unemployment rate is close to the national average, about 11,000 people showed up to apply for 400 jobs.

Because employers don't have to raise wages to get workers, wages are lagging behind the cost of living. According to Labor Department statistics, the purchasing power of an average nonsupervisory worker's wage has fallen about 1.5 percent since the summer of 2003. And this may understate the pressure on many families: the cost of living has risen sharply for those whose work or family situation requires buying a lot of gasoline.

Some commentators dismiss concerns about gasoline prices, because those prices are still below previous peaks when you adjust for inflation. But that misses the point: Americans bought cars and made decisions about where to live when gas was $1.50 or less per gallon, and now suddenly find themselves paying $2.60 or more. That's a rude shock, which I estimate raises the typical family's expenses by more than $900 a year.

You may ask where economic growth is going, if it isn't showing up in wages. That's easy to answer: it's going to corporate profits, to rising health care costs and to a surge in the salaries and other compensation of executives. (Forbes reports that the combined compensation of the chief executives of America's 500 largest companies rose 54 percent last year.)

The bottom line, then, is that most Americans have good reason to feel unhappy about the economy, whatever Washington's favorite statistics may say. This is an economic expansion that hasn't trickled down; many people are worse off than they were a year ago. And it will take more than a revamped administration sales pitch to make people feel better.

Corrections: In my column last Friday, I cited an inaccurate number (given by the Conyers report) for turnout in Ohio's Miami County last year: 98.5 percent. I should have checked the official state site, which reports a reasonable 72.2 percent. Also, the public editor says, rightly, that I should acknowledge initially misstating the results of the 2000 Florida election study by a media consortium led by The Miami Herald. Unlike a more definitive study by a larger consortium that included The New York Times, an analysis that showed Al Gore winning all statewide manual recounts, the earlier study showed him winning two out of three.

Originally published in The New York Times, 8.26.05