Most manufacturing firms, looking at both their top and bottom lines, would be pardoned for assuming the U.S. economy is still in a recession. But that's not the way federal government statistics tell the economic story. The official numbers show that real GDP has risen at an annual rate of almost 3% since the unofficial end of the recession in December 2001 and that the index of industrial production has increased over 2%. In the meantime, however, U.S. private-sector employment has declined by 800,000. Why the divergence? The answer is easy, say the government statisticians, whose views are echoed by none other than the chief pooh-bah of the Federal Reserve, Alan Greenspan himself. Last year, productivity grew 4.8%, the highest rate in over 50 years. And if productivity is growing so rapidly, says Greenspan, then all must be well with the economy. We just need to wait a little while longer for robust growth to return. But suppose Greenspan is wrong, those government statistics are wrong, and the economy is still slumping. Basically, the "current-dollar" figures are OK. It's the estimated inflation rate that is wrong. And the problem is that by understating the inflation rate, the data overstate the rate of growth in real output, production and productivity. A decade ago, Greenspan got this bright idea that if the government could reduce the reported rate of inflation, it also would reduce government spending for Social Security and other programs tied to the cost of living. In turn, the federal budget deficit would be reduced, and the combination of lower inflation and a smaller deficit would permit the Fed to keep interest rates relatively low, thus stimulating the stock market and capital spending. Greenspan suggested that Michael Boskin, chairman of the White House Council of Economic Advisers under Bush the First, head up a commission to estimate the amount that the inflation rate was overstated. After the appropriate amount of time, Boskin et al dutifully reported that the average annual overstatement of CPI was 1.1%. Sufficiently chastised, the statisticians at the Bureau of Labor Statistics (BLS) made adjustments, even though many privately realized these changes would end up understating the inflation rate. In fact, many of Boskin's suggestions were appropriate. The problem is that in the intervening years, the BLS has been cowed into severely understating the growth rate in the cost of medical care. Last year it rose 13%, and this year, informed estimates put the growth at 15%. But you won't find these numbers in the government statistics. Instead, the BLS says medical-care costs rose 5% last year, and in January, increased only a minuscule 0.1%, despite the fact that many health-care providers put through their big increases of the year in January. What's more, about 15% of consumer spending goes to health care, but BLS counts only the amount paid directly by consumers -- even though indirectly consumers end up paying for it all. As a result of these and other misdemeanors, the inflation rate is understated by about two percentage points. Instead of being 2%, it is closer to 4%. That means growth adjusted for inflation is overstated. Which means that instead of being 2.5%, average annual economic growth since 1947 has been 0.5%. If the government said the rate of inflation were 4%, interest rates wouldn't be as low, and the federal budget deficit would be even bigger. But manufacturers would know where we really are in these troubled times. Michael K. Evans is chief economist for American Economics Group, Washington, D.C., and president of the Evans Group, an economics consulting firm in Boca Raton, Fla.