Why do Americans feel so insecure in an economy that, by most standards, is chugging along just fine? 

Public opinion polls may themselves be to blame. Respondents are usually more optimistic about their own economic prospects than the country's—a bias that invariably skews poll results downward. Moreover, as Steve Sailer notes re immigration polls, wording matters greatly.

On the other hand, the economy may simply not be as healthy as the economic statistics indicate. A recent memo by the Economic Policy Institute titled What's Wrong with the Economy, highlights areas in which economic weakness coexists with ostensibly strong economic data.

"Profits are up, but the wages and incomes of average Americans are down. Inflation-adjusted hourly and weekly wages are below what they were at the start of the recovery in November 2001. Yet productivity—the growth of the economic pie—is up by 14.7%." 

Source: What's wrong with the economy? by EPI President Lawrence Mishel and Policy Director Ross Eisenbrey, based on their analysis of BLS data.

Labor productivity measures growth in output produced by the average American worker in an hour. Rising productivity should mean that employers can pay workers more without reducing profits or raising prices.

The (alleged) 14.7  percent gain in output per hour since 2001 is well above the historical average, and—if accurate—should by now have produced the best of all worlds for business and  labor, as well as for  Ben Bernanke.

Profits have indeed risen sharply since 2001. But incomes are stagnant and inflation is higher than the Fed and most economists had anticipated.

What went wrong?

One possible answer: labor productivity may not be growing nearly as fast as the official statistics indicate. That's because BLS calculates productivity using payroll survey employment figures instead of the larger, more rapidly growing, household survey numbers.

While the most recent payroll survey estimates 134.7 million workers held jobs in the first quarter of 2006, the household survey counted 143.3 million – nearly 9 million more.

More importantly, employment growth since the fourth quarter of 2001 is estimated at 3.8 million by the PS versus 7.1 million in the HS. (Table 1.)

Implication: labor productivity is growing about half as fast as the official statistics indicate.

It gets worse. As noted above, productivity is measured per hour, not per worker. If the "missing" workers are largely illegal Hispanics who work longer hours than other workers, productivity growth will be overstated by even more than the employment undercount alone would suggest.

Some economists argue that the cyber-commuters, temporary workers, and others who are not on payrolls but show up in the household survey, are the major reason why employment growth is understated.

But there's a better explanation—unmentioned (not for the first time) by the left-leaning EPI: illegal aliens. They do not show up in the PS for the simple reason that employers who admit to hiring them risk stiff penalties. (Absent such an admission, employers can—and do—break the law with impunity.)  

Estimates of the illegal alien workforce range from 7.2 million ( Pew Hispanic Center) to 20 million (Bear Stearns) The consensus gravitates to between 8 to 9 million—a figure strikingly close to the gap between the two employment surveys.

The illegal immigrant labor force may be the missing link between apparently strong economic data and weak economic reality—for American workers, if not for American capitalists.

Edwin S. Rubenstein (email him) is President of ESR Research Economic Consultants in Indianapolis.