Every presidential election is an exercise in hope and exaggeration. We hope "our" candidate's triumph will uplift the nation. But the usual campaign exaggerations may deceive and disillusion us. In two recent columns, I have tried to dampen unrealistic expectations on the issue of jobs. I've argued that you should discount the candidates' rhetoric that they can easily affect the number or quality of new jobs. Outside forces -- the business cycle, new technologies, mass psychology -- eclipse a president's powers, I wrote.

The message hasn't taken. Many readers were appalled. Especially offended were unemployed workers. "I am a 40-year-old college-educated white male who has an electrical-engineering degree," wrote one. "Until three years ago, I was a middle-class American with very few bills and loads of disposable income. [Then] many of my college-educated friends and I were laid off from [our] software company. . . . While I do not blame Republicans for my layoff, as the economy was starting to go into recession before they took office, I do fault them for doing absolutely nothing for millions of jobless high-tech workers."

I take these objections seriously. Let me try to answer some readers' comments. As we proceed, remember a few facts: First, the number of U.S. jobs now exceeds 131 million; second, the economy needs to add 2 million to 3 million jobs annually to absorb new workers and reduce unemployment.

"You downplay the quality of the jobs being lost to offshore outsourcing -- information technology is reeling from the offshore menace."

The recession and the bursting of the "tech bubble" -- not offshoring -- explain the worst job losses. Recall that in the late 1990s companies couldn't get enough computer specialists. The Internet boom and Y2K drove demand, says Andrew Bartels of Forrester Research. In 2000 alone, information technology jobs rose 9.8 percent, to 3.37 million, he says. Then Internet, Y2K and telecom spending collapsed. Layoffs followed. Now hiring has resumed, though it's hurt by offshoring. No one knows how many IT jobs are leaving. Forrester estimates 102,000 in 2003 and 143,000 in 2004. Unless someone abolishes the Internet, it will be hard to stop offshoring. Still, Forrester expects IT jobs to grow 2 percent annually from 2003 to 2008.

"You're wrong in saying that 'many jobs being lost abroad to other countries are actually low-skilled and low-paying (that's why they're lost).' "

Most highly skilled and well-paid Americans don't face competition from poorer countries, because these countries don't have lots of highly skilled workers. Offshored IT jobs represent something of an exception. The more common threat is to low-paid workers in labor-intensive industries, where wage gaps with workers in poorer countries are staggering. Consider the U.S. textile and apparel industries, whose hourly wages averaged $12 and $9.56 in 2003. That's well below the average U.S. manufacturing wage ($15.74) but well above wages in many poor countries. In Sri Lanka, factory wages average about 50 cents an hour. American IT jobs are still growing. Textile and apparel jobs aren't. Since early 1999 they've dropped from about 1.2 million to 700,000.

"The U.S. government is a [big] employer and can add large numbers."

True -- but the cost would be huge. In 2003 the federal government had 1.9 million civilian workers. The military (1.5 million) and the Postal Service (800,000) bring the total to 4.2 million, equal to about 3 percent of U.S. jobs. Now, suppose the government hired another 1 million workers at $50,000 each ($40,000 in wages, $10,000 in fringes). The cost would be $50 billion a year forever. To hire 1 million more workers the next year would require another $50 billion commitment. Exactly what would all these workers do?

"Recall Franklin Roosevelt? How many jobs was he instrumental in creating?"

Good question -- but whatever the answer, it wasn't enough to end the Depression. FDR prevented the Depression from getting worse, mainly by relaxing the gold standard. In 1933 unemployment was 25 percent; by 1937 it was 14 percent. But other policies had mixed results; some studies conclude that certain New Deal programs hurt job creation. In any case, unemployment rebounded to 19 percent in 1938. World War II ended the Depression.

Job creation is a market process. Companies hire when they think that demand and profit prospects justify more workers. I am not arguing that government policies don't matter. Taxes, regulations and subsidies create an economic climate -- for better or worse. In Europe, excessive taxes and regulations have stymied job creation. Sometimes government can, through changes in interest rates and budgets, affect the business cycle. But the whole process resists the purposeful manipulation implied by George Bush and John Kerry when they claim they can create a given number of jobs or a higher quality of jobs.

On this subject, the candidates seem equally driven to hype and overpromise. The "jobs issue" is said to be fading in political importance. If so, it's no great loss.