Earlier this month, President George Bush's chief economic adviser, Gregory Mankiw, once Harvard's youngest tenured professor, attracted a storm of abuse. He told Congress that if a thing or a service could be produced more cheaply abroad, then Americans were better off importing it than producing it at home. As an example, Mr Mankiw uses the case of radiologists in India analysing the X-rays, sent via the internet, of American patients.

Mr Mankiw's proposition, in essence, is the law of comparative advantage, first postulated by David Ricardo two centuries ago and demonstrated to astonishing effect since. Yet the Republican speaker of the House of Representatives, Dennis Hastert, joined Democrats in their rebuke of Mr Mankiw for approving of jobs going overseas; another Republican called for his resignation. The White House gave Mr Mankiw only lukewarm support - unsurprisingly, since Mr Bush recently signed a bill forbidding the outsourcing of federal contracts overseas. And the Democratic presidential contenders? Mr Mankiw had just written their attack ads.

As if to underline the point, this week's Wisconsin primary was dominated by the subject of jobs, and the failure of the Bush administration to do enough to protect them from going off to India. In John Edwards, who wants to rewrite the North American Free-Trade Agreement (NAFTA), the American left may have found its cuddliest protectionist yet; support for the southerner surged after he spent much of a debate drawing implicit comparisons between his own skills as a jobs-defender and those of John Kerry, who has stuck to free trade only a little more loyally. The Democratic front-runner defends NAFTA, but rants about "Benedict Arnold" bosses betraying American workers by moving jobs overseas (presumably to boost returns for fat-cat investors, like, er, Mr Kerry's family).

As for what might be called the business lobby, this is in disarray. "Tech jobs are fleeing to India faster than ever," moans the cover of Wired. Watch "Lou Dobbs Tonight", America's main business show, and every factory-closing is hailed as proof of America's relentless "hollowing-out" at the hands of dark forces in China, India and indeed the White House. Strangely, no mention is made of the fact that a pretty tiny proportion of all jobs lost actually go overseas.

So what is really happening? Three themes emerge:

Although America's economy has, overall, lost jobs since the start of the decade, the vast majority of these job losses are cyclical in nature, not structural. Now that the economy is recovering after the recession of 2001, so will the job picture, perhaps dramatically, over the next year.

Outsourcing (or "offshoring") has been going on for centuries, but still accounts for a tiny proportion of the jobs constantly being created and destroyed within America's economy. Even at the best of times, the American economy has a tremendous rate of "churn" - over 2m jobs a month. In all, the process creates many more jobs than it destroys: 24m more during the 1990s. The process allocates resources - money and people - to where they can be most productive, helped by competition, including from outsourcing, that lowers prices. In the long run, higher productivity is the only way to create higher standards of living across an economy.

Even though service-sector outsourcing is still modest, the growing globalisation of information-technology (IT) services should indeed have a big effect on service-sector productivity. During the 1990s, American factories became much more efficient by using IT; now shops, banks, hospitals and so on may learn the same lesson. This will have a beneficial effect that stretches beyond the IT firms. Even though some IT tasks will be done abroad, many more jobs will be created in America, and higher-paying ones to boot.
Just You Wait
The "jobless recovery" first, then. Despite strong productivity growth and an accelerating recovery from the recession of 2001 (the economy grew by an annual 4% in the fourth quarter of last year), jobs are being created at a feeble rate of 100,000 or so a month. The jeremiahs point out that a net total of 2.3m jobs have been lost since Mr Bush came to office.

Although this date is often used as the starting-point from which to make a comparison, it is a silly one. In early 2001 the hangover effects from the investment boom of the late 1990s were only starting to be felt. Unemployment, at 4.2%, was unsustainably below the "natural" unemployment rate, consistent with stable inflation, that most economists put at around 5%. In other words, perhaps two-thirds of those 2.3m jobs were unsustainable "bubble" ones. Given the scale of job losses - along with the shocks of a stockmarket bust, corporate-governance scandals and terrorist attacks - it is a wonder that the recession was so mild. By the same token, a mild recession is now being followed by a commensurately mild recovery.

This week, the White House retreated from a claim that 2.6m new jobs would be created this year. But there are reasons to think that job growth will be more robust. In particular, the remarkably strong productivity growth, running at twice its long-run average of 2.1%, must slow down eventually. In the face of rising order books, businesses will have to hire more workers.

This may already be happening in some parts of the country. William Testa, director of regional research at the Federal Reserve Bank of Chicago, points out that the downturn began in the mid-west (because of its relative emphasis on manufacturing, notably business equipment, the mid-west was hit first by the slump in business investment) and then spread to the coasts. Now a recovery is spreading in the reverse direction - starting on the coasts and ending up, alas for Mr Bush, in the key electoral states of the industrial heartland.

In the absence of an obvious jobs recovery, it is perhaps not surprising that the myth arose that the American economy was being buffeted by structural, not cyclical, forces. Yet it nevertheless is a myth - as three notable economists, William Baumol, Alan Blinder and Edward Wolff, point out in a recent book ("Downsizing in America: Reality, Causes and Consequences," Russell Sage Foundation).

Churning, they point out, has being going on in the American jobs market for years, and "the creation of new jobs always overwhelms the destruction of old jobs by a huge margin." Between 1980 and 2002, America's population grew by 23.9%. The number of employed Americans, on the other hand, grew by 37.4%. Today, 138.6m Americans are in work, a near-record, both in absolute terms and as a proportion of the population (see chart).

Of course some firms wither - Reynolds Tobacco's workforce shrank by nine-tenths between 1980 and 2002 - but others grow: Wal-Mart's by 4,700%. During the 1990s, about a quarter of all American businesses shed jobs in a typical three-month period, equivalent to 8m jobs. Yet jobs created greatly outnumbered these, to the tune of 24m over the decade.

The process leads to incremental shifts that can have profound cumulative consequences for some sectors of the economy. In 1960 only one in 25 workers was employed in the business-services and health-care industries. Today, one in six is. In terms of output, manufacturing has risen, but, thanks to that productivity spurt, these goods are produced by fewer people - 12% of the workforce, less than half the proportion of three decades ago.

And what of China? Still piffling. Certainly, China competes with some labour-intensive American industries that have long been in decline, such as textiles and stuffed toys. In the mid-west, metal-furniture makers and small tool-and-die foundries face growing competition. Yet most Chinese imports are of consumer goods, competing with imports from other poor countries, whereas America's manufactures are chiefly capital goods. Even at their peak in 2001, the number of all "trade-related" layoffs represented a mere 0.6% of American unemployment.

As for the Indian threat, "offshoring" is certainly having an effect on some white-collar jobs that have hitherto been safe from foreign competition. But how big is it, really? The best-known report, by Forrester Research, a consultancy, guesses that 3.3m American service-industry jobs will have gone overseas by 2015 - barely noticeable when you think about the 7m-8m lost every quarter through job-churning. And the bulk of these exports will not be the high-flying jobs of IT consultants, but the mind-numbing functions of code-writing.

Meanwhile, there is another side to the ledger. Instead of focusing on jobs lost to the globalisation of information technology, Catherine Mann of the Institute for International Economics in Washington looks at globalisation's power to reduce prices and so help spread new technology, new practices and job-creating investment through the economy.

She uses the example of cheaper IT hardware, one of the main aspects of globalisation in the 1990s. Most of the drop in prices for PCs, mainframes and so on was caused by the relentless advance of technology; but she still thinks that trade and globalised production - all those Dell Computer factories in China, for instance - was responsible for 10-30% of the fall in hardware prices. These lower prices led to higher American productivity growth and added $230 billion of extra GDP between 1995 and 2002, equivalent to an extra 0.3 percentage points of growth a year.

These days, software spending is increasing at twice the rate of hardware spending, as businesses struggle to make their new computers work better. The manufacturing sector is where such integration has gone furthest. In many other parts of the American economy, the process has barely begun - particularly among smaller- and medium-sized businesses. Mr Mankiw's example of the Indian radiologist shows how the internet could help lower costs and raise productivity in health care. Who would object to that?

Ms Mann concludes that if IT software sees falls in prices, thanks to globalisation, similar to those that IT hardware has seen, then the second wave of productivity gains - notably in the service sector - could be greater than the first, which was based mainly on manufacturing. Some service sectors, such as construction and health care, are ripe for gains, because their efficient use of IT is low.

Will the trend lead to jobs going overseas? You bet, but that is not a disaster. For a start, America runs a large and growing surplus in services with the rest of the world. The jobs lost will be low-paying ones, such as bank tellers and switchboard operators. Trade protection will not save such jobs: if they do not go overseas, they are still at risk from automation.

By contrast, jobs will be created that demand skills to handle the deeper incorporation of information technology, and the pay for these jobs will be high. The demand for computer-support specialists and software engineers, to take two examples, is expected by the Bureau of Labour Statistics (BLS) to double between 2000 and 2010. Demand for database administrators is expected to rise by three-fifths. Among the top score of occupations that the BLS reckons will see the highest growth, half will need IT skills. As it is, between 1999 and 2003 (that is, including during the recession) jobs were created, not lost, in a whole host of white-collar occupations said to be particularly susceptible to outsourcing.

Yes, individuals will be hurt in the process, and the focus of public policy should be directed towards providing a safety net for them, as well as ensuring that Americans have education to match the new jobs being created. By contrast, regarding globalisation as the enemy, as Mr Edwards does often and Messrs Kerry and Bush both do by default, is a much greater threat to America's economic health than any Indian software programmer.

© CFO Publishing Corporation 2004. All rights reserved.