Concerns about "offshoring"—shorthand for offshore outsourcing, or the migration of jobs from rich countries to poor—have subsided since the 2004 presidential election. John Kerry looked for traction there, but with no real success: His focus on the issue lent his campaign an un-Clintonian pessimistic tone, and his proposals, including tax penalties for American companies resorting to the practice, met with derision from most economists. The orthodox position on the subject, then and now, was expressed by Gregory Mankiw, at that time chairman of the White House Council of Economic Advisers. His view, politically awkward but applauded by most economists (and in this space as well, for what that's worth), is that offshoring is not such a big deal and, to the extent that it matters, is good for the economy and not bad.

A fascinating new article by Alan Blinder (in the March/April issue of Foreign Affairs) dissents from the first part of that consensus. Blinder argues that offshoring is potentially a very big deal. In fact, he believes that what we have seen so far is just the timid beginning of a third Industrial Revolution.

Aside from the merits of the article in question, Blinder deserves to be listened to. He is an outstanding economist—a professor at Princeton, the co-author of one of the best introductory texts on economics, a former member of the Council of Economic Advisers, and a vice chairman of the Federal Reserve. He wrote a marvelous series of lectures on central banking ("Central Banking in Theory and in Practice"), and his book Hard Heads, Soft Hearts: Tough-Minded Economics for a Just Society is probably the best treatise on economic policy for the general reader, bar none.

What makes Blinder's new angle on offshoring so interesting is that he is not afraid of the trend, nor does he want it stopped: He believes that nothing valuable can be done to deflect or retard it. He is well disposed to market forces, and in that sense is an optimist. Yet at the same time, he envisages enormous economic disruption, and urges policy makers to think hard, and urgently, about how to prepare for this.

His is a distinctive and very persuasive take on the issue. Most people opining on the subject have been either blasé (I plead guilty) or unduly alarmed. Blinder says that neither response is appropriate. Offshoring will be good for us in the end, but it could be a huge and wrenching phenomenon, painful for very many people, and a much bigger thing than generally acknowledged.

What is new in offshoring these days is that, for the first time, many jobs in services are at risk as well. Improvements in computing and telecommunications mean that when you pick up the phone to talk to your bank, you are quite likely to get Bangalore on the line. Yet, up to now, most economists have been relaxed about offshoring, partly because the numbers involved—large as they may sound—are really quite small. The evidence, such as it is (and Blinder rightly complains that the data on this are thin), suggests that less than a million jobs in services have been moved abroad to date. A million sounds a lot. Is it? No, in fact, not when you remember that ordinary job turnover in the United States displaces more than half a million people every week. As I noted in a previous column, Forrester Research, a consulting company, has estimated that 3 million jobs might move over the next 10 years (a forecast that aroused a lot of agitated commentary); but that figure is just 300,000 jobs a year, barely 1 percent of normal job turnover.

Blinder comes up with a very different ballpark figure of jobs potentially at risk. First, he says, there is the manufacturing sector, more or less the whole of it: about 14 million jobs. Because the vast majority of manufacturing workers make things that can be put in a box and shipped, those jobs are movable. Construction and mining jobs, nearly 8 million, will stay ("You cannot hammer a nail over the Internet"). Ditto, for political reasons, jobs in local, state, and federal government: an additional 22 million. Most retail trade, 16 million jobs, also requires a physical or face-to-face presence—but that is changing with the growth of online commerce: Some of those jobs will migrate. In addition, more than 70 million other jobs are in services of different kinds, and some of these services can be delivered electronically, either now or in the near future—education and health (17 million jobs); professional and business services (17 million); leisure and hospitality (12 million); finance (8 million); wholesale trade (6 million); transportation (4 million); and so on.

Looking in turn at each sector, Blinder makes some educated guesses about how many jobs might be movable—not now, with technology in its present state, but over the next decade or two, assuming further improvements in information technology and telecommunications.

Most of the services provided by the health sector need to be delivered in person—this applies to the "vast majority," says Blinder. But, as he notes, there are exceptions. Radiology is one. Lab work can be outsourced, too. And there may be more scope for IT-enabled health care than even Blinder supposes: Remote diagnostics and monitoring of chronic patients over the Internet is an infant industry, to be sure, but more than a mere gleam in an entrepreneur's eye. Companies like Health Hero Network are doing it.

Parts of the education sector are at risk as well. Personal interaction is everything, no doubt, so far as K-12 is concerned, but is it necessary in colleges? Reading about the travails of Lawrence Summers at Harvard, one is reminded that personal interaction with academics (as opposed to graduate-student teaching assistants) is something that undergraduates get rather little of in any case—despite the colossal fees. As for professional and business services such as accountancy, it may be that many of these jobs can be performed remotely. The same goes for banking (where offshoring is already well advanced) and insurance. And so on.

Adding it all up, Blinder concludes that "the total number of U.S. service-sector jobs that will be susceptible to offshoring in the electronic future is two to three times the total number of current manufacturing jobs"—in other words, between 28 million and 42 million jobs, or 10 or more times the Forrester estimate (remember, though, that this was for the next decade only). And in some ways the picture is more dramatic even than this. Blinder points out that the kind of face-to-face services that are likely to remain in the United States will probably experience rising relative prices, because they tend to be resistant to increases in productivity (think of haircuts, or bus drivers). This, in turn, implies falling relative demand—and, as jobs migrate, other things equal, falling real wages for the workers concerned.

The danger, as Blinder rightly argues, is not rising unemployment. People will find work. That happened during the first Industrial Revolution (farming to manufacturing) and the second as well (manufacturing to services). Tens of millions of jobs were lost; tens of millions were created. The American labor market is superbly efficient at moving people out of old jobs and into new ones. (It will be a different story in Europe.) The problem is that the adjustment will be painful. Blinder therefore wants to see a much larger commitment to trade-adjustment assistance. He also seems to favor a thicker and stronger social safety net: unemployment insurance, public health insurance, and so on. (There, I hesitate: Europe's safety net is surely one of the reasons why its labor market is so bad at creating new jobs.)

What else might be done? The constant refrain from orthodox economists is "education" (I plead guilty, again). Blinder raises doubts about this as well. The United States is going to retain lots of low-skill jobs; it is already apparent that high-skill jobs are not at all immune to offshoring. "People skills may become more valuable than computer skills. The geeks may not inherit the earth after all—at least not the highly paid geeks in the rich countries." In other words, while it is true that education may need to be rethought for the third Industrial Revolution, that task is going to be a bit more complicated than parroting such easy and uncontroversial nostrums as, "We need more scientists and engineers."

Blinder is not claiming to have the answers. And, as I say, he is fundamentally an optimist: The first two Industrial Revolutions were wrenching, all right, but who would wish to turn the clock back? But Blinder has convinced me, at least, that this is going to be a much bigger deal than most economists have so far recognized.