There is a symposium on Should We Still Make Things? at Dissent Magazine. Here's part of Dean Baker's entry:
Should We Still Make Things?, by Dean Baker: I have often thought that economists should be required to have a better grasp of simple arithmetic. It would prevent them from repeating many silly comments that pass for conventional wisdom, such as that the United States will no longer be a manufacturing country in the future.
Those who know arithmetic can quickly detect the absurdity of this assertion. The implication of course is that the United States will import nearly all of its manufactured goods. The problem is that unless we can find some country that will give us manufactured goods for free forever, we have to find some mechanism to pay for our imports.
The end of manufacturing school argues that we will pay by exporting services. This is where arithmetic is so useful. The volume of U.S. trade in goods is approximately three and half times the volume of its trade in services. If the deficit in goods trade were to continue to expand, we would need an incredible growth rate in both the volume and surplus of service trade and our surplus on this trade in order to get to anything close to balanced trade.
For example, if we lose half of our manufacturing over the next twenty years, and imported services continue to rise at the same pace as the past decade, then we would have to see exports of services rise at an average annual rate of almost 15 percent over the next two decades if we are to have balanced trade in the year 2028. ... It would take a very creative story to explain how we can anticipate the doubling of the growth rate of service exports on a sustained basis. ...
[Also], the idea that U.S. workers are somehow too educated to be doing for manufacturing work, but instead will be making the beds, bussing the tables, and cleaning hotel toilets for foreign tourists is a bit laughable. Of course, with the right institutional structure (e.g. strong unions) these jobs can be well-paying jobs, but it is certainly not apparent that they require more skills than manufacturing. ...
In short, the idea that the United States can survive without manufacturing is implausible: It implies an absurdly rapid rate of growth of service exports for which there is no historical precedent. Many economists and economic pundits asserted that house prices could keep rising forever in spite of the blatant absurdity of this position. The claim that the U.S. economy can be sustained without a sizable manufacturing sector is an equally absurd proposition.
I thought that if you looked at the value of US manufacturing, it hasn't fallen nearly as much as manufacturing employment. Thus, much of the change that has affected workers is due to changes in technology, not the exporting of jobs (this comes from a study done by the Peterson Institutute, more here). But from a worker's perspective, it doesn't matter all that much whether it's technology or jobs moving to other countries, the job is gone either way. The key, then, is to have good jobs waiting for workers when they are displaced due to inevitable (and desirable) technological change or to jobs moving overseas, jobs that are every bit as good or better than the jobs they left. That is where we are falling short. The new jobs we are creating are not as good as the jobs we are losing, when workers are forced to find new jobs they don't tend to do as well as they did in their previous job, and that is the source some of the stagnation we have seen in middle class incomes over the last few decades.