>But the problem with the "churn" comparison is that assumably almost all of
>those losing jobs in one place gain them in the new jobs. So the 310,000
>job losses per year are concrete jobs lost to the economy as a whole.
Huh? That makes no sense at all. The "churn" numbers (see 'em for yourself at <http://www.bls.gov/bdm/home.htm>) measure gross job losses and gross job gains. Gross gains less gross losses = net gains or losses, which are roughly comparable to the monthly establishment survey gains or losses in employment. The 310,000 losses would go in the gross loss column. If the job losers are re-employed, that would go into the gross job gains column. The numbers are quite comparable.
>The other problem with these numbers is that by multiplying the quarterly
>numbers by four is deceptive, since it assumes that the same person is not
>losing jobs multiple times, but the reality is that the "churn" sector of
>jobs involve many workers who are continually moving between such temporary
>jobs. It's a reasonable bet that the 7 million jobs lost each quarter often
>involve less and possibly far less than 7 million people, so actual PEOPLE
>losing jobs is far less than numbers of jobs lost. It's a bit like the
>statistic that one out of two marriages end in divorce; that may be true but
>a far smaller percentage of people are ever divorced, since serial divorce
>participants make up a disproportionate share of divorces.
>
>So it is far more likely than 2% that a person losing their job is losing it
>to overseas competition.
So it's 4% then? That would be way too high, since I doubt that many people lose two jobs in a year. But let's grant it - then what about the other 96%? Where do they come from?
Glancing at Kletzer's book (downloadable at <http://bookstore.iie.com/merchant.mvc?Screen=PROD&Product_Code=110>), I see she also points out that plenty of manufacturing job loss in industries without significant import competition. From 1979-99, the annual risk of job loss in a high-import-competing sector was 5.9%; in low-import sector, 4.2%.
>And then of course, the real effect of import competition is not necessarily
>job loss, but the threat of job loss. Most workers being rational and
>recognizing the threat accept lower and lower pay in such sectors in order
>to stem job loss. So even where jobs are not lost to import competition,
>its existence plays a very large role in wage loss.
"Very large"? Then how do you explain a job loss risk in low-import-competing industries that's 71% the risk of that in high-import-competing industries?
This is important for at least two reasons, aside from the importance of separating truth and falsehood. One, if the risk from trade isn't as high as people like you paint it, then the working class is more scared than it needs to be. And two, if trade is really the culprit, then it encourages dangerous ideas like banning imports from China, a move that would almost certainly result in millions of job losses around the world. If it's not, then more sensible policies can be pursued. As long as trade is fingered as the culprit, our crummy support for displaced workers loses political salience. And it gets unions off the hook for their failure to organize the service sector.
Doug