My friend Tom Palmer sent to me today this study done by the Coalition for a Prosperous America. It seems to be authored by one Jeff Ferry, and its bottom line is that, at least in recent years, every $1 billion increase in goods imports into America causes a loss of 4,552 American jobs. Pretty scary stuff, we are to conclude.
Recovering as I am from recent surgery, I have more time to waste than usual, so I spent a couple of hours today trying to make sense of this study. It contains some glaring errors – such as when its author writes that
In general, imports are a straight substitute for domestic production. For example, for each dollar spent on a vehicle or a computer produced (or partially produced) in a foreign country, there is a domestic alternative.
Both common sense and economic theory tell us that this assertion is mistaken. The global economy features distinct patterns of specialization, and these patterns aren’t random: people in each country specialize in those production activities for which they have a comparative advantage. Countries with people who have a comparative advantage at producing lumber tend to export lumber and not import lumber. These ‘lumber’ countries import non-lumber goods and services that the people there cannot produce at a comparative advantage.
Of course, these patterns of specialization and trade are never perfect – that is, these patterns are never in perfect equilibrium. And even if they were today in perfect equilibrium, not only would this ‘perfect’ pattern change over time, there would at any moment in time still be a great deal of overlap between the comparative advantages of people in country A with those of people in some other countries, such that producers of (say) lumber in country X will find it profitable to export to country Y while producers of lumber in country Y will find it profitable to export to country X. Nevertheless, it’s plainly mistaken to assume (as the author of this study seems to assume) that all goods and services, or even most goods and services, that Americans import compete directly with currently on-going American production.
(For a more detailed and elaborate exploration of this point, see, for example, Lawrence Edwards’s and Robert Z. Lawrence’s excellent 2010 paper “Do Developed and Developing Countries Compete Head to Head in High-Tech?” The answer, as Edwards’s and Lawrence’s empirical analysis shows, is clearly no – as we would expect. Quoting Edwards and Lawrence (page 8):
We will show in the paper that there are distinctive patterns of international specialization that suggests developed and developing countries produce fundamentally different products. Judged by export shares, the US and developing countries specialize in quite different product categories that for the most part do not overlap. Moreover, even when they do overlap and exports are classified in the same category, there are large and systematic differences in unit values that suggest the products made by developed and developing countries are not very close substitute – developed countries produce are far more sophisticated.
So there.)
But beyond identifying this error (and few other obvious ones), I’ve failed to make sense of this CPA paper. It reads, to me, like gibberish. Perhaps my medication has me loopy, or perhaps I’m simply a dunce. What, for example, is to be made of the author’s first table (pictured just below)?
The author claims that the figure of 4,522 job losses per one-billion-dollar-increase in American imports is shown in the figure. And, indeed, I see the number “4,552” in the southeastern-most part of the figure, in the bottom line titled “Total.” But how is it derived? From where does this figure come? Unlike the other numbers on that bottom line, 4,552 is clearly not the sum of the numbers in the column above it. How does the author come by this figure? I’ve read this piece at least a dozen times in the past couple of hours and have no idea where this number comes from.
Anyway, rather than wrestle further with this confusing essay, I am taking the lazy way out and asking you, dear Cafe patron, to help me. What do you make of this CPA essay? How is the 4,552 figure derived? You may e-mail me, you may put your answer in the comments section, or both.
