Mitchell’s Musings 5-8-2017: Your Final Exam Question

05 May 2017 1:04 PM | Daniel Mitchell (Administrator)

Mitchell’s Musings 5-8-2017: Your Final Exam Question

Daniel J.B. Mitchell

I recently placed online the PowerPoint slides (in pdf format) and other materials for courses I used to teach at UCLA in their final formats, i.e., as of 2008 when I retired.[1] One of them was an introductory course in international economics. From time to time in that course’s final examination, I included a recent newspaper article that related to the course’s topics and asked the students for a critique. As it happened, at about the time I was uploading the course files, an article appeared in the Los Angeles Times that would have been a good candidate for such an exam essay, were I still teaching the class.[2]

There are various themes in the article such as whether President Trump overestimates the bargaining leverage that running a trade deficit with a particular country gives him. But the article also goes into whether improving the trade deficit would create more jobs, particularly in manufacturing. In prior musings, we have noted that bringing the trade balance to zero would not come close to restoring the proportion of manufacturing workers in the labor force to what it was in the 1950s. But we also noted that given the current shrunken share of the workforce in manufacturing, about one in ten U.S. jobs, moving to balanced trade would have a significant impact on that sector.

If I had assigned the article as a final exam essay question, I would have hoped that the students would have directed their attention to the chart it contained below:



The chart represents an effort to show that changes in the trade balance don’t correlate positively with changes in employment in manufacturing. But here’s the problem. The trade balance as a time series tends to be negatively correlated with the business cycle. That is, during booms the trade balance tends to become more negative. During busts it improves. Manufacturing is highly cyclical, but in a positive direction. That is, other things equal, Good Times in the general economy tend to be Good Times in manufacturing (and for manufacturing jobs). Hard Times are especially hard in manufacturing. So in a time-series sense, an improvement in the trade balance is likely to be negatively correlated with jobs in manufacturing.



There is a simple explanation for this seeming paradox. Real imports are strongly correlated with real GDP. When times are good, there is more consumption, more investment, etc. Some of what is consumed and invested comes from abroad (the demand link). Some of what is produced domestically in the U.S. uses inputs, directly or indirectly, from abroad (components, fuel, etc.; the supply link).  Indeed, a hypothetical student taking my final exam might have recalled the slide from the class above showing that a one percent change in real GDP is associated with about a two percent change in real imports (e=2 on the slide).[3]

Imports are only half the trade balance story. What about the export side? We can crudely think of imports as our demand for goods and services from them (the rest of the world). We can think of exports as their demand for goods and services from us. So, since our exports mainly reflect conditions abroad, they are going to be less associated with our business cycle than are imports. Put another way, the import side tends to dominate in determining the movement of our trade balance in response to our business cycle.[4]

The bottom line here is that the Los Angeles Times article’s use of a time-series chart involving manufacturing jobs and the U.S. trade balance – while the numbers underlying the chart are correct – is highly misleading. The chart is dominated by the business cycle. Consider the two most recent recessions. You can see the trade balance improving during the dot-com bust of the early 2000s (and the simultaneous recession-induced fall in manufacturing jobs). The same effect is apparent from the chart during the Great Recession of 2008-09. All the chart really tells you is that manufacturing is hurt by recessions and helped by booms, and that the trade balance is helped (improved) by recessions and hurt by booms. The chart tells you nothing about the connection between improving the trade balance and adding more jobs in manufacturing.

You’ll likely get an “A” on the final exam’s essay if you knew that already.


[1] Management 205a - International Economics, available at

[2] “Why Trump's obsession with trade deficits is misguided,” available at

[3] I re-estimated the relationship for this musing, updating the data through 2016 and got the same e=2 (not surprisingly since I was only adding a few years to a long time series). The course slide reproduced in the text is at

[4] There is some positive correlation between our business cycle and the business cycle(s) abroad, partly because our imports are their exports. So you might expect real exports to show some positive correlation with real GDP through that channel and because they are a component of real GDP. But a similar estimation with regard to exports produces e<1. And since we start from the fact that imports>exports, the trade balance of the U.S. reflects the U.S. business cycle.

Employment Policy Research Network (A member-driven project of the Labor and Employment Relations Association)

121 Labor and Employment Relations Bldg.


121 LER Building

504 East Armory Ave.

Champaign, IL 61820


The EPRN began with generous grants from the Rockefeller, Russell Sage, and Ewing Marion Kauffman Foundations


Powered by Wild Apricot Membership Software