Tables 16 and 17 provide percent change in employment estimates for the United States by sector,
under each of the six scenarios. It should be noted that the model assumes no change in the overall
level of the US workforce or employment; the results displayed in tables 16 and 17 simply show the
redistribution of employment across sectors, always assuming full employment. For both unskilled (table
16) and skilled workers (table 17), the biggest changes come in the wearing apparel and gas sectors under
the NTB reversion scenario, with both increasing roughly 15 percent. The largest negative impact for
unskilled workers comes in the transportation cost scenario, where the reversion to 1980 transportation
cost levels would trim employment in the wheat sector by 10 percent. The motor vehicles sector would
see skilled and unskilled employment increases under every scenario except for the repeal of US FTAs
(Scenario 4) and the reversion to circa 1990 applied tariffs (Scenario 6).
Benefits of Trade Expansion
The cost to US and world real income under each scenario using Gilbert’s CGE modeling is divided
into efficiency and terms-of-trade effects. The efficiency effect is akin to comparative advantage gains or
losses—in other words, the cost of reallocating resources to less productive sectors. Importantly, the CGE
model does not encompass other channels by which trade increases real income—notably productivity
gains by US firms when they face import competition, the spreading of overhead costs when US firms sell
more in export markets, and greater variety gains for US consumers.
A negative impact on real income, by way of efficiency loss, occurs in each of the six scenarios. The
terms-of-trade effect—changes in the price of exports relative to the price of imports—can be a benefit
or loss to a country. For large countries higher tariffs can sometimes result in real income gains via the
terms-of-trade effect, because they pay less for imports and get more for exports. In the six scenarios only
the NTB reversion and transportation cost reversion scenarios generate positive terms-of-trade effects for
US real income. Even in these cases, the negative impact on US real income via the efficiency effect results
in an overall net loss for the United States.
Tables 18 and 19 summarize the real income effects for the six scenarios. Surprisingly, among the
six scenarios, the reversion in transportation costs to 1980 levels is estimated to be the most costly for the
United States, even though this scenario had the smallest trade impact. The net negative impact amounts
to around a $21 billion loss in annual real income (measured in 2004 dollars). The disparity between a
large real income loss and a small trade impact reflects the fact that declining transportation costs generate
a productivity shock to the whole economic system.
The scenario with the smallest impact on US real income is the NTB reversion, cutting only $5
billion annually. However, this small result reflects the combination of a large positive terms-of-trade
effect ($25 billion) and a large negative efficiency effect ($30 billion). We place more stock in estimates of
efficiency gains and losses, so the small net estimate reported for the NTB scenario may be misleading.
12