well named
poorly undertitled
- Oct 2, 2018
- 432
- 84
- 80
From a new working paper from the National Bureau of Economic Research (full text here). Some interesting data, seemed worth sharing.
Background:
They also considered some counterfactual scenarios extrapolated from the data:
Background:
The border wall expansion we study was a result of the Secure Fence Act of 2006, which authorized the construction of reinforced fencing in California, Arizona, New Mexico, and Texas. Between 2007 and 2010, 548 miles of wall were constructed along the 1,954-mile U.S.-Mexico border, bringing the total fencing to 658 miles. Since unauthorized migrants typically cross the border by foot (Massey, Durand, and Malone (2003)), the extension of the wall altered the relative costs of migration across origin and destination pairs. By combining this geographic variation with a confidential version of the Matrícula Consular (Consular ID card) database that provides detailed information on the bilateral flows of (primarily unauthorized) Mexican migrants between 3,000 different regions in the United States and Mexico, we examine how the patterns of migration changed after the expansion of the border wall.
On the impact of the new fencing on migration:
...we find that the border wall expansion caused a decline in migration flows: a 10% increase in the total travel time necessary to avoid the border wall resulted in a 1.4% reduction in bilateral migration flows. This result is robust to accounting for the response of border patrol enforcement, controlling for different types of border walls, and instrumenting the location of the wall expansion using geographic predictors of where the wall was built.
On the economic impact:
This decline reduced the welfare of both low-skill and high- skill Mexican workers by an equivalent decline in per capita income of $1.34 and $2.99, respectively (driven primarily by the direct increase in the cost of migrating) and reduced the welfare of high-skill U.S. workers by $4.35 (driven primarily by the decline in relative scarcity of high-skill labor). However, low-skill U.S. workers benefited by $0.36, as low-skill labor in the United States became more scarce. These figures do not include the direct cost of wall construction, which is approximately $7 per person (not worker) in the United States.
So, we spent $7 per person to increase low-skilled laborers wages by $0.36, at least using their estimates. Doesn't sound too great.
They also considered some counterfactual scenarios extrapolated from the data:
First, we consider experiments that “fill in” some of the gaps in the wall to understand if our small effects are driven by the fact that the wall only partially covers the U.S.-Mexico border. We find no evidence of such nonlinearities: filling in half of the remaining gaps on the border would reduce the number of Mexican migrants by 144,000 yet increase the economic benefit to only $0.58 per low-skill U.S. worker.
Second, we consider an experiment that reduces the international trade costs between the United States and Mexico. We find that, like the border wall expansion, this reduction in trade costs also reduced the number of Mexican workers in the United States. For example, a trade policy that reduced the impact of distance on international trade flows one-quarter of the way toward the impact of distance on domestic trade flows would have reduced the number of Mexican workers residing in the United States by about 123,000. However, unlike the border wall expansion, reducing trade costs results in large economic benefits for both U.S. and Mexican workers. For example, a 25% reduction in the additional international cost of distance would would yield a benefit of equivalent to a $59 increase in income for each low-skill U.S. worker and even greater gains for low-skill Mexican workers and high-skill workers of both nationalities.
Certainly makes trade policy seem way more effective than wall building, both in terms of reducing net immigration inflows and in terms of supporting American workers. There's a lot more detail on the methods and data in the linked PDF.Second, we consider an experiment that reduces the international trade costs between the United States and Mexico. We find that, like the border wall expansion, this reduction in trade costs also reduced the number of Mexican workers in the United States. For example, a trade policy that reduced the impact of distance on international trade flows one-quarter of the way toward the impact of distance on domestic trade flows would have reduced the number of Mexican workers residing in the United States by about 123,000. However, unlike the border wall expansion, reducing trade costs results in large economic benefits for both U.S. and Mexican workers. For example, a 25% reduction in the additional international cost of distance would would yield a benefit of equivalent to a $59 increase in income for each low-skill U.S. worker and even greater gains for low-skill Mexican workers and high-skill workers of both nationalities.