April 10, 2017 12:15PM

Wage Effects of Immigration Are Small

Immigration has small long-run relative wage impacts on American workers by education (Figure 1). These estimates are the most popular and widely cited in the immigration debate. They were completed by George Borjas and Gianmarco Ottaviano and Giovanni Peri. Their findings are very close but diverge most appreciably for the wages of dropouts, even though the effect is small and positive for all native-born workers lumped together. According to the 2015 American Community Survey, 9.4 percent of native-born Americans over the age of 25 are dropouts. Thus, over 90 percent of American workers are in education-skill categories where immigration increased relative wage, according to the most negative finding (Figure 1).

Figure 1

Relative Impact of Immigration on Native Wages by Education

Media Name: immwages1.png

 

Sources: Borjas, p. 120; Ottaviano & Peri, Table 6.

Note: Borjas looks at 1990-2010. Ottaviano and Peri look at 1990-2006.

Borjas and Ottaviano and Peri find that the wages of immigrant workers are most affected by new immigrants (Figure 2). That’s because new immigrants have skills and education levels most similar to previous immigrants, so they compete against each other more than with natives who have very different levels of skill and education. As we point out in Figure 25 of this bulletin, immigrants still support liberalized immigration despite the negative wage effects they experience. There are at least three explanations for this.

Figure 2

Relative Impact of Immigration on Immigrant Wages by Education

Media Name: immwages2.png

 

Sources: Borjas, p. 120; Ottaviano & Peri, Table 6.

Note: Borjas looks at 1990-2010. Ottaviano and Peri look at 1990-2006.



First, most of these immigrants want to bring over their family members, so they support expanded immigration even if they know they will face more wage competition. People are willing to pay a lot to have their families nearby. Second, wage competition does not generally cause anti-immigration opinions. As Jens Hainmueller and Daniel J. Hopkins wrote in their wonderful literature review of the literature on opinions toward immigration: “As an explanation of mass attitudes toward immigration, the labor market competition hypothesis has repeatedly failed to find empirical support, making it something of a zombie theory.” Immigration attitudes show little evidence of being strongly correlated with personal economic circumstances. Instead, people are most concerned with immigration's nation-wide impact on many issues.

Third, the wage gains from immigrating are so relatively gargantuan that the small single-digit decline is swamped. Research on the wage premium by Michael Clemens, Claudio E. Montenegro, and Lant Pritchett illustrate this point. A 35-year-old Mexican-born male urban worker with nine years of education can expect a monthly wage 2.53 times as large just by immigrating to the United States. In other words, his monthly wage income rises from $580.90 to $1,470.80 by immigrating. Without immigrants during the time period, as measured by Borjas and Ottaviano and Peri, his monthly wage income would have instead been $1,541.40 or $1,548.75, respectively, or 2.65 to 2.67 times as great as in Mexico. That’s a maximum difference of about $78 a month.

The Borjas and Ottaviano and Peri findings are at ends of the academic literature as most findings are even closer to zero. These differences are small for natives and, with the exception of dropouts, all point in the same direction by skill-education. Borjas and Ottaviano and Peri reach slightly different conclusions because they measure the immigrant impact on population differently. Borjas measures the immigrant impact by their fraction of the population by skill-education level while Ottaviano and Peri measure it by looking at the inflow of the share of immigrants in the labor force by skill-education level compared to the previous Census.

Regardless, Borjas and Ottaviano and Peri agree on at least two points. First, immigrants raise relative native-born American wages overall by +0.6 percent. Second, immigrant workers compete with other immigrant workers and lower their wages in every education group reported. Native-born American workers do not face much wage competition with immigrants. 

January 13, 2017 5:20PM

Further Thoughts on Mariel Boatlift

Jason Richwine just blogged about my recent Mariel Boatlift post that confirmed George Borjas’ finding of wage increases for those with only a high school degree in post-Mariel Miami.  George Borjas understood my quick extension of his research.  Below are some of Richwine’s points and my quick responses. 

“The point is not especially interesting, since the standard immigration narrative has always been that efficiency gains come at the expense of the natives with whom immigrants most directly compete – high school dropouts, in the case of Mariel.”

It’s important to identify which skill-group of Miamians could have benefited from the Boatlift.  George Borjas pointed out in his report for the Center for Immigration Studies:  “Economic theory predicts that immigration will redistribute income by lowering the wages of competing American workers and increasing the wages of complementary American workers as well as profits for business owners and other “users” of immigrant labor.”  Borjas focused on the benefits for business owners and other “users” of immigrant labor in that paper. 

Although he argues that dropouts and workers with only a high school degree are not substitutes, he doesn’t provide evidence of potential complementarity.  In at least the Mariel case, there is some evidence of that.  My post shifts the narrative from “only businesses and the rich gain from low-skilled immigration” to “the real beneficiaries could be a much larger pool of workers who actually bothered to finish high school.”  That matters. 

Furthermore, if “efficiency gains come at the expense of natives with whom immigrants most directly compete,” then who gains in Richwine’s analysis?  The cross-skill wage elasticities for those above a high school degree are not statistically significant.  They are for workers with only a high school degree according to Table 4 in this paper by Borjas and Joan Monras.  My Mariel post is perfectly consistent with those findings. 

“But it would have been simple to determine the HS-and-below impact directly from the Borjas and Monras paper: Just take a weighted average of the dropout and graduate wage effects reported in their Table 4.  Doing that yields a negative number that does not reach statistical significance.”

I didn’t work backward from Borjas and Monras because they ended their analysis in 1984.  Using 1984 in my post would have unfairly biased my results against Borjas’ findings in his Mariel paper.  I chose 1986 because wages for dropouts were at their nadir in that year and to be consistent with Borjas’ Mariel paper. 

Furthermore, Richwine’s point that combining wages for dropouts and high school only workers yields a result that is statistically insignificant shows yet again that high school graduate wages increased to outweigh the negative wage effects on dropouts.  I don’t think Richwine intended to neutralize Borjas’ findings in exactly that way.    

“Overall, Nowrasteh frames his argument as the following: Borjas may or may not be right about Mariel lowering the wages of dropouts, but if he is, then we must also conclude that Mariel raised the wages of dropouts and high-school-only natives put together. The argument doesn't work. Across several different methodological scenarios that do not alter Borjas's conclusion, Nowrasteh's numbers – both his own and those of economists who previously studied the same question – do not tell a consistent story. His claim that accepting Borjas requires accepting that HS-and-below natives benefited from Mariel is therefore unconvincing.”

Those who swooned over Professor Borjas’ Mariel paper have consistently failed to note the increase in wages for high school graduates that Borjas and Monras discovered.  To quote their paper:

“As before, the estimated own wage effect is negative and significant, with a wage elasticity of about -0.9. Similarly, the estimate of the own employment effect is not distinguishable from zero.  The analysis, however, shows that the cross effects are numerically important.  Although the supply shock of the predominantly low-skill Marielitos lowered the wage of high school dropouts, it raised the wage of workers with a high school education, and this effect is both numerically and statistically significant. The cross-wage elasticity is about +0.7.  In addition, the unemployment rate of workers with more than a high school diploma also fell significantly [Emphasis added].”

That such dramatically different findings can result from such minor changes in methodology weakens all of the research on this topic – including mine.  Reading the work by Card, Peri and Yasenov, Borjas, and Borjas and Monras, and then conducting my own replication and extension of their results has significantly weakened my confidence in any of these findings.  For instance, weekly earnings for Hispanic dropouts between the ages of 25 and 59 increased in Miami shortly after the Boatlift (Figure 1).  Hispanic dropouts should be the most substitutable for Marielitos but they’re not according to Borjas’ methods.  This graph would well fit into David Card’s research. 

Figure 1

Weekly Earnings for Hispanic Dropouts Age 25-59

 

 

Media Name: french.png

It’s baffling how anybody can replicate these findings, look at these graphs, and maintain their confidence in this research. 

January 4, 2017 1:24PM

Rebuttal of Senator Tom Cotton’s Anti‐​Legal‐​Immigration Op‐​ed

Senator Tom Cotton (R-AR) recently penned an op-ed for the New York Times in which he calls for a large reduction in legal immigration, something he believes will raise American wages. It’s nice when immigration restrictionists are honest about their intention to cut legal immigration, but Senator Cotton would be disappointed if his policy ever came to fruition. Senator Cotton does make some cursory arguments for expanding high-skilled immigration—a positive policy—but I will focus here on his argument to restrict it. I will respond to a few of Senator Cotton’s comments below. His will be in block quotes while my responses will follow. 

Higher wages, better benefits and more security for American workers are features, not bugs, of sound immigration reform. For too long, our immigration policy has skewed toward the interests of the wealthy and powerful: Employers get cheaper labor, and professionals get cheaper personal services like housekeeping. We now need an immigration policy that focuses less on the most powerful and more on everyone else.

Senator Cotton argues that skilled native workers are complementary to low-skilled immigrants, meaning that the former's wages rise rather than fall when more of the latter arrive. This is because low-skilled immigrants and higher skilled workers don’t compete for the same jobs but instead work together, expanding productivity and compensation for both parties. These complementarities do exist, but there is also much evidence that lower-skilled American workers are actually complementary with low-skilled immigrants. Economists Gianmarco Ottaviano and Giovanni Peri found that immigration had a small positive relative effect on the wages of native workers with no high school degree (between +0.6 percent and +1.7 percent) and a small positive effect on average native wages (+0.6%) from 1990 to 2006. Immigrants are complementary to native workers but substitutable for other immigrants who experienced a substantial relative negative effect (−6.7 percent) from immigration. It should not be surprising that new immigrants compete with older immigrants who both share similar skills while native-born American workers benefit overall.



Language differences are a major reason why immigrants and natives with the same skill level are complementary according to economists Peri and Chad Sparber. Low-skilled immigration incentivizes low-skilled natives to specialize in jobs that require communication in English. Meanwhile, immigrants specialize in jobs that are more manual-labor intensive and require less English-language proficiency. Communication jobs are more highly compensated than manual-labor jobs. This complementary task specialization reduces the downward wage pressure because natives react by adapting and specializing in more highly paid occupations, not by dropping out of the job market or accepting lower wages. This effect decreases wage competition between lower-skilled natives and immigrants by around 75 percent. Economist Peter Henry also found that low-skilled immigrants to an area induced natives to improve their school performance so that they wouldn’t have to compete with lower-skilled immigrants. Immigrants tend to push Americans upward rather than downward.

More low-skilled workers incentivize more Americans to enter the labor market. A good example of this is provided by economists Patricia Cortes and Jose Tessada, who found that skilled American women with young children reentered the workforce faster when they lived in cities with many low-skilled immigrants who could work as nannies. In this case, an immigrant with a job increases the number of working skilled American women. 

After all, the law of supply and demand is not magically suspended in the labor market. As immigrant labor has flooded the country, working-class wages have collapsed. Wages for Americans with only high school diplomas have declined by 2 percent since the late 1970s, and for those who didn’t finish high school, they have declined by nearly 20 percent, according to Economic Policy Institute figures.

Senator Cotton rightly tells us to pay attention to the law of supply and demand but then promptly ignores demand. Immigrants boost demand by buying goods and services, which create more jobs than are occupied by the immigrant workers themselves according to research by Gihoon Hong and John McLaren. If immigrants are removed from an area they take their purchasing power, and the jobs that their purchases support, with them. An example will help explain this. Let’s say you own a small local business. One day the government rounds up half of your customers, some of whom are your employees, and deports them. As a business owner, would you be eager to hire more workers at a higher wage to replace those who were deported or would you refrain because your revenues are about to take a substantial hit? The answer is obvious. Now multiply that by 11 million unauthorized immigrant consumers.

Some will claim that money sent abroad by immigrants in the form of remittances, about $135 billion in 2015, does nothing to help the U.S. economy. First off, most of that money wouldn’t have been made had it not been for immigrants earning it—the economy is not a fixed pie. Secondly, that money eventually returns to the United States in the form of exports or foreign investment.

There is not a fixed number of jobs in the economy, so an employed immigrant does not automatically force a native out of the job market. Immigrants “taking” jobs from natives, also known as displacement, is a minor phenomenon when it exists at all. One prominent study found that an increase in the foreigner share of a population by 10 percent reduces native employment by 0.2 to 0.7 percent, a result exacerbated by labor market regulations. Many of those Americans displaced by immigrants tend to get better-paid jobs that exist due to the complementarities described above. David Card and Ethan Lewis found that most of the workers displaced were actually immigrants themselves, although this was confined to just a handful of cities.

The wage decline figures provided by the left-wing Economic Policy Institute have been soundly rebutted when proper inflation indices are used.  A forthcoming Mercatus Center paper by Scott Winship, a visiting fellow at the Foundation for Research on Equal Opportunity, finds that wages for workers in the 20th percentile who are most likely to be only high school graduates have increased slightly since 1970. Wages for workers with less than a high school degree have likely fallen somewhat since 1970.

However, the U.S. workforce is a lot more educated than it was in 1970. In that year, there were 47.2 million native-born American high school dropouts who were 25 years or older. They comprised 46.4 percent of the native population in that age category. In 2015, there were only 16.8 million native high school dropouts in the same age range and they comprised a mere 9.4 percent of the natives. Both the absolute and percentage of native-born Americans with less than a high school degree has crashed since 1970.

The number of natives who are at least 25 years old with only a high school degree increased from 43.5 million in 1970 to 51.1 million in 2015, but their percentage of the population dropped from 42.8 to 28.6 percent. Native high school dropouts and those with only a high school degree numbered almost 91 million in 1970 and comprised 89 percent of natives who were at least 25 years old. In 2015, they numbered 67.9 million and a mere 38 percent of natives in the same age range. Even if Senator Cotton’s figures were correct (they aren’t) the percentage and number of native-born Americans who suffer is much reduced from 45 years ago.

Interestingly, workers with only a high school degree might be the most complementary to those with less than a high school degree. There is some evidence that Miamians with less than a high school degree suffered wage declines after the Mariel Boaltift dropped about 125,000 mostly low-skilled Cubans in Miami in 1980. However, the wages for native-born Miamians with only a high school degree shot up afterward and likely overwhelmed the wage decline for those with less than a high school degree. If these results are generalizable (a huge if) then halting immigrants with less than a high school degree could stop wage growth for high school graduates. That’s disincentives climbing the skills ladder.

It’s been a quarter-century since Congress substantially reformed the immigration system. In that time, the population of people who are in this country illegally has nearly tripled, to more than 11 million. We’ve also accepted one million legal immigrants annually — and a vast majority are unskilled or low-skilled.

If controlling immigration to the United States was as easy as flipping a policy switch, then there would be no debate over immigration reform. There would be no illegal immigrants and the system would behave as its designers intended. In the real world, people respond to incentives and are not passive objects that just accept government laws. When laws, like our heavily restrictive immigration laws, erect legal barriers to voluntary exchange then the result is a vast black market represented by 11 to 12 million illegal immigrants. More restrictions are unlikely to reduce the size of the black market. Assume, for the sake of argument, that Senator Cotton is correct that wages for Americans with less than a high school degree will rise if millions of illegal immigrants are deported and the future legal flow is halted. Foreigners will then have even more of an incentive to sneak in illegally to work, overstay visas, or find other ways to circumvent American labor market regulations. More comprehensive enforcement will just raise wages, which will attract more illegal immigrants, who will then lower wages again, which in turn will be countered by better enforcement—and so on in a familiar and predictable cycle. All is not hopeless, however: there is a way out.

The only times in American history when our immigration laws were largely obeyed was when the Great Depression turned off the “jobs magnet,” a world war prevented the crossing of borders, and a large-scale guest-worker program funneled would-be illegal immigrants into the legal system (the Bracero program). Since 1964, we have not had a Great Depression (thank God), world war (double-thank God), or a functional guest-worker visa program for lower-skilled workers. As a result, we have a large problem with illegal immigration that enforcement cannot halt except by triggering a world war or economic depression. Foreigners want to sell their labor to Americans and Americans want to buy it. The law does dent that flow but there will always be a large black market so long as the government tries to enforce laws that conflict so much with reality. A functional legal immigration system can prevent or substantially reduce illegal immigration far more cheaply and effectively than expanding an already vast border bureaucracy.

Immigration produces a net-benefit for Americans. Economist David Card called research on the topic “the elusive search for negative wage impacts of immigration.” Although some noted economists like George Borjas at Harvard do find relative wage declines for some groups of American workers, even his work shows that wages for native-born Americans benefit overall, although by a small amount. An honest discussion of immigration policy must consider the known economic benefits of immigration as well as the costs.

January 2, 2017 4:12PM

The Mariel Boatlift Raised the Wages of Low‐​Skilled Miamians

Harvard economist George Borjas recently published an important paper on how the unexpected surge of 125,000 Cubans (henceforth Marielitos) to Miami in 1980 lowered the wages of native-born male Miamians with less than a high-school degree. Because at least 60 percent of the Marielitos were high school dropouts, Borjas found that the negative wage effects were concentrated on Miamians with the same level of education.  

There are excellent criticisms of Borjas’ paper that show his results hinge on the control cities he chose, his exclusion of women, the age group of the workers, whether Hispanics are included, whether high-school-or-less or no-high-school-at-all are included, and whether datasets with the larger samples are used. For the sake of argument, supposing that Borjas made the correct methodological choices on every single point above, the Mariel Boatlift still raised the wages for low-skilled U.S. workers collectively due to wage complementarities. That’s because native-born Miamians with only a high school degree (no associate degree, no education after high school) experienced significant wage increases immediately after Mariel relative to workers with the same levels of education in the control groups, or placebos, of other cities. Borjas’ supporters ignore this finding but he does not. 

In his Mariel paper, Borjas reports the wage of high school dropouts relative to high school graduates in Figure 3(C) and the wage of Miami high school graduates across an all cities permutation in Figure 4(B), but he doesn’t have a dramatic graph like this that shows what happened to the relative wages of high school graduates after Mariel. 

Another working paper by Borjas and Monras on the wage effects of refugees also found that “the rate of wage growth for high school graduates, a group whose size was only increased modestly by the Marielitos, is noticeably higher in Miami than outside Miami.”  They go on to write that, “the predominantly low-skill Marielitos . . . raised the wage of workers with a high school education, and this effect is both numerically and statistically significant. The cross-wage elasticity is about +0.7 [compared to -0.9 for high school dropouts].”  They do not find any employment effects for high school dropouts but they did uncover positive and statistically significant employment gains for those with a high school degree. Furthermore, Figure 7.5 on page 148 of Borjas’ new book We Wanted Workers hints at a wage increase for high school graduates immediately after Mariel.              

My intern Cole Blondin and I followed Borjas’ methods to create graphs for the wages of high school graduates before and after the Boatlift. We used the March Current Population survey (March CPS) and combined the May Current Population Survey and the May Outgoing Rotation Group (May CPS-ORG) datasets. The only differences are that we present the figures in dollars rather than logs, we did not use three-year averages to smooth the data, and we did not recreate the synthetic control. One final note, the wage effect of the Marielitos must be compared to placebos because there was only one Miami in 1980 and we can’t actually observe what would have happened to that city had the Marielitos not arrived. We used the same sets of placebo cities as Borjas.

Even under Borjas’ assumptions, native-born male Miamians with a high school degree or less saw a ­net-wage increase after the Mariel Boatlift.    

March CPS 

Just as Borjas reported, there is a significant drop in the wages for male native-born Miamians with less than a high school degree who were 25 to 59 years old after 1980 relative to workers with the same characteristics in the placebos (Figure 1).  By contrast, the wages for Miamians with only a high school degree also increased after the Mariel Boatlift (Figure 2). The wage changes in both figures are statistically significant relative to the placebos. Miamians with just a high school degree are complementary to high school dropouts. Figure 3 shows the wages for high school dropouts and high school graduates in Miami.       

Figure 1

High School Dropouts, March CPS

Media Name: borjas1.png

 

Figure 2

High School Only Graduates, March CPS

Media Name: borjas2.png

 

Figure 3

High School Dropout and High School Only Graduate Wages 1977-2002, March CPS

Media Name: borjas3.png

 

Relative wages for high school dropouts reached their nadir in 1985 and 1986 while the wages for workers with only a high school degree reached their peak in 1984. A mere 19 percent of Miami workers in the March CPS sample had less than a high school degree but 36 percent had only a high school degree which means that more low-skilled American workers experienced a wage gain than a wage decline after Mariel.  

We compare the pre- and post-Mariel Miami wages, weighted by the relative size of the native educational cohorts in the March CPS, to produce a rough estimate of Mariel’s net impact on native wages.  For instance, if there are 10 high school dropouts who each lose $1 and 20 high school only graduates who each gained $0.60, then the net wages earned by Americans increased by $2 [(20*$0.60)-(10*$1)]. In such a scenario, total wages rise for Americans who are high school dropouts and high school graduates. 

We compared the post-Mariel period of 1981 to 1986 to the pre-Mariel period of 1976-1979. The total wage gain for high school graduates outweighs the total losses for high school dropouts in the “All Cities” placebo (the placebo that produced Borjas’ stunning graph). The Card and Low-Skill placebos also returned slightly net wage losses while the employment placebo returned a deeply negative result (Table 1).  The average for all the placebos from the March CPS data was -5.85.       

Table 1

Net Wage Effects, March CPS

         
March CPS Differences in Differences Changes (1976-1979, 1981-1986)
  Card Cities Employment Cities Low-Skill Cities All Cities
No HS, Wage Changes Times Pop

-9.7804


-21.2793


-10.0078


-9.41663

HS Only, Wage Changes Times Pop

5.471363


4.892769


6.571003


10.13576

Net effect

-4.30904


-16.3865


-3.4368


0.71913

May CPS-ORG

Professor Borjas also used the May CPS-ORG data to test whether Mariel lowered the wages of similarly-skilled Miamians. The May CPS-ORG dataset returned a much smaller wage decrease for high school dropouts in Borjas’ paper. This is important by itself because the May CPS-ORG dataset is superior for two widely-reported reasons: First, the May CPS-ORG contains fewer errors because it asks respondents about their wages last week rather than last year (which is the case for the March CPS). Second, it has a larger sample size than the CPS in every year from 1979 onward.  

Regardless, we confirmed Professor Borjas’ findings that the wages of male high school dropouts in the relevant age range fell after the Mariel Boatlift, although they fell less than in the March CPS data (Figure 4). We also found that the wages for high school graduates in the May CPS-ORG dataset increased relative to the placebo cities after the Mariel Boatlift (Figure 5).

Figure 4

High School Dropouts, May CPS-ORG

Media Name: borjas4.png

 

Figure 5

High School Only Graduates, May CPS-ORG

Media Name: borjas5.png

 

The wage gains for high school graduates outweigh the losses of male high school dropouts in Miami for each placebo group—using the March CPS weight of the population by education (Table 2). The Card Cities, Low-Skill Cities, and All Cities placebos all returned very positive results.  The Employment Cities placebo was the only one that returned a slightly negative result. The average net gain for high school graduates and dropouts across the placebos is a positive +17.43. 

Table 2

Net Wage Effects, May CPS-ORG

May CPS-ORG Differences in Differences Changes (1976-1979, 1981-1986)
  Card Cities Employment Cities Low-Skill Cities All Cities
No HS, Wage Changes Times Pop

-2.20452


-8.33419


-1.67613


-1.91032

HS Only, Wage Changes Times Pop

28.30903


7.618387


25.00548


22.89419

Net effect

26.10452


-0.71581


23.32935


20.98387

To further check our results, we also weighted the wage changes by the May CPS-ORG survey’s estimate of Miami’s education level. The May CPS-ORG survey found that the share of Miami’s native population without a high school education is higher and the percentage with only a high school degree is lower than in the March CPS. Using the May CPS-ORG education weights returns a reduced net positive wage impact for the Card, Low-Skill, and All Cities groups while the Employment Cities group is more negative. Using the May CPS-ORG education estimates, the average is still a positive +10.09.    

Evaluating the Complementary Effects from the March CPS and May CPS-ORG

We ran Borjas’ four different sets of placebo cities on two educational groups of male workers in two different datasets—the March CPS and the May CPS-ORG. There were eight final estimates of the net effects, four in the March CPS and four in the May CPS-ORG.  Four of the eight showed a positive net-wage impact – three in the May CPS-ORG and one in the March CPS.  Four of the eight also showed a negative net-wage impact—three in the March CPS and one in the May CPS-ORG. For each placebo, the positive net-wage gains in the May CPS-ORG results were far larger than the estimated loss in either the May CPS-ORG or the March CPS.

Conclusion

The Mariel Boatlift provides a wonderful natural experiment to test how a sudden, exogenous surge of immigrants affects wages. Professor Borjas’ examination of how Marielitos substituted for native-born Miamians with the same level of education is an important component of that story. However, most have ignored the important complementary effects of the Marielitos on the wages of workers with only a high school degree. A full understanding of Professor Borjas’ contributions to this subject requires acknowledging these complementary effects. To borrow the language used by Professor Borjas, the Marielitos’ redistributed wages from dropouts to workers with only a high school degree with a net positive effect on all low-skill workers.  

Special thanks to Cole Blondin for his excellent work on this blog post.  

March 28, 2016 3:36PM

The Hidden Costs of ObamaCare’s Millennial Mandate

Guests mingle during the second annual Future of America gala at the House of Sweden in Washington, D.C., U.S., on Friday, Oct. 3, 2014. "Our guests are not people that are traditionally struggling with unemployment," said David Pattinson, founder of David Pattinson's American Future, a Washington-based nonprofit, which aims to get the millennial generation more fully employed. Photographer: T.J. Kirkpatrick/Bloomberg

There is a current running through the ObamaCare debate that goes something like this:

Every other advanced country provides health insurance to all its citizens for a fraction of what Americans spend on health care. ObamaCare emulates what those countries do. Anyone who complains about ObamaCare increasing premiums or imposing other costs is therefore a right-wing nut who doesn’t understand that universal coverage results in lower spending, not higher spending.

This line of reasoning, so to speak, leads supporters to believe ObamaCare is a free lunch. Their ignorance is not accidental. MIT health economist and ObamaCare architect Jonathan Gruber helpfully explained some years ago that he and his co-architects deliberately designed the law to hide its costs and make the benefits seem like a free lunch.

ObamaCare’s “Millennial mandate”—the requirement that employers who offer health coverage for employees’ dependents continue to offer such coverage until the dependents turn 26 years old—is one of those supposed free lunches. This mandate’s benefits unquestionably come at a cost. Expanding health insurance coverage among adults age 19-26 leads them to consume more medical care. When those people file insurance claims, health-insurance premiums rise. Yet ObamaCare does an amazing job of hiding those costs from voters.

Does ObamaCare impose a special tax that the IRS collects to pay for that extra coverage? No. That would be far too transparent. The cost just gets added to your premiums.

Does ObamaCare require employers to include a line-item on your premium payments, to show you how much this additional coverage is costing you? Absolutely not. That, too, would make the costs dangerously noticeable. The additional cost just gets thrown onto the pile, hidden among the costs of all the other mandated coverage you don’t want, and the coverage you actually do want.

Maybe workers see their premiums rising, and are merely ignorant of the fact that the Millennial mandate is part of the reason? Nope. ObamaCare hides the cost further still. Explaining how requires a little bit of labor economics.

Workers pay about one-quarter of their health premiums themselves. Employers pay the rest. But employers get the money to pay that three-quarters share of the premium by taking money out of other types of worker compensation. So when your employer pays $13,000 toward your health premiums, it is you—not your employer—who bears that cost, because you otherwise would have gotten that $13,000 in salary or other benefits. Workers lose control over that money without ever knowing it belongs to them.

Workers, therefore, pay 100 percent of the cost of their health benefits, which means they pay 100 percent of the cost of the Millennial mandate as well. But you don’t realize you bear those costs because you never see your employer taking money away from other forms of compensation. In fact, since your employer is writing the check, you probably think it’s your employer’s money rather than yours. You probably don’t know—because, importantly, you have no way of knowing—that the Millennial mandate is preventing you from getting a raise this year. Or that the Millennial mandate led your employer to jettison other health benefits you value more, like lower deductibles or broader physician and hospital networks. And even if you notice those changes, you are more likely to blame their employer than ObamaCare.

ObamaCare’s Millennial mandate may be the perfect crime. Politicians hand out benefits, but no one can tell who’s paying. Workers don’t see a special tax. They don’t see a special premium surcharge. At best, they see only a fraction of the increase in their premiums. They may not see their premiums rise at all. And if they feel any harm, they are likely to blame someone other than the politicians who enacted ObamaCare.

How can voters possibly weigh whether the Millennial mandate’s benefits justify the costs when ObamaCare so systematically keeps them in the dark? They can’t. And, as Jonathan Gruber let slip, that is by design.

Thank God, as always, for economists. Some of them are actually trying to reveal the costs and benefits of this mandate.

  • Stanford University’s Jay Bhattacharya and his coauthors estimate the mandate has reduced wages for workers in affected firms by $1,200 per year. That’s about $6,000 per affected worker since the mandate took full effect in 2011. Note that all workers subject to the mandate saw this reduction in wages, not just workers with dependents.
  • Asako Moriya of the federal Agency for Health Care Research & Quality and her coauthors find evidence that adults age 19-26 used more inpatient care and mental-health services, and that those who were hospitalized were more likely to be insured.
  • Gregory Colman (Pace University) and Dhaval Dave (Bentley University) find evidence that the mandate: induced adults age 19-26 to reduce their labor supply; resulted in those adults spending less time waiting for care; induced them to spend more time socializing; and improved these Millennials’ subjective well-being.

Thanks to such scholars—and only thanks to them—we can have a reasonably informed debate about whether the benefits of this mandate are worth the costs. When it comes to ObamaCare, having information about costs and benefits is the exception rather than the rule.

Even so, don’t expect that information or that debate to do a whole lot of good. The political system doesn’t much care about weighing benefits against costs. Politicians obey whoever shouts the loudest.

In this case, that would be the small number of employers, insurers, health care providers, and (to a lesser extent) Millennials who get relatively large benefits in the form of a competitive advantage or other indirect subsidies. The political system will obey them rather than the much larger population of workers, employers, and others whom this mandate harms.

Why? Because even though it is larger, that group doesn’t shout as loudly because it is much harder to organize. And one of the reasons they are so hard to organize is, as Jonathan Gruber reminds us, ObamaCare’s authors designed the law to make sure that its victims don’t even realize they are victims. That’s why ObamaCare's Millennial mandate may survive even if its costs vastly exceed the benefits.

Note: I will be hosting a discussion on ObamaCare’s Millennial mandate featuring Prof. Jay Bhattacharya (Stanford) and Asako Moriya (AHRQ) at the Cato Institute in Washington, D.C., this Wednesday, March 30, from 12-1:30 pm ET. Register to attend the event here. If you can’t make it to the event, you can watch it live online at www.cato.org/liveFollow the discussion and ask questions on Twitter at #MillennialMandate.

March 10, 2015 11:13AM

CIS’ All Job Growth Since 2000 Went to Immigrants’ Report Is Flawed

The Center for Immigration Studies (CIS) has released a number of reports purporting to show that all employment growth since the year 2000 has gone to immigrants. The CIS report does not include econometrics. However, the report includes a few references to the economic literature (those few references present have little to do with native job displacement caused by immigration, which is the topic of the CIS report). Nonetheless, the CIS report has gained significant attention.

The CIS method of measuring job displacement caused by immigration is not used by professional economists to study this issue. Fundamentally, CIS assumes a static number of jobs that is unchanging based on immigration and does not consider what the job market would look like with fewer immigrant workers, entrepreneurs, and consumers—estimates essential for understanding the actual labor market impact of immigrants.  I discuss those actual effects here, here, and here

Regardless of their flawed methods, I decided to recreate CIS’s research in order to exactly understand how they got their results.

The study did not find any evidence of immigrants pushing natives out of the job market. After spending hours recreating their data and checking it, all I can conclude is that immigrants hold about a percentage of jobs in the economy that is roughly equal to their percent of the population. I am underwhelmed by that finding. 

Below I will present the academic literature on immigration-induced job displacement, explain how CIS got its results, and detail why its analysis of the data does not prove that “All Job Growth Since 2000 Went to Immigrants.” (If you just want the meat, scroll down to the hed "CIS’s Three Big Conclusions Are False").

Brief Literature Survey

The academic literature finds very little native job displacement caused by immigration. 

David Card and John DiNardo looked at native responses to immigration in American cities to test the so-called “skating rink” model of native location decisions, a model that assumes each new immigrant knocks an American out of the workforce. If the skating rink model is correct, natives with skills similar to immigrants should vacate areas where immigrants move to and not move to areas where immigrants are residing. Instead, Card and DiNardo found that an increases of the immigrant population in specific skill groups leads to small increases in the population of native-born individuals in the same skill group. Changes in the local economy, such as the creation of new businesses and types of industries affected by an increase in immigrants, made up for any displacement of native workers. 

Another paper by Card did not find any offsetting native mobility responses to immigrants in the same skill level but did find negative wage effects for some skill sets as a result of immigration. However, he also found that in the very short run at least, inflows “of new immigrants in the 1985-90 period reduced the relative employment rates of natives and earlier immigrants in laborer and low-skilled service occupations by up to 1 percentage point, and by up to 3 percentage points in very high-immigrant cities like Los Angeles or Miami” (emphasis added). 

Card’s findings in his second paper are consistent with the later findings of Gianmarco Ottaviano and Giovanni Peri that newer immigrants compete with the immigrants who preceded them, not with native-born Americans who have similar skills. The labor market effects of new immigrants appear to fall most heavily on immigrants who preceded them, not Americans, which would seem to cut against the theory that immigrants have a large negative effect on American workers. 

Even then, Card and Ethan Lewis in another paper looked specifically at how new Mexican immigrants displaced older Mexican immigrants and found decidedly small effects. Only in Los Angeles and El Paso, TX did new Mexicans push out older Mexicans.  In all of the other cities they examined, new Mexican immigrants complemented the existing Mexican immigrant workforce rather than displaced it. To a remarkable degree, the U.S. economy is very good at attracting Mexican immigrants, providing incentives for them to settle in areas where they are most demanded, and responding in ways that increase net production and employment.

A paper by George Borjas seems to find the greatest effect of immigration on the wages of native-born American workers, a wage elasticity of –0.39. Borjas’s finding has been criticized by many, including in this recent paper that extended his period of analysis by 10 years but found only a –0.2 wage elasticity as well as other potential problems with his methods. Another paper by Peri and Chad Sparber also questioned Borjas’s paper, finding that less educated immigrant workers and native born workers specialize in different tasks, thus inducing natives to reallocate their task supply, thereby reducing downward wage pressure. Foreign born workers specialized in occupations that required manual labor and physical skills while natives pursued jobs that required more intensive communication and language skills. Immigration induces natives to specialize accordingly, reducing the negative wage effect of immigration by roughly 75 percent. In other words, natives do not react to immigration by leaving the workforce or moving to different areas, but by changing their skill sets and occupations. 

A paper by Christopher L. Smith found that for every 10 percent rise in employed immigrants with at least a high school degree, high school students worked about 3 percent fewer hours and adults decreased their number of hours worked by 1 percent. A Chicago Fed Letter and research paper authored by Daniel Aaronson, Kyung-Hong Park, and Daniel Sullivan discovered that teens are less likely to work for reasons other than immigration. They found that an increase in the relative benefits of education versus work, government financial incentives for schooling, merit-based scholarships with minimum grade requirements, and education grants were the primary causes of a decline in teen labor force participation. In other words, teens allocated their scarce time to education and away from work to increase their investment in acquiring human capital and, hence, a higher future income. Low-skilled immigration and stiffer labor market competition was not a compelling explanation for their decline in labor force participation. Another report by the U.S. Bureau of Labor Statistics in 2002 echoed the findings of Aaronson, Park, and Sullivan when it concluded:

In summary, the increasing proportion of teens enrolled in school during the summer and a drop in students’ labor force participation rates contributed to the overall decline in teen summer labor force participation during the recent expansion. Data for October each year indicate that labor force participation among high school students also dropped during the school year, although nonstudents were increasingly likely to participate in the labor force. Together, these facts suggest that, among teens, an increased emphasis was placed on school rather than work during the summer and school year.

Patricia Cortes does find some displacement effects across cities. These effects are not large enough to equalize wages across the country, and thus not large enough to induce the displacement of one American worker for each immigrant worker. Cortes found that three natives move out of a city for every 10 immigrants who move in. 

These academic papers do not produce a compelling reason to believe that immigrants displace native-born workers in large numbers. There is likely some job displacement caused by immigration, but the effect is small and does not produce a scenario where all job growth goes to immigrants.  

Reproducing CIS’s Study

I reproduced CIS’s entire study and data set to properly understand how they reached their conclusions. 

Data and Methodology

The data for this investigation were collected through the U.S. Census Bureau’s Consumer Population Survey, and then further transformed using the DataFerrett mining program. Monthly data on employment, citizenship, and population are available through these sources for the period 1994–2014. For some points of analysis, I chose a start year of 1994 as opposed to CIS’s start date of 2000 because that year's very low historical unemployment rate added bias to CIS's results. 

Findings

Each subsection below details the findings. 

Immigrant Share of Employment

There is a proportional relationship between growth in the immigrant share of the total population and the immigrant share of employment. Unsurprisingly, as immigrants grow as a proportion of the population, they occupy more jobs as a percentage of total employment. The immigrant share of employment has increased over the period 1994–2014, but this increase corresponds with an increase in the share of immigrants as a percentage of the population. The proportionality between the two variables was confirmed by a statistical test, which showed a correlation between the immigrant share of employment and the immigrant share of the population of 96.4 percent. Over the entire 20-year period (1994–2014), the immigrant share of employment has grown a paltry 2.53 percent.

Extending this analysis back to 1994 revealed an interesting pattern. Since 2000, immigrant share of employment has been slightly higher by less than a percentage point than the immigrant share of the population would predict. Prior to 2000, the immigrant share of the population was higher than their respective share of jobs. In that sense, during the period 1994–2000, the percentage of jobs given to immigrants was lower than the respective percentage of the population. It is only after 2000 that the percentage of immigrants in the population is less than the immigrant share of employment—again by less than a percentage point. This could be another reason why CIS picked the year 2000 as the start date for its study.      

The Effect of Immigrant Employment on Native Employment

Immigration restrictionists claim that an increase in employment among immigrants has a detrimental effect on native employment. The CIS study did not produce an economic model, any econometrics, or a single regression to support that claim. Instead, I ran a few regressions to see if there is any relationship between immigrant job gains and native job losses.  

There is an 86.75 percent correlation between native employment and immigrant employment. A regression was then performed on the data to discern the relationship between the variation in native employment and non-citizen employment, which resulted in a statistically significant positive relationship between the variables. The variation in native employment represents the change in the number of jobs held by natives, while the variation in immigrant employment represents the change in the number of immigrant jobs. According to the results of the regression, an increase in immigrant employment by 1 is correlated with an increase in native employment by approximately 1.614. Basically, every additional job performed by an immigrant is associated with more than one additional native working. This relationship indicates that increases in immigrant employment are not damaging to natives but instead could produce a net benefit. Correlation does not prove causation in this situation, but the preliminary findings of this regression show that immigrant job gains are not correlated with net-job losses for natives. It is likely that both immigrants and natives get jobs when the job market is growing—hardly a controversial result.

Additionally, a negative linear relationship was found between the native share of employment and the foreign-born share of employment (Figure 1). The negative slope indicates that an increase in the share of immigrants with jobs correlate with a decrease in the share of natives with jobs. Although this may seem as though immigrant employment growth is injurious to native employment growth, this is not the case when compared with the immigrant share of population (Figure 2). If immigrants are a larger share of the population, we would expect them to occupy a larger share of jobs. Since there is no fixed number of jobs in the economy, this finding does not suggest that immigrants take jobs from natives.   

Figures 1 and 2 are almost identical. Taken together, the figures demonstrate the proportionality between the share of the immigration population and the share of jobs occupied by immigrants. CIS’s main finding appears to be that the immigrant share of all jobs is about the same as their share of the entire U.S. population. 

Figure 1: Immigrant and Native Shares of Total Employment, 1994–2014

Media Name: nowrastehcis_1.jpg

 

Figure 2: Immigrant and Native Shares of Total Population, 1994–2014

 

Media Name: nowrastehcis2.jpg

 

Employment and Unemployment Rates by Nativity

Employment and unemployment rates are not significantly different between native and immigrant workers. The unemployment rate for immigrant workers age 16–65 is higher than that for native workers of the same age (Figure 3). The employment rate for immigrant workers age 16–65 is lower than for native workers of the same age (Figure 4). The findings show that, on average, an immigrant worker is less likely to be employed than a native worker, and this trend has held true for the 16–65 age bracket from 1994 to 2000.

CIS limited its analysis to natives and immigrants in the 16–65 age range. However, when all persons over age 16 are analyzed, the employment rate for immigrants is slightly higher than the employment rate for natives (Figure 5). Despite the lower employment rate for natives in this case, the percentage difference between the two groups is small; for example, in December 2014, immigrants were only 4.6 percent more likely to be employed than natives.  

Looking at the over-65 age bracket for the years 1994–2000, the native employment rate was higher than the immigrant employment rate (Figure 8). The switchover occurred only in the years immediately after 2000. Moreover, the recent employment rate difference seemingly favorable to immigrants can be explained by the demographic differences between native and immigrant populations. The employment rate for workers over 65 is lower than that of other age groups, and immigrant populations tend to be younger than 65.  

 

Figure 3: Unemployment Rate by Nativity, 16–65

Media Name: nowrastehcis3.jpg

 

Figure 4: Employment Rate by Nativity, 16–65

 

Media Name: nowrastehcis4.jpg

 

Figure 5: Employment Rate by Nativity, All Ages

 

Media Name: nowrastehcis5.jpg

 

Figure 6: Unemployment Rate by Nativity All Foreign Born, 16–65

Media Name: nowrastehcis6.jpg

 

Figure 7: Employment Rate by Nativity All Foreign Born, 16–65

 

Media Name: nowrastehcis7.jpg

 

Figure 8: Employment Rate by Nativity All Foreign Born, All Ages

Media Name: nowrastehcis8.jpg

For those in the 16–65 age bracket, an increase in the unemployment rate of all foreign born by 1 corresponds with an increase in the unemployment rate of natives by 0.799. The unemployment rates move in the same direction for both natives and the foreign born (Figure 6). An increase in the employment rate of all foreign born by 1 corresponds with an increase in the employment rate of natives by 0.345, again meaning that the two rates move in the same direction. An increase in the employment rate for all foreign-born by 1 corresponds with an increase in the employment rate for natives by 0.279, again meaning that they move in the same direction. If immigrants took the jobs of natives, we would not find that their levels of employment or unemployment move in the same direction.  

Different Time Periods and Populations

Below I make two major changes to augment the CIS results. First, I compare CIS’s time period of 2000–2014 to the period of 1994-2000. The year 2000 is a problematic year to begin this analysis because the unemployment rate was abnormally low at 4 percent, the lowest over the 1994–2014 period when the CPS data are available and the lowest since 1969 when it stood at 3.5 percent. Picking a year like 2000 with such a low unemployment rate will make subsequent years look bad by comparison. Using 1994 as a start date for this analysis makes more sense because the unemployment rate in that year was 6.1 percent compared to 6.2 percent in 2014. Using start and end years that have similar unemployment rates should allow an analysis to more easily judge the effect of immigration on native employment. 

Second, I include all workers who are 16 years old or older. CIS excluded workers over the age of 65 who are now more likely to work than they were in the past. To CIS’s credit, footnote 5 in their paper admits this omission but very few people reporting on these finding bothered to read the fine print—especially when the title of the paper is “All Employment Growth Since 2000 Went to Immigrants” (emphasis added). CIS’ exclusion of the 65+ portion of the American workforce biased their results against natives. 

The percentage of total jobs in the economy occupied by natives was 88.48 percent in 2000 and 83.25 percent in 2014. For immigrants in 2000 and 2014 it was 12.17 percent and 16.90 percent, respectively. That is a very small shift in the relative, not absolute, employment of these two big employment groups in the U.S. economy that is roughly parallel to their increase in the population. When compared with each group’s share of the population, native-born Americans and immigrants maintain roughly proportional growth in shares of employment and population throughout the period (as shown in Figures 1 and 2).

Furthermore, the relative share of employment for the two groups reveals only a small shift in employment. The native share of employment went from 90.03 percent in January 1994 to 83.25 percent in 2014 (Figure 1), and the immigrant share of employment grew from 9.97 percent to 16.90 percent during the same time period. Again, the change in the native and immigrant shares of employment corresponds with proportional changes in each group’s share of the population. 

The number of people older than 16 increased by 53.1 million between January 1994 and December 2014. Some 63.86 percent of this increase was native-born Americans, while 36.14 percent went to immigrants. Over that period, 53.3 percent of net job growth went to natives while 47.7 percent to immigrants.   

CIS’s Three Big Conclusions are False

I list CIS’s conclusions, numbered and in italics, below. My responses follow.

1. The long-term decline in employment for natives is a clear indication that there is no general labor shortage.

There is no evidence for a long-term decline in the number of natives employed. The data indicate that between January 2000 and December 2014, there were two recession years (2009, 2010) that exhibited a significant decline in jobs. However, all other years saw increasing native employment. Between January 2000 and December 2014, 5.19 million additional natives held jobs. There is, however, strong evidence of declining labor force participation. Between January 2000 and December 2014, native labor force participation saw a fairly steady decline from 64.01 percent to 58.46 percent, which was true both for recession and non-recession years. So there isn’t a shortage of jobs, but there is a decreasing percentage of natives who want or are able to work. CIS did not demonstrate that immigrants are the cause of that. 

Some 14.4 million more natives held jobs in December 2014 than held jobs in January 1994. From January 1994 to December 2014, native labor force participation steadily declined from 61.54 percent to 58.46 percent.

2. The decline in work among natives over the last 14 years of high immigration is consistent with research showing that immigration reduces employment for natives.

The academic research about immigrant displacement of native-born workers is thin and much of it demonstrates the opposite (see “Brief Literature Survey” above). Interestingly, Robert Hall found much of the decrease in the U.S. labor force participation rate (LFPR) occurred for members of wealthier households while the LFPR increased for individuals in poorer households since 1999. This directly contradicts CIS’s story that immigrant job competition is the cause of the decline in native employment opportunities, and Hall's findings are consistent with the findings of one of the few academic papers actually cited by CIS. Lower-skilled immigrants can only really compete for jobs with young, uneducated, and low-skilled native-born workers (when there is any competition at all). That younger, uneducated, and low-skilled native-workers in low-skilled households have increased their LFPRs in recent decades is evidence that little competition or job displacement even occurred. 

However, Salim Furth of the Heritage Foundation has challenged many of Hall’s findings. Both Furth and Hall have found that teenagers in the lower half of the family income distribution have seen smaller declines in labor force participation compared to teenagers in the upper half of the family income distribution, excluding the effects of the Great Recession. Furth rightly notes that more research is needed on this issue.      

The CIS report suggests that 2000–2014 was a period of high immigration, although the Census reveals that immigration during that period had slowed down compared to the previous decade. Between 1990 and 2000, the immigrant population grew from 19.8 million to 31.1 million, an 11.3 million person increase that grew the size of the immigrant population by 57 percent. From 2000 to 2010, the immigrant population grew from 31.1 million to about 40 million, an 8.9 million person increase that grew the size of the population by 29 percent—half of the previous decade. The increases in the absolute number and relative percentage of immigrants in the first 10 years of the millennium were smaller than in the last decade of the 20th century. 

From the beginning of the Great Recession in 2007 to 2010, the immigrant population increased by 3.2 million. From 2010 to 2013 the immigrant population increased by only 1.3 million—far below the average for any three-year period in recent decades. The growth in the immigrant population is slowing dramatically but the job market is still poor. Unsurprisingly, immigration is not an entirely exogenous shock; rather, it ebbs and flows based on American demand for immigrant workers. 

3. Trends since 2000 challenge the argument that immigration on balance increases job opportunities for natives. Over 17 million immigrants have arrived in the last 14 years, and native employment has declined.

CIS didn’t run a regression for the period 2000-2014, but if it had it would have found that an increase in noncitizen employment by 1 correlates with an increase in native employment by 0.696. That is not the result that we would expect if CIS’s conclusion were correct. CIS did not identify an actual correlation between native job losses and immigrant employment gains. There is nothing in CIS’s report that demonstrates that immigration causes native unemployment.    

Conclusion

Immigrants' share of jobs is very similar to their share of the U.S. population. That is not a surprising finding and it certainly doesn’t show that they are “taking all the jobs.” On a bigger level, labor markets in every country are changing.  Labor’s share of GDP is declining even in countries that accept very few immigrants, like Japan. This trend will not change by radically decreasing the number or share of immigrants.  

Special thanks to Kristina Pepe. Her quantitative skills contributed mightily to this blog post.

January 17, 2014 12:32PM

Damning Trade with Faint Praise

A Washington Post editorial today pushes back against the argument that a Trans-Pacific Partnership agreement would exacerbate income inequality. Amen, I suppose. But in making its case, the editorial burns the village to save it by conceding as fact certain destructive myths that undergird broad skepticism about trade and unify its opponents.

“All else being equal,” the editorial reads, “firms move where labor is cheapest.”  Presumably, by “all else being equal,” the editorial board means: if the quality of the factors of production were the same; if skill sets were identical; if workers were endowed with the same capital; if all production locations had equal access to ports and rail; if the proximity of large markets and other nodes in the supply chain were the same; if institutions supporting the rule of law were comparably rigorous or lax; if the risks of asset expropriation were the same; if regulations and taxes were identical; and so on, the final determinant in the production location decision would be the cost of labor. Fair enough. That untestable premise may be correct.

But back in reality, none of those conditions is equal. And what do we see? We see investment flowing (sometimes in the form of “firms mov[ing],” but more often in the form of firms supplementing domestic activities) to rich countries, not poor. In this recent study, I reported statistics from the Bureau of Economic Analysis revealing that:

Nearly three quarters of the $5.2 trillion stock of U.S.-owned direct investment abroad is concentrated in Europe, Canada, Japan, Australia, and Singapore. Contrary to persistent rumors, only 1.3 percent of the value of U.S.-outward FDI [foreign direct investment] was in China at the end of 2011.

Meanwhile, the United States (not China or Mexico) is the world’s #1 destination for FDI:

With a stock valued at $3.9 trillion, the United States is the top single-country destination for the world’s FDI outflows. There are plenty of reasons for that being the case, including the facts that the United States is the world’s largest market and has a sound legal system and a relatively transparent business environment. More than $4 out of every $5 of that stock (84.2%) is owned by Europeans, Canadians, and Japanese, with the U.S. manufacturing sector accounting for a full third of its value, making it the primary destination for inward FDI.

Are BASF, Michelin, BMW, Siemens, Airbus, InBev, Honda, Kia, Ikea, Shuanghui (recent Chinese purchaser of Smithfield Hams) and thousands of other foreign-headquartered companies invested here because labor is cheapest in the United States? They are here because firms conduct value-added activities wherever it makes the most sense to do so, given all of the considerations and restrictions that affect costs. For so many reasons, the United States is still the top destination for investment in manufacturing and most services industries. 

So stop. Just stop.

The editorial also indulges the most persistent myth of all, that increasing exports while minimizing imports is the purpose of trade agreements. As I’ve written on countless occasions in numerous different ways, increased imports are the real benefits of trade. Our exports are what we use to pay for our imports. If you prefer paying less for your products at the grocery store, you should prefer exporting less for the products you import. And for those concerned about income inequality—the editorial’s presumed audience—it is worth understanding that import competition increases choices and reduces prices, which means imports increase real incomes. 

The editorial concedes that imports from Vietnam may increase under the TPP (“suppose [the bilateral deficit] were to double”), but that the deficit “would still be tiny relative to the overall U.S. trade balance.” Instead of apologizing and rationalizing, as though this development were a cost, why not point out how lower-income Americans, in particular, would experience a lower cost of living because trade would reduce the cost of their clothing and footwear?

We need to do better a job explaining how trade does not lend itself to sports metaphors. Exports are not our "points." Imports are not "their" points. The trade account is not a scoreboard. It is not Team America against the world. Trade is about mutually beneficial exchange between individuals in different political jurisdictions, and to the extent that those kinds of transactions are subject to the whims of politicians, more and more resources will be diverted from economic to political ends.

Though it may have had good intentions, the Washington Post should know better than to perpetuate simplistic myths spun by well-compensated K Street consultants on behalf of the business, labor, and environmental interests that benefit financially from restrictions on trade and investment.