A Shortage of Low-Wage Workers?? That’s Not the Right Defense of Immigration Reform

June 14th, 2013 at 9:24 am

Readers know I’m a supporter of immigration reform for many reasons, some of which go beyond economics into the realm of America as a welcoming country for those seeking opportunities.  There’s no doubt in my mind that both my own life and that of our nation has been enriched by those who have left their homes to come to America in search of a better life.

But when we defend reform, I think we have a responsibility to stick to the facts as best we can, and I thought this piece went far beyond that threshold, particularly in claiming a shortage of low-wage workers (h/t: KR in comments):

In 1950, according to the Census Bureau, 56% of U.S. workers were high-school dropouts. Today, the figure is less than 5%.

The result is that the pool of people available to fill low-skilled jobs has shrunk dramatically.

The argument that we have a shortage of high-skilled workers in this country is dicey, but there’s a case to be made (a weak case, but that’s a different discussion).  But I know of no credible arguments that we have a shortage of low-skilled workers (obviously, we’re not talking about right now, when shortages are clearly on the demand side–not enough job slots).  If their wage and employment trends over the past thirty years doesn’t convince you that there’s no supply shortage in the low-wage sector (here’s the wage evidence; for jobs evidence, see the unemployment rates of the least skilled/educated—it’s consistently way above the average), then you’re playing with a very different set of cards than the rest of us.

These facts have led some to worry—reasonably—about the impact of immigration reform on the wages of less advantaged domestic workers.  That’s been carefully examined and here, in contrast the illogic of the WSJ piece, is what the research shows:

–Most contemporary research finds that immigrants don’t place significant downward pressure on the wages of domestic workers because they’re more often complements than substitutes.  But, and here’s where the WSJ is especially off, when they are substitutes (i.e., domestic workers or recent immigrants with low skills) the wage effects from immigrant competition are significant and negative.

–We have historically absorbed large immigrant flows in ways that have been positive for them and for the economy.  First, the flows are small relative to the size of the overall labor market, and second, demand curves also move out.  IE, an exceptional period to the negative wage trends for low-wage workers was the latter 1990s, when immigrant flows were very strong (and welfare reform was also driving up the supply of low-wage work), yet demand for low-wage work grew faster than supply.

–Bringing undocumented workers who are already here “out of the shadows” does not push out the labor supply curve–they’re already here–and can only help dampen unfair competition.

In the end, as I stressed before on these pages, I think much of what’s in the Senate plan on low-wage immigration makes sense (W-visas—see previous link).  The fact that this part of the bill is supported by folks like EPI, an institution with tremendous credibility on these low-wage issues, former Labor Sec’y Ray Marshall, who understands both the research and the on-the-ground reality of low-wage competition, and the AFL-CIO also should lead one to believe we’ve got the balance generally right on this.

Of course, then there’s the House…

Print Friendly, PDF & Email

8 comments in reply to "A Shortage of Low-Wage Workers?? That’s Not the Right Defense of Immigration Reform"

  1. smith says:

    The statement “Most contemporary research finds that immigrants don’t place significant downward pressure on the wages of domestic workers because they’re more often complements than substitutes.” is misleading since the leading figure in the field disputes those findings.


    Saying “We have historically absorbed large immigrant flows in ways that have been positive for them and for the economy.” is also misleading since the previous great wave of immigrants at the turn of century could fight for labor rights without fear of deportation. 11 million undocumented workers waiting 13 years for naturalization aren’t going to risk arrest and deportation from labor actions. Neither are those on work visas (low and high skill) able to make any demands since dismissal means deportation.

    You don’t need fancy equations* to understand reform is just corporate welfare aided by sellout unions and politicians looking for votes.

    True reform is immigration without requiring employers sponsorship, same as we had 100 years ago, which led to a robust labor movement.
    http://en.wikipedia.org/wiki/Constant_elasticity_of_substitution *(CES)

    • Kevin Rica says:


      Where the “like” button on this contraption?

    • Procopius says:

      Just by chance I’m re-reading John Dos Passos’s “The Big Money” the third book of his U.S.A. Trilogy. If you read the preceding two books as well, I don’t think you’d feel that the current labor relations situation is all that different. Before 1933 the number of deportations was not insignificant, especially after passage of the Espionage and Sedition Acts in 1918. Living conditions of the newly arrived immigrants were terrible, unless they were lucky enough to have connections through relatives who had already established themselves. Incidentally, life was much less economically secure for most people back then. I was born at the tail end of the Depression and reading the headlines from the turn of the century through the twenties brings back memories from the bad old days.

      When I was in high school we read things like “Grapes Of Wrath,” “The Octopus,” “The Jungle.” I don’t think people born after about 1960 who read those books get the same feelings we did — the environment is just too alien to them, where it seemed almost familiar to us.

  2. Fred Donaldson says:

    Using the term 11 million “undocumented workers” is incorrect. About 40% of all Americans are employed full time, so of the 11 million “undocumented”, expect about 4.5 million employed and the rest dependent.

    Entering the country illegally does not mean that you are a job seeker. You could be a criminal, a family dependent, a wanderer, or just confused by numbers and believe you are escaping the 4.85$ unemployment rate in Mexico to “benefit” from the 7.6% rate here among citizens not in the shadows.

    Anyone without a job will not come out of the shadows and many folks with jobs will see the danger of being, in effect, an indentured servant for a decade to some boss who could care less.

    • smith says:

      102 million people work full time. Workforce is around 150 million (includes 34 million part time and 12 million unemployed) That’s out of 250 million workforce representing 63% participation. The rest of the population is mostly 16 and under.

      I would imagine work participation of 11 undocumented immigrants could be higher or lower than 63%, and depending on unemployment level, it’s a good question to ask, how many are currently working. For native Americans only 33% of the population work full time.

      I favor immigration with no employment restrictions, no requirements for employer sponsorship or temporary work contract, free labor, free men, free soil.

    • Kevin Rica says:


      The number of illegal workers is probably much higher. The number of SS “no matches” peaked around 9 million. Conservatively 7 million of those were illegal. The Westat appraisal of E-verify:


      implies that we could have an equal number of people using stolen identities, but not detected by E-verify.

      And then there are those off the books.

      Maybe we only have 11 million, but maybe far more are here and working. Eleven million is, at best, a plausible bottom of the estimated range.

  3. Kevin Rica says:


    You keep referring to the 1990s, the golden age of Clinton and Rubinomics when immigrants poured across the border without depressing wages.

    Bad example! With that sort of job and GDP growth, we should have expected serious wage growth — which did not happen. wage growth was positive but anemic. (But of course, the 1% did fine.)

    Compare that to the 1950s and 1960s (before the “jaws of the snake” opened up). Because of the 1924 restrictions, from the 1930 until the 1970 decennial census, the number and proportion of immigrants in the population was low and declining and headed to the century’s low. Then everybody’s income went up — and people nostalgically idealized immigration.

    Then a sudden surge in immigration began in the late 1960s and from the early 1970s the good wage trends reversed.

    Conclusive proof or coincidence? As Damon Runyon said: “The race is not always to the swift nor victory to the strong, but that’s the way to bet.”

  4. Frederic Mari says:

    Thank you for an interesting post.

    I think I agree with most of your descriptions and comments and yet I come to a slightly different conclusion: http://theredbanker.blogspot.com/2013/07/immigration-trade-disturbed-macro.html

    Basically, in a weak economy, where the lack of labour supply, whether highly-skilled or low-skilled, cannot be said to be the main issue, I am wary of betting that complementary effects will be so strong as to supplant the first order effect of increasing wage competition.

    It doesn’t mean that the USA shouldn’t reform its immigration policies but it should, imho, be wary of importing too many new people, from anywhere, regardless of qualifications. Until the economy starts looking up…