Migration and wages

Though I am generally a big fan of the FT, this article by Andrew Balls (FT.com – subscribers only, reproduced free here) created a false impression of the contents of a recent Congressional Budget Office report (pdf) on the impact and role of immigrants in the US labour market.

Even the title, Increased migration to US holds back wages, is misleading. (I realise that journalists do not control the headlines that sub-editors put on their articles; but when the headline is comprehensively wrong it does suggest that the article may not have conveyed its meaning very effectively.)

Here is the first paragraph:

Increased immigration of low-skilled workers from Mexico and Central America helps to explain the pattern of low average wage growth in the US in recent years, the Congressional Budget Office said in a report released on Thursday.

You might reasonably conclude that migration is bad for the wages of native-born workers, since it slows down wage growth.  But you’d be wrong.  The CBO report is clear what causes this: the migrant workers themselves accept jobs for lower wages than native-born workers, and so depress the measured average wage of workers as a whole.

The CBO report does not find that increased migration reduces the wages of native-born workers.  It says:

More recent studies based on differences across a large number of local labor markets have continued to find little, if any, adverse effect on native workers. For example, based on his analysis of data from the 2000 census for about 300 metropolitan areas, David Card concluded  “Although immigration has a strong effect on relative supplies of different skill groups, local labor market outcomesof low skilled natives are not much affected by these relative supply shocks.

Now that is rather an important distinction.  The report does caution that there may be other changes which dull the impact of migration on native-born unskilled wages:

To the extent that employers or workers adjust their location decisions in ways that offset the otherwise adverse impact of immigration,the effects will be diffused.

The CBO paper has an interesting discussion of Borjas’s estimates which try to take account of this effect, making nationwide estimates of the impact of migration on wages of unskilled workers.  The estimates are inconclusive about the size of the effect, because the effect is muted by secondary adjustments that occur in response to the increased labour force (such as more investment, or greater educational attainment by native-born workers), whose size is unknown.

So what the CBO report tells us is:

  • The migrant workers are better off (presumably, else they wouldn’t come)
  • If there is a fall in the wages of native-born workers, it is too small to detect statistically, and the effect appears to be very small.
  • The economy is more competitive, goods are cheaper and consumers (including native-born workers) are better off.

That sounds like a win for everyone.  It is a shame that the article, and especially the headline, in the FT chose to sound a misleading warning about a fall in average wages rather than highlight the reports conclusions of gains for everyone from more open migration policies.

3 thoughts on “Migration and wages”

  1. If you look only at workers you neglect those who might otherwise have been workers – rather a selection bias, I should have thought.

  2. Dearime

    That’s true. However, I think it is right to keep in mind that, in principle, the dynamic benefits will eventually benefit everybody, including any workers initially displaced by migrant labour. The fact that any initial negative effects on native born workers are so small as to be statistically invisible suggests that the dynamic benefits may outweight any static costs rather quickly.

    Owen

Leave a Reply

Your email address will not be published. Required fields are marked *