Migration & wages: more evidence

Here’s yet more evidence (pdf) that immigration is good for natives’ wages:

We find that all European countries experienced a decrease in their average wages and a worsening of their wage inequality because of emigration. Whereas, contrary to the popular belief, immigration had nearly equal but opposite effects: positive on average wages and reducing wage inequality…
Public fears in European countries are misplaced; immigration has had a positive average wage effect on native workers.

This seems counter-intuitive. Immigration represents an increase in labour supply, and increased supply means lower prices. Conversely, emigration reduces labour supply which should raise wages.
However, this intuition is wrong, for three reasons:
1. In the long-run, the capital stock should increase in response to an increased labour supply, and this raises the demand for labour.
2. Immigrants are imperfect substitutes for native workers: the arrival of a Polish builder, Indian IT worker or American investment banker does not mean increased competition for electricians or lawyers.
3. Skilled immigration can have a positive external effect upon others’ wages. The IT specialist who helps get a foreign firm operating in the UK, in effect, helps that firm increase employment of native workers. The American investment banker creates a demand for nannies and cleaners. The Polish brickie helps builders and electricians to get more work done, and so on.
Now, the effects here are small - only a couple of percentage points over 10 years. And they are contingent upon the premise that immigrants, on average, have a higher skill composition than natives.  And I appreciate that concerns about immigration are not really rooted in its effect upon wages. But even so, this is more evidence that immigration does not have adverse labour market effects.
I reckon there are three policy implications here.
1. A cap on highly skilled immigration is a stupid idea.
2. Insofar as emigration reduces the wages of stayers, there is more reason to worry about the possibility that higher incomes taxes might encourage high earners to migrate. Fortunately, there is - so far? - little sign of them doing so, but this weakens a little my prior support for higher taxes.
3. It’s important to ensure that the capital stock can adjust upwards in response to immigration. Exactly how to do this is of course a matter of debate: is it more important to maintain aggregate demand and get banks lending, or to reduce red tape and capital taxes and barriers to entrepreneurship? Whatever the answer, the question becomes a little more important.

Comments

[Luis Enrique]

I'd like to see some research that is not based on averages, but says something about heterogeneity (perhaps this paper does: I haven't looked). I think I have read you question the justification of imposing costs on one group in order to confer benefits on another. I'm sure immigration has held down wages in some sectors, I'd like to know how bad the worst effects are.

I reckon it's a tricky empirical problem - hard to construct a counterfactual.

[John Rentoul]

I can't read the paper properly, but there is an obvious question about the direction of causation here. Surely immigration is a symptom of economic growth and emigration of the opposite.

[maplestory mesos]

good luck. plz keep it away.

[nm]

Your second point argues that immigrants are imperfect substitutes for native labour, but surely a Polish builder is in fact a near perfect substitute for a native builder. As immigration is perhaps skewed towards certain trades / industries it is not really suprising that we get some resentment from workers whose wages are being pushed down by increased supply. This doesn't make the case against immigration, just that we should recognise that the costs and benefits are not equally shared out.

Of course many of those who talk loudest on the subject are often those who are entirely on the positive end of the deal (think Daily Mail journalists) but no amount of evidence will stop them from venting their bullshit ideas.

[chris]

Here's some evidence that immigration does reduce wages for the low-skilled - although the effect is small:
http://stumblingandmumbling.typepad.com/stumbling_and_mumbling/2008/11/immigration-wages-more-evidence.html
I'd suggest:
1. Insofar as immigration does reduce low-skilled wages and raise higher skilled one, the answer is not to control migration, but to take measures to support low earnings.
2. However you look at it, immigration is only a minor contributor to the worsening labour market conditions of the unskilled. Technical change, globalization, the decline of unions, rising power of bosses etc are more important influences. Anyone who was seriously concerned for the low paid would think far more about these issues, and less about immigration.

[Trey - Swollen Thumb Entertainment]

I tend to fall in with the belief that Immigration is often a byproduct of a productive marketplace. I don't know if I believe that Immigration will usually cause higher labor wages, but I'm not opposed to the idea.

In the case of America, I do believe that, with Mexican labor, it has caused an artificial shift, as there are fewer low income jobs available for natives, (ahem, I meant whites, not "natives") so a byproduct of that may have been this boom in entrepreneurship that America has seen lately. More people have been "forced" to create their own jobs, since they are under the impression that the immigrants are taking them.

In this case, it's difficult to argue that emmigration hurts an economy, because a good chunk of the money that Mexican immigrants make goes back to Mexico, yet it remains a 3rd world country.

[charlieman]

@nm, point 4: "Your second point argues that immigrants are imperfect substitutes for native labour, but surely a Polish builder is in fact a near perfect substitute for a native builder."

A 1900 mill worker in the north of England would have been interchangeable for one from a neighbouring factory or town. The environment, processes and machinery were similar.

Houses and business premises are not made in the same way in different countries. I'm not familiar with how they do things in Poland, but here's a for instance. A detached house in the USA will typically be wood framed, but that would be unusual in the UK. The order of construction is thus different, as is the role of a plumber in its construction. A UK plumber would not be a direct substitute for an American plumber.

In IT, customers services and data centres are managed differently according to business sector and nation. A worker might hop conveniently between between countries inside one organisation. The same worker might flounder when flipping between businesses within his/her home nation.

[vimothy]

Regrettably little time to think or read about this, but let me just wade right in regardless.

My first thought is that agg effects of immigration are dominated by questions of distribution of income and that for a cheaper supply of un-skilled or semi-skilled labour, the top of the income dist gets no negative pressure on their wage for increased negative pressure on those beneath them and cheaper goods and esp services.

E.g., lots of Iranian civil engineers trying to find work as taxi drivers means that taxi drivers’ wages are lower, and therefore so is the price of taxis. As a taxi driver, you don't give a fsck about the price of taxis, but you sure do about your wage. City bankers on the other hand face no downsides but hey, the price of getting a taxi to and from work every day is getting a lot cheaper and goddamn if I can’t get my suits dry cleaned for next to nothing.

The payoffs there are totally asymmetric. Which is why it doesn't make sense to average them out, I don't think. And also why immigration is bad, or something.

Also, how does that study find that average wages are increased under immigration and decreased under emigration? Anyone with time fancy offering a quick précis? Sounds dubious!

[Andrea]

I really liked your post. It has definitely left me thinking about immigration and its consequences. Although I think that immigration has to be controlled for safety reasons, I agree with you on the fact that skilled or non skilled immigrants willing to work for the country can be more beneficial to all of us than having people flying off the States just because they cannot make a living here.

[shaiya gold]

good luck. plz keep it away.

[BenSix]

"Andrea" is about the most eloquent spambot I've seen. Perhaps "Discount ID" has an interest in immigration.

[Laban]

"And they are contingent upon the premise that immigrants, on average, have a higher skill composition than natives."

At which point the whole building collapses.

With regard to IT, only a minute fraction of the techies coming over here, mostly via intra-company transfer (ICT), are more skilled than the local techies. In five years of working alongside them I've only met one code god. Their USP is that they're cheap.