Several observers believe that insofar as the government's points-based immigration system reduces immigration (which as Jonathan says is "far from certain"), it will do real economic harm. Ian Dunt says it "is a bitterly stupid and small-hearted thing to do." And Anthony Painter writes:

The scale of the change and its suddenness risk very significant negative impacts. Businesses and public services could struggle to fill vacancies, seasonal businesses could especially suffer, and costs could escalate impacting business and public service viability.

These harms arise because, to the extent that the new policy is based upon economics at all, it rests upon a fallacy - the idea that, as Iain Duncan Smith told The World at One (20'58" in), immigration has a "very negative effect on earnings."

This is plain false. The nearest thing we have to evidence for it is a paper by Stephen Nickell and Jumana Saleheen which estimates (pdf) that a "10 percentage point rise in the proportion of immigrants is associated with a 2 percent reduction in pay." A 10 percentage point rise is, however, a humungous number; it's equivalent to over three million workers across the whole economy. In reality, their estimate implies only a tiny actual effect on the earnings of the low-skilled. As Jonathan says, it suggests that:

the impact of migration on the wages of the UK-born in this sector since 2004 has been about 1 per cent, over a period of 8 years. With average wages in this sector of about £8 an hour, that amounts to a reduction in annual pay rises of about a penny an hour.

This tallies with the general consensus, that migration has little impact on the wages of natives. A survey by the Migration Advisory Committee concluded (pdf):

Migrants have no or little impact on the overall employment and unemployment outcomes of the UK-born workforce...Migration is not a major determinate of the wages of UK-born workers. We found some evidence suggesting that lower-skilled workers face a negative impact while higher-skilled workers benefit, however the magnitude of the impacts are generally small.

This is consistent with my chart. It shows that the share of wages in GDP is higher now than it was in the mid-90s - despite a near-trebling in the number of foreign-born workers during this time.

It's not just in the UK that migrants don't depress wages much. Studies of big waves of immigrants in other countries have found a similar thing. For example, the mass migration of Europeans to the US in the 1920s (pdf) "had a positive and significant effect on natives' employment and occupational standing, as well as on economic activity." And a study (pdf) of Russians' migration to Israel after the collapse of the USSR found that it "did not have an adverse impact on native Israeli labour market outcomes."

The evidence then, seem clear.

But it is also counter-intuitive. Surely, if you increase the supply of something, its price will fall.

No. There's a lot more going on than just increased supply. In some ways, migrants can boost wages.

First, they don't just add to the supply of workers. They also add to demand. When the Polish builder spends his wages in Wetherspoons, he is adding to demand for bar staff.

Secondly, migrants are not always substitutes for native workers. They can also be complements, helping them earn more. If there's a shortage of roofers, plasterers and electricians won't be able to work on new houses. Importing roofers solves this problem, allowing plasterers and electricians to earn more. And if women can use cheap immigrant childcare they can go out and work longer hours.

Thirdly, migrants give natives more opportunities for management jobs. If we import a lot of fruit-pickers, who is going to oversee them? As Giovanni Peri says in a study (pdf) of Danish immigration:

An increase in the supply of refugee-country immigrants pushed less educated native workers (especially the young and low-tenured ones) to pursue less manual-intensive occupations. As a result immigration had positive effects on native unskilled wages, employment and occupational mobility.

Now, you might think these mechanisms limit the impact on natives' wages but don't stop it entirely. If this is the case, however, some other mechanisms operate.

One, say Banerjee and Duflo in Good Economics for Hard Times is that the reduction in wages "makes it less attractive to adopt labour-saving technologies", which helps sustain demand for native-born workers.

Another is that lower wages mean lower prices. That raises real incomes and hence demand for labour.

Even if all these mechanisms don't operate, though, another would. The Bank of England would respond to lower wage inflation by cutting interest rates. And this would raise labour demand and hence wages.

And remember: locking people out of the UK will not stop them depressing wages. If migrants were to stay at home manufacturing goods on low wages, the import of these would cut natives' wages. The factor price equalization theorem tells us that trade can reduce wages even if there isn't a single immigrant.

Evidence and theory therefore tell us the same thing - that migrants do not significantly depress wages. There is, therefore, no economic justification for immigration controls.

So, why do people think otherwise? There is perhaps an analogy here with the MMR scare. Some people found it easy to believe that the MMR vaccine caused autism because there were so many stories of children having the vaccine and then being diagnosed with autism. Similarly, there are stories of migrants arriving and natives then suffering worse wages and job opportunities. In both cases, though, a simple fallacy is at work - the "after it, therefore because of it" fallacy.

In truth, though, there are other reasons for flat-lining real wages - the many failures of British capitalism which have been amplified by bad economic policy.

And herein lies the reason why Duncan Smith is so keen to claim that migrants depress wages a lot. He's simply using them as a scapegoat to deflect blame from the bigger causes of low pay.

And it is scandalously easy for him to get away with this, because the media prefers to give a platform to racists* than to educate people about economics.

* I'll not link to that now-notorious Question Time tweet; we shouldn't satisfy the BBC's lust for clickbait.