If you had to name a single government policy that ruins the greatest number of lives, what would you pick? The 45p tax rate? Saver-hurting inflation? Green energy subsidies?

I’d say that the biggest one is the one that free marketeers are largely silent about: migration controls.

In 2011 Michael Clemens looked at the economic estimates of the global GDP growth that would come if every country in the world abolished restrictions on the movement of goods, capital and labour across national borders. According to the papers Clemens looked at, removing all barriers to trade would increase global GDP by between 0.3% and 4.1%; removing all barriers to capital flows by between 0.1% and 1.7%. Those are big gains that would make the world a substantially richer place.

Completely removing barriers to migration, though, could increase global GDP by between 67% and 147.3%. Think about that: simply letting anyone work anywhere could more than double global GDP. And that would be a long-term boost to economic growth, not a one-off. Even the bottom end of that, 67%, is an astonishingly huge figure.

It’s not as far-fetched as it might sound. As Clemens points out, workers can often create wildly different amounts of value by doing the same thing in different places (or doing them with different people). A taxi driver who might expect to make $1,500/year in a city in (say) Benin might be able to make $31,000/year in New York City by doing exactly the same thing. That shouldn’t be a surprise: bringing someone like Sergey Brin to work quickly, saving him an hour, is much more valuable in terms of his opportunity cost than, say, saving me an hour.

The institutions that most successful countries have are extremely valuable too. Corruption, instability and political uncertainty all have the potential to be extremely costly for firms, and they often prefer to pay a higher up-front cost in labour terms to locate their production in stable countries with good institutions. That’s one reason why Nissan still prefers to build some cars in Sunderland than Haiti: the institutions effectively boost Sunderlanders' productivity enough to make their higher wages worth paying. If we let Haitians move to Sunderland, they could take advantage of those institutions and make a living for themselves too.

The counterargument will be that a Sunderland filled with Haitians will quickly stop being like Sunderland: Haitians might vote badly, or might be so culturally incompatible that the social institutions that are so important to Sunderland's success, like trust, would break down and ruin things for everyone. That’s a valid argument and probably the main thing we should be talking about when we talk about immigration. But it’s also ambiguous: immigrants tend to have lower rates of crime than natives, and increased contact between immigrants and their neighbours can mostly overcome the cohesion problem.

But even if these arguments did prove to be true, they would be a case for country-specific immigration controls: even if Haitians proved to be too culturally incompatible to come to Britain en masse without undermining what’s valuable about Britain, that would not necessarily be the case for Chinese or Sri Lankans. If this seems ugly it is much, much less ugly than our existing blanket controls on immigration. Letting more people come to Britain should be the priority, not preserving the appearance of cultural neutrality.

What puzzles me is that my fellow free marketeers are often very indifferent (if not openly hostile) to policies that make it easier for foreign people to work in Britain. They cannot believe the economic claims that immigrants 'steal jobs' in an overall harmful way unless they also think that free trade does. There are many keyhole solutions to prevent immigrants from sponging off the welfare state. The cultural arguments, if they can be classed as such, are worth considering but certainly not so powerful that they invalidate the economic arguments. And free marketeers are usually pretty happy to let society adjust itself rather than try to engineer it to become or remain the way they like it.

Fundamentally, migration controls are not just laws about what foreign people can do, they’re laws prohibiting businesses from hiring people and property owners renting or selling to people who were unlucky enough to have been born in the wrong place. On the fact of it, these laws are so staggeringly invasive that no free marketeer could be comfortable with them; when you realise the economic costs it is amazing that anyone can tolerate them at all.

There are lots and lots of bad things governments do that ruin people’s lives. But few cause as much harm to the poorest people as the state controls of where people can work and live that we call ‘migration policy’. Even a marginal step towards a more liberal immigration policy would allow people to create an enormous amount of wealth, and probably do more good than almost any other possible policy. So why don’t more free marketeers start talking about it?