Weak and Strong Arguments for Free Migration

[This is a post in my series on free migration. I will keep an overview with short summaries updated that you can find here: Synopsis: Arguments for Free Migration.]

Immigration restrictionists, even rather sophisticated ones, often have a simple worldview: People from elsewhere are worse than those who already live in the country or they are even inherently bad. There are sometimes exceptions restrictionists grant, but that is the prima facie assumption. From there, they jump to a claim that immigration is bad, and not just somewhat bad, but catastrophically bad.

If you look at the argument as it stands, the conclusion does not follow that outcomes will be catastrophically bad. It is reasonable to conclude from the premise that the direction is “bad,” but not how far it will go. But then this is intuitive thinking where standards are sloppy. The conclusion is actually not a conclusion, but already part of the worldview. (See my post: “Worldviews, Narratives, and Ideologies” for more on what I mean by a “worldview.”)

If you listen to immigration restrictionists they have a clear vision in mind how a country is in a civil war and mired in chaos, everything has gone down the drains: society, culture, the economy, etc., the native population is swamped by immigrants, and “bad hombres” have taken over. This is treated as a self-evident fact. It is immune to contradictory evidence that your country is not in such a state, not even by a far stretch, or moving in that direction at fast enough a pace to make the vision a reality anytime soon even if you could extrapolate a trend.

A sure indicator for a worldview at work is that emotions are intertwined with arguments. You can easily sense a deep unease, anguish, and even fear with many restrictionists. But once emotions are involved, rational thinking is out the window because you have to make decisions — and fast.

The only option here is to start fighting right away before it is too late which is today or rather yesterday. Entering into a discussion seems like a dangerous distraction in such a situation. It is like the lion already jumps at you, and you want to debate whether 6268,6615,3491,9154 is a prime number. (If you wonder: No, it is an even number and divisible by two. So we have time for handling the lion.)

My point about a worldview is not to discard any arguments from immigration restrictionists. Some of them are sophisticated and are worth your attention. What I only want to explain is why certain arguments seem obvious to someone although they are patently false. If you have a worldview, all it takes are some superficially plausible connections and some pattern matching to view something as a proof.

— — —

Take for example this type of argument:

Your country will soon be taken over by Muslims and that will be very bad. Note that this is a factual question, and the judgments in the latter part should not play a role for what the answer is. But then intuitive thinking does not distinguish between facts and judgments.

To prove the point someone then argues that indeed Al Qa’eda or Da’esh has a horrible ideology and commits terrorist acts. They are bad people, are you so foolish to doubt that? Sure, that is not contentious in my view. But how do you get from there to your conclusion?

Or an argument might be that immigration from societies where you have Islamist movements might lead to terror acts in cour country. And becoming the victim of terrorism is very bad, or do you mean you like that? The point is plausible as far as it goes. But how do you get to the conclusion that there will be a takeover in the foreseeable future?

Or it is an argument that in your reading of Islam the end-goal is to conquer the world and to make it wholly Muslim by force. They really want this, how can you deny that? But then that has been the case for more than thousand years, and it has not happened. How do you get from someone’s wish to its realization, even if it were so? Why doesn’t your wish that it be otherwise lead to the opposite outcome?

— — —

The missing part here is how it would come to such a takeover and whether that is possible with immigration at current levels, with people like actual immigrants, and with the current mix of sending countries, or even with much more immigration. Maybe there is an argument, but I have not yet seen it from immigation restrictionists. And it is a factual question, not one whether you think it would be good or bad. A judgment call is beside the question.

But then you will get a dramatic vision instead what it would be like if the horror scenario were already the case, and how it would be very bad. Or how certain things at present look eerily similar to what would be so in the scenario. Wouldn’t pork be outlawed under such a regime and was there not some public institution that began to serve less pork? That proves it, this is just the low-level version, the first step down a slippery slope. Never mind that at that pace, it should take perhaps 100,000 years before we were there, again assuming that an extrapolation is warranted.

Or you get a scenario with an endless list of assumptions: All people from Muslim-majority countries are Muslims who are all Islamists who are all on board with one terrorist faction. And then suddenly 100 million of them and noone else immigrate next year. How could that not lead to a takeover with disastrous consequences?

Or you get a lecture on how fertility of 10 will make a tiny group a huge one in no time. You know it is the exponential function which grows so fast. But then average fertility in Muslim-majority countries is just around and often even beneath the replacement level.

Or there will be mass conversions to Islam. Proof: There are some people who convert to Islam. It is possible, so everybody will convert soon. Etc. Etc. Etc.

The point here is that all these arguments do not prove the contention. They are only loosely associated with the conclusion, which is just assumed as a fact. And hence no more than plausible connections and pattern matching are needed as confirmation. This is good enough for intuitive thinking.

— — —

Unfortunately in my view, many who argue against such a worldview try to do it with their own worldview that only multiplies everything with -1:

People from elsewhere are better than those who already live in the country or they are even inherently good. There are sometimes exceptions that are granted, but this is the prima facie assumption. From there, such people jump to a claim that immigration is good, and not just somewhat good, but miraculously good.

One symptom of a worldview is that facts that do no fit in are either hammered into the grand picture or are just ignored. Immigration restrictionists will admit when cornered that, yes, some immigrants might sometimes be also good people or bring something to the table. But then, that is an exception and it would be naive to miss that they are always worse and even inherently bad.

And the opposing view has the same trouble of owning up to contradictory evidence. If cornered, they might also concede that, yes, some immigrants might sometimes be also bad people or cause problems. But then, that is an exception and it would be naive to miss that they are always better and even inherently good.

You can then get equally sloppy arguments where at most the direction is established, but the leap is immediately to extreme claims. The conclusion is primary: With lots of immigration, everyone will live happily ever after. On every dimension things will be perfect because of immigrants. They will create jobs, develop great technologies, save the pensions system, bring peace and stability, and what have you. Maybe there is an argument here, but I have not seen it either.

— — —

What’s my point then?

I would simply point out that this is a false alternative between “immigrants are worse or even inherently bad” and “immigrants are better or even inherently good.” Maybe the first stab should be: Immigrants are roughly like natives with all their good and bad sides. We are all human beings after all. Sometimes immigrants are better and sometimes they are worse in some regards, but maybe mostly they are not extremely worse or extremely better.

The assumption to start from should then be that the impact could be more or less neutral. There might be some things that improve with immigration and some that might deteriorate, but perhaps on a modest scale. Maybe this is just not the choice between heaven and hell.

And yes, it is also conceivable that in some regards things could improve a lot or also deteriorate a lot. In the latter case that would be a serious problem. But maybe not one that crushes everything within a short time. If so, the only response here need not be: No immigration. It could also be: Immigration and tackling a specific problem head-on, which seems doable if things otherwise more or less stay the same.

If you absorb this, it is also a worldview. But then my argument is not that you should not have one, which I think is not in the cards for human beings. There is no problem with jumping to intuitive conclusions as long as you are aware that that’s what you are doing and as long as you don’t get carried away by a feeling of self-evidence.

Just don’t forget the next step and check whether the first stab was right. If not, work on your worldview to keep it close to reality. And my specific claim here is that the assumption that immigration has a more or less neutral impact almost always leads you to the right conclusions that also survive rational scrutiny.

— — —

Let me illustrate my point with two examples that I have come across just over the past few days:

The first is from an immigration restrictionist (or so I assume from his short argument) on Facebook. The exact wording does not matter because the argument is pretty well-known: Immigration will lead not only to an expansion of the welfare state, but a government default, extremely high taxes, economic catastrophe, and the whole “road to serfdom” thing.

My point here is not that I can exclude this under any circumstances. You can make up scenarios where this could play out: There are 100 million Chavistas or North Korean Communists in the world who are fanatics. They all immigrate next year to Germany and noone else. And they are also well-armed and determined to take over. Then maybe I would agree with you that this could end in a catastrophe for everybody, not only natives.

However, that does not answer the question whether it also possible without such fancy assumptions. Can it also happen with immigration at current levels, a diverse mix, and people who are not that fanatical? I would say the simple answer is: No. And you can grant even a lot of leeway here: More immigration, a less diverse mix, and you could even throw a contingent of Chavistas and North Korean Communists in, and the answer would still remain: No.

The argument on Facebook was narrower. It only turned around an expansion of the welfare state. As I have argued in another post, that does not imply an inevitable “road to serfdom,” so even further arguments would be necessary to arrive at catastrophic consequences. See my article here for more details: “Both Hayek and his Opponents Were Wrong.”

My point is so simple that my only explanation how you could miss it is that you have a worldview where it does not fit in:

Let’s say someone in Germany gets 10,000 euros a year on welfare, and GDP per capita is 40,000 euros. GDP per capita is actually higher, and maybe also what someone gets on welfare. However, the argument is so safe that you can drive a truck through.

So, someone on welfare receives a quarter of GDP per capita. If you add 1% to the population and they stay on welfare for the rest of their lives, that will mean expenditures of 0.25% of GDP.

If you have 0.3% immigration, the average for the US and the EU, the share of immigrants will grow to roughly 50 times that (be careful: this only works for low rates, the relationship is not linear). That is so because immigrants typically come at an age of 25 or so. They then live another 50 years before they die. With fixed immigration, new immigrants will only replace previous immigrants after some time who die, and the share levels off after about half a century. It is trickier than that, but as a first approximation that is quite good. There is a simple fallacy for intuitive thinking here: If there is always an inflow of new immigrants, then the share must rise and rise. But that is not true if you think about it.

Now, the end result might be something like 15% immigrants in the population. Hence if all immigrants remained on welfare for the rest of their lives that would cost about 4% of GDP. Expenditures for all welfare programs in Germany are now about 20% of GDP. That would be a noticeable change, but not a catastrophic one. It is hard to see why 24% would lead to a government default, economic collapse, etc.

But then the argument is silly. Not all immigrants will remain on welfare for the rest of their lives. Let’s suppose it were 25% of them at any point in time, which as far as I know is absurdly high. Then this cuts the costs from 4% of GDP to 1% of GDP. And that is not yet all. The other 75% also contribute something to finance those on welfare. So you would have to subtract that, too.

Actually, there is a ferocious debate whether the impact of immigration is minimally negative or maybe minimally positive on the whole. It could also be that immigrants add something here. But in any event, it has to be very little. It is patently silly to say that a change from 20% to 21% will lead to a catastrophe. It cannot be so just because of the orders of magnitude involved.

You can make arguments all you want how bad immigrants are and how they want to exploit the welfare system and all that. That does not change the order of magnitude, which is small. The realistic assumption here is that the impact of immigration is neutral with maybe a minimal deviation on either side, no more.

If you are concerned about a rise of welfare expenditures from 20% to 21% over decades, the first thing to note here is that you have plenty of time to study the question and fix the problem. This is not a lion that jumps at you and where you have to make a decision in a split second.

Okay, you think this a problem? Then there are many ways to keep the level at 20%. Surely, that will have some effects, but perhaps no major ones. I mean a rise by one percentage point could also happen for many other reasons apart from immigration. Is the contention here that there are no conceivable remedies apart from no immigration at all?

How about seeing to it that the economy grows a little faster? In that case an absolute increase could go along with a decrease in the share. All you need here is that the economy grows once by an additional 5% while welfare payments grow along without the 5%. Do the math: 21/105 is about 20% what it would be without immigration under rather pessimistic assumptions.

Are there no ways to achieve this? Especially libertarians baffle me here who seem to think that potential growth cannot be improved by something like 0.15 percentage points over 30 years once (1.0015³⁰ is about 1.05). You would have decades of time for that. And that is, to repeat, only under very pessimistic assumptions about the impact of immigration that might be wrong. Therefore, I don’t see how you can believe you are faced with a catastrophic development that cannot be handled without shutting immigration down ASAP.

— — —

Now this was the restrictionist side. Here is the side of someone who is enthusiastic about immigration that I found on Twitter. Again, I don’t need the specifics, the argument is well-known: Immigrants will save the pensions system that is supposed to be in deep trouble because of a rising share of older people. Note that that has nothing to do with immigration and would happen anyway.

Immigrants typically come at an age of perhaps 25 years on average. They will hence work for decades and pay into the pensions system without receiving anything. That’s why there is a positive effective effect on the welfare system and also why it is not obvious why the net contribution has to be negative. However, that mostly only applies for more immigration compared to a base level. With continuing fixed immigration, the immigrant population will also grow older, and then there are also immigrants who draw pensions after paying into the system.

Basically you have a stable immigrant population after a few decades. The age structure is different from that for natives because immigrants join the population not at birth, but later. Actually, what that means is that investment for education and upbringing happened elsewhere. This is a present immigrants bring along. Sending countries might not like that, but receiving countries have no point here. You can complain that immigrants have little education. However, if you don’t pay for it, the baseline is zero, and then it is just a smaller present, which is still a present.

Since immigrants crop up on average only at age 25 or so, this means that you will have fewer children relatively in the immigrant population at the start than in the native population. Note that this does not mean they will have no children, only that they do not bring them all along. Most grow up as natives, and the technicality of where someone was born does not change that. Those are not immigrants, but natives. Otherwise the rest of the population structure is similar to that for natives because it is only like immigrants were born at an age of 25. What happens later is determined by mortality, which should be roughly the same as for natives.

That means that the immigrant population after it has leveled off has a larger share of people of working age than the native population, but not by much. Let’s assume that the missing children (who are not gone, but still in the source countries before they emigrate) make up 20% if you added them in. If you have 50 out of 100 of working age in the native population, you will have 50 out of 80 in the immigrant population. That is roughly 60% or about 10 percentage points more.

Now, if the share of immigrants levels off at about 15% of the total population, a rough estimate for the impact on the share of those of working age would be 1.5 percentage points versus no immigration, or maybe a change from 50% to 51.5%, rather minimal.

My calculation here is very approximate, but even if you fiddle with it a lot, it is hard to see how you could get more than a change by a few percentage points in the lower single digits. Of course, that will not make a major difference. No matter how many arguments you have that immigrants are hardworking or how good they are in general, you cannot overcome this. It doesn’t matter much because a minority in the population which is only slightly different from the rest cannot yield a large order of magnitude for its effect on anything.

Note, though, that this only applies over the long run. At first immigrants can boost the share of working age more. The “guestworkers” who came to Germany in the 1970s initially had labor market participation of 80%, not even people of working age, which is incredibly high for a population. But then it had to go down over time to like 50%. The general fallacy here is that this proves immigrants somehow got lazy and went on welfare. But 50% is normal, just as for natives. And you can only have a share of 80% when immigration goes up massively and only for some time. To keep that up, you would have to have ever rising immigration, which is not my assumption here, but fixed immigration.

All in all, the answer is: No, immigration cannot change the ratio a lot between those of working age and those who draw pensions. It can at best make a small difference. Although the argument is very popular, it is mostly false.

— — —

As an aside: There is also another problem with this argument how immigration can avert a catastrophe for the pensions system: Maybe the supposed crisis from a larger share of old people is not that serious.

To give you a rough idea of my argument that I will develop in further posts: The share of old people mostly depends on the level for fertility. If it is low, there will be relatively fewer children, more pensioners. If it is high, there will be relatively more children, fewer pensioners. However, the two sides practically cancel out: The share of those of working age is almost independent from the level of fertility. (If fertility changes you can have various effects in both directions, but not forever.)

However, if the share of working age remains the same, the question is whom they will have to support. With higher fertility you have to pay for more children, fewer pensioners, with lower fertility for fewer children and more pensioners. So the correct approach is to net the two sides. It is not so that only pensioners cost something, so do also children. That makes the overall effect much smaller than it appears in dramatic scenarios where so and so many people who work have to support a pensioner. The counterfactual is that they have to support as many people young and old, just fewer pensioners and more children.

— — —

To get back to the main line of thought after this short digresseion: My point was about how to argue for free migration and how not.

I would suggest this approach:

It is neither reasonable to assume that immigration leads to catastrophic results nor that it will work miracles. The natural first stab should be: Immigration has more or less a neutral effect. There can be deviations from this on both sides: good or bad, but they have to be very modest with current levels of immigration and even much higher levels.

You can go to extremes here with your assumptions, and then that may no longer be clear or even false. However, that can only lead to an argument against these extreme scenarios, not against current or even much higher levels of immigration. There can be more serious problems even at those levels. But it is plausible that they should be only in a few domains and that they are not systemic. They can hence be tackled head-on. And before something builds up you have plenty of time, and do not have to make a decision on the fly.

If that is so, restricting immigration is not necessarily the best or even the only possible counter-measure here because such serious problems could also arise without immigration. It is not so that there cannot be terrorist groups without immigration. The RAF of the 1970s in Germany was as native as Steve Bannon would wish.

The lesson here is that you should clamp down on the problem itself, not immigration in general (BTW, I’m fine with keeping terrorists out of a country). Since it has to be a modest problem and not systemic because of the numbers, there is no reason why you could not do that. And heading for the problem itself and not immigration in general seems like the sensible thing to do.

— — —

The strong arguments for free migration lie elsewhere in my view, and I will develop them in further posts. I would call the central points the Liberal, the Social Democratic, and the Conservative arguments for free migration, where I view all these political directions as branches of one common old Liberalism (see my post: “The Liberal Order — An Explanation”).

The foundational principles are: Universality, Equality, Individualism, and Liberty. Free migration is not about certain groups, but about humankind understood as individuals who have all an equal right to be free and live well. Different directions descended from on old Liberalism diverge as to what that means and how to trade different elements off against each other and when restrictions are warranted, but they agree on the principles. That is not a trivial point because there are many political directions that challenge some or even all of the principles and which are fundamentally anti-liberal.

Here are now the three arguments in a nutshell:

• The Liberal argument for free migration is that every human being should be free to decide where to live and work. Any restrictions on this liberty need a very strong argument that has to be better than: We have the power and can force it through.
• The Social Democratic argument for free migration is that humankind would be much wealthier if we could tap into the economic potential that is now only latent. Restrictions on migration are a form of oppression of those who had the bad fortune to be born in the wrong place.
• The Conservative argument for free migration is that to defend civilization, understood as the Liberal Order, it should be strong. The more people live in a Liberal Order the stronger it is in general. Only in this way can we preserve and even expand it over the whole world, which would make civilization safe because we can drive barbarism out once and for all.