About Those Burdensome Refugees…

Trump’s executive order this weekend has got everyone talking. A lot of people out there are now taking positions on issues that most people know very little about, and which involve an extremely complicated policy domain - immigration. Even at the most basic level, there seems to be some confusion over the many moving parts of the executive order. Trump’s executive order temporarily ends travel to and from seven middle eastern countries for those with non-immigrant travel visas. When it was first announced, the order affected those possessing immigrant visas (green cards) as well, but this has since been walked back by the administration. Then, completely separately from this, the executive order also temporarily bans all entry by refugees, indefinitely bans entry by Syrian refugees, and cuts the number of refugees allowed into the country this year by more than half, from 110,000 to 50,000 in FY 2017.

This article is going to focus very specifically on this very last piece of the policy — the reduction in the refugee admission ceiling by 60,000 in 2017. A lot of well-meaning people right now are trying to figure out where they stand on this issue. What should we do with refugees? Do they pose a threat to national security? Can we properly vet them? When people don’t know anything about a topic, but are nevertheless concerned about it, they often argue their points in the abstract, even when data exists that could allay many of those concerns. Here, I am going to try to convince you that, regardless of your partisan affiliation or stances on illegal immigration, visa checkpoint security, or Islam generally, everyone should be opposed to a policy that lowers the number of refugees admitted to our country. In fact, everyone should be supportive of a policy that increases the number of refugees admitted to our country.

I’m focused on the refugee ceiling for a few reasons. First, I’m choosing the policy that I think most clearly maximizes harm to people. Refugees, by definition, are those that have been able to demonstrate they are at great risk of persecution by remaining in their home country. Not being able to travel on a non-immigrant visa is inconvenient and harrowing; not being able to travel to the US as a refugee might mean imprisonment, rape, torture, or death. Second, I’m choosing the policy that I think is least defensible. One can at least demonstrate, for instance, that the non-immigrant visa program can be used by would-be terrorists to gain entry to the country — the 9/11 hijackers, for example. On the other hand, the refugee program, as I will show, is extremely unlikely to bear similar risks to American citizens. Third, it is the least temporary action in the executive order. All other actions are, at least at the moment, temporary measures that will last a few months. The change in the refugee ceiling is, in effect, final for the entire year. Moreover, the ceiling cap in the first year of a president’s term generally gives us a good idea of what that cap is going to look like for the next years. To the extent that the lowered ceiling is bad for refugees, it’ll be worse for them every year the cap is set at 50,000…or lower.

Refugee Admission in the United States

Source: New York Times, https://www.nytimes.com/interactive/2017/01/25/us/politics/trump-refugee-plan.html

In order to make it into the United States from a foreign country, one can take a number of different paths. If you are a member of a group of developed nations that are in part of the Visa Waiver Program, you can stay for a few months without really any requirements. If you are not a member of one of those countries, or you want to become a resident of the United States, you’ll have to obtain a visa. Quick visits can be handled on non-immigrant visas (for tourism/business/etc), and planned residence is handled through immigrant visas (green cards for legal permanent residents). You can also gain entry by applying to the United States as a refugee. If you gain refugee status, you will be allowed into the US for a year before then being required to become a legal permanent resident.

Many people seem to be under the impression that refugees all pile onto a boat in Country X, and are dumped into Country Y and told “best of luck” without anyone making sure the crowd doesn’t contain criminals or terrorists. This is emphatically not the case. Refugee status is extremely difficult to obtain, given that there is a long and relatively complicated screening process. This process requires security checks by four separate United States agencies, and given that less than 1% of global refugees are even eligible to make it through the first step of this process, it is safe to say that your chances of getting into Harvard are much higher than they are to make it into the country as a refugee. The process also involves finding a community to place refugees in, such that they have a reasonably high chance of becoming self-sufficient in America within a short amount of time.

In order to be admitted as a refugee, you must specifically be able to demonstrate that you are at high risk of persecution in your country of origin for your religion/race/politics/nationality, and that the government in that country is not providing sufficient protection for you and those like you. The people who enter the US through these programs are people that are in great danger. Their entrance to the United States will likely save them and their family from death, or fates worse than death. Anyone who is accepted as a refugee meets this criteria, so while Syrian refugees have gotten most of the attention recently, and for good reason (they made up a large percentage of refugees admitted in 2016), there are many refugees from many countries around the world.

Even if you pass the security checks and qualify for refugee admission, there is no guarantee that you will be allowed in. This is because we cap the number of refugees we allow into the country every single year at a specific number — the ceiling I was referring to earlier. Like other forms of immigration, refugees who meet the criteria but only after the yearly quota has already been filled will have to wait for future opportunities…but refugees often do not have time to wait another year for the quota to drop again. The number of refugees we admit every year has changed quite a bit over time, but is almost always over 50,000, and has been steady at about an average of 70,000 for the past couple decades. Obama recently increased the cap in 2016 to help accommodate the large growth of global refugees due to the Syrian conflict. The drop from 110,000 to 50,000 just implemented by Trump is the largest single-year change since 9/11.

Are Refugees Dangerous?

Source: Financial Times, https://www.ft.com/content/a8798b58-e347-11e6-8405-9e5580d6e5fb

Most people tend to have some measure of sympathy for refugees, but many of these people might also be conflicted by serious concerns that the refugee program could be used to accidentally let terrorists into the country, particularly when many of the refugees are coming from regions occupied by ISIS. This is a reasonable concern — but only if you don’t know very much about the refugee program.

First, because we have allowed refugees in the country for decades, we have data on how refugees have behaved during their time in the United States. Since 1975, there have been about 3.4 million refugees allowed into the United States, and only three murders by refugees during that time — all occurring in the 1970s, and all involving Cuban political activists. To use more recent numbers, since 9/11, we have resettled about 800,000 refugees. Of these refugees, three have been arrested for terrorism-related activities, none of them involved actual planned attacks in the United States (they wanted to send money and weapons to organizations abroad), and they were all caught by our intelligence agencies before they did anything. So, if history is to be our guide, refugees do not pose a serious danger to us.

It is the case that if you go looking, you will be able to find news reports of refugees committing crimes. That’s true — with a population of 3.4 million people, it would be pretty bizarre if precisely none of them were committing crimes. Still, the people who actually try to carefully estimate these sorts of things don’t think immigrant populations produce more crime than citizen populations, and this should be especially true of refugee populations, which are vetted more strictly than legal immigrants and, of course, far more strictly than illegal immigrants.

Perhaps, though, history cannot be our guide. Many people worry about the Syrian refugees specifically. Is it possible that the 2016 and 2017 refugee classes would be much, much more dangerous than in previous years? Answering this question would require the kind of factless speculation that ruins these debates in the first place, but there is a strong theoretical reason to not think the new refugees should be very different from the old ones. As previously stated, the refugee process is the last process you’d want to use if you were a terrorist, as it requires the most scrutiny. Instead, you’d probably want to try to gain entry through a temporary non-immigrant visa — again, precisely what the 9/11 hijackers did. While it may look, to the uninitiated, like the Syrian refugee crisis is a perfect opportunity to sneak into the country undetected, this analysis only makes sense if you know literally nothing about the refugee program.

But aren’t they currently checking to make sure that the vetting process is up to standards? Sure, but that doesn’t mean there was anything previously wrong with the standards. Trump, in particular, has a habit of trying to fix things that were never broken in the first place. That’s something you should expect from someone who doesn’t care about getting good information; they will assume that any problems that exist represent a total lack of proper care, rather than the end result of a process that has already mostly been optimized over decades of work. Reinventing the wheel does not make us safe, it just gets innocent foreigners killed unnecessarily.

Are Refugees A Burden?

Source: ORR 2015 Annual Report to Congress, page 18

Some may oppose the influx of refugees not on the grounds of national security, but rather budget matters. If refugees come into the country and immediately begin to use all of our federal welfare programs, then perhaps they are a substantial drain on U.S. taxpayers. While refugees are more vulnerable than your average citizen, given that they have few familial support structures in the U.S. and likely were able to take very little wealth with them when they resettled, refugees have demonstrated themselves to be surprisingly successful in terms of assimilation. All of the facts about this I am about to present can be found here and here.

While refugees are immediately eligible for all regular benefits upon getting resettled, the vast majority tend to be able to get off of assistance within several years. Most do not use welfare even in the beginning, and most who do are off of it after just one year. Most refugees do use the food stamp program at first, but after a decade of resettlement, just about 25% of refugees remain on food stamps. In fact, just about 6% of refugees use any form of cash assistance by their fifth year. Refugees also tend to have comparable or even higher employment rates than other U.S. citizens. After about a decade in the country, refugee yearly incomes are nearly at the U.S. median.

This may be very surprising if one pictures refugees to be confused third-world dirt farmers, but this is not the typical demographic profile of a refugee. Refugees actually are, overall, 75% high-school educated and are college educated at a rate (28%) that is essentially the same as US citizens. How is this possible? One has only to think about the hoops one must jump through to make it through the application process, which is long (18–24 months) and complicated enough that it is likely that only somewhat well-educated people with verifiable backgrounds would be able to enter in the first place.

It is somewhat tricky, but we can get a rough sense of how much we spend on refugees a year when factoring in resettlements costs plus all other usage of federal programs. For every refugee we accept, we spend about $12,800 a year on them in the first five years. Multiplied across the little under 70,000 immigrants we accepted in 2013, as an example, that’s a cost of a little under $1bn. However, for context, it is worth understanding that a large percentage of this spending goes to refugees who are also unaccompanied minors, and payouts to adults incapable of working. One billion dollars a year might sound expensive, but the federal government deals with large populations, and therefore unavoidably deals with large numbers. For comparison, the Iraq war has cost us (yes, we’re still paying lots of interest on the money we borrowed to help finance it) about $170 billion every single year since 2003, we effectively subsidize Wal-Mart to the tune of about $8 billion a year, we give $3 billion in aid to Israel every year, and even spend more ($1.5 billion) on government advertising each year. Across every adult in the United States, a billion dollars on refugees is about $4 a year per person. If everyone were willing to give up a cup of coffee a year, that’s the budgetary equivalent of saving thousands or tens of thousands of lives.

Finally, many of you might be concerned about whether refugees end up “stealing” jobs that would otherwise go to Americans. Academic research has found little reason to be concerned about this. Humans have a tendency to easily recognize why adding a human to a population removes one job from that population, but they have a much harder time visualizing the fact that a nearly equivalent amount of labor in that population must increase by about the same amount (business needs to expand to meet the needs of the growing population). So what is the tradeoff? While it is somewhat difficult to analyze cleanly, most estimates suggest that they actually create more jobs than they take, or at the very least have a neutral effect on the economy. This is true for several reasons, but in particular, refugees are more likely than the average American to open a small business when they arrive, and their willingness to accept low-end labor often encourages the growth of specialization and job training programs for other citizens in the area. The evidence that they are “taking our jobs” is weak.

America First

I have seen many people who oppose the entry of refugees on the basis of caring more about the lives of fellow Americans than those of non-citizens. I will accept this as a reasonable starting point, even if it is one I do not necessarily share. We could respond to this argument by saying it’s offensive or xenophobic, but that won’t win very many arguments, and it isn’t taking their argument seriously. It’d be hard to argue that humans generally value all other humans equally — at the extreme, we certainly value our family and friends over strangers. Valuing citizens over non-citizen is just another form of group difference, though certainly one that is much more abstract. A nation is more or less set up to protect itself, its territory, and its citizens, and very few understandings of the responsibility of the nation-state prior to the last several decades would include expectations of caring for the world at large.

What I do, however, object to is people using this argument as a way to shield themselves from all rational argumentation. My point is this — if you want to claim that you value citizens over non-citizens that’s fine, but if you are unwilling to make a rough claim about how much you value one over the other, you’re being disingenuous. Very few people would be comfortable with the idea of valuing citizens over non-citizens infinitely. For instance, would you allow every non-American human on the planet to die a painful death in order to save the life of one American? Saying yes stretches credulity, but if you feel comfortable saying yes to that, the rest of this section is not addressed to you.

So what is an acceptable ratio of citizen to noncitizen lives? A better question is this: what ratio would justify not letting refugees into the country? Let’s say we’re just trying to maximize the number of humans who don’t get killed in a given year. Every refugee brought into the country carries some level of risk to American citizens, and every refugee who is refused admission bears a certain risk of themselves getting killed as a result. If we value citizen and non-citizen lives equally, then to figure out whether or not we should admit refugees, we simply need to figure out which is greater: the odds of a refugee killing an American once admitted, or the odds of the refugee dying if not admitted. Using the existing data, this is not a very difficult calculation — based on the past forty years of refugee admissions, the chance of a refugee killing an American is zero, so we should let refugees in regardless of what the chance of them being killed upon refusal is. Even if you valued an American life a thousand times higher than a non-American life, because the risk to American lives is virtually zero, you should let the refugee in.

To give this thought experiment a little weight though, let’s just assume that the chance of a refugee killing an American is somewhat greater than zero. In fact, let’s assume that a major terrorist attack will result from allowing the extra 60,000 refugees into the country (a couple terrorists slip through), and this attack will claim 100 American lives. Again, there’s absolutely no reason whatsoever to think this will be the case, and many reasons to think terrorists would use completely different entry points, but what would happen if we didn’t let those 60,000 refugees in? The data on this is not available, but let’s just make a very conservative assumption that 1% of refugees will be killed if they are not able to flee their country. Remember, these are people who by their very designation have proven that they face a great deal of danger in their home country — the figure is probably much, much higher than 1%. That’s 600 people dead. If we value American lives equally to the lives of refugees, then we should clearly let the refugees in, even if a terrorist attack on American soil is the result.

But let’s say we don’t value these lives equally. How much more value would we have to place on American lives compared to non-citizens? Using the numbers above, we’d have to think an American life is worth 6 times the amount of a non-American life. That seems extreme to me, but may seem reasonable to others. Keep in mind, though, that the numbers I used previously were stacked against me in ridiculous ways. If we assume that it’s more like 10% of refugees who die if we do not let them in, which seems more likely to me, then an American life would have to be worth 60 times a non-American life. Or if we assume that a major terrorist attack is not committed by a refugee, but maybe one of them snaps and kills 2 people (neither of which we have a shred of empirical reason to believe would happen), then we’d have to assume an American life is worth three thousand times that of a refugee’s life to justify not letting them in.

You can play with the numbers as you want, but you’d have a difficult time engineering a plausible scenario that would result in numbers that justify not letting refugees into the country, even if you weigh American lives more heavily. Those who hide behind such an argument should be ready to volunteer exactly what level of human misery they are willing to accept abroad to receive negligible security improvements at home.

Source: Financial Times, https://www.ft.com/content/a8798b58-e347-11e6-8405-9e5580d6e5fb

Another version of the “America First” argument might be that a better use of our charitable funds would be to spend it on struggling citizens in our own country than on those abroad. While this is admirable, a couple points. First, the amount we spend on refugees is a virtual rounding error on the total amount we spend federally on domestic poverty, which is about $800 billion a year. $1 billion is likely a marginally effective amount to spend, given that the money is used to literally prevent people from being brutally murdered. Second, harboring refugees is an international mission, chartered through the UN, of which we are obviously a key member. Despite being one of the five permanent voting members of the UN, the US, compared to most other developed countries, takes in just a fraction of the normal amount of refugees; a little under 1 in 1000 American are refugees, while this number is typically double or triple in other countries such as the United Kingdom or Canada, and is well over 20 times as high in the Scandinavian countries. Choosing to spend even a paltry sum on refugees can be easily read by our allies as a dereliction of our basic duties as a world leader — especially one that is responsible for much of the displacement and misery that occurs abroad, particularly in the Middle East. Third, I note curiously that support for increasing the federal budget on poverty always seems to be much higher whenever someone suggests spending any amount of money on vulnerable foreigners. It sure would be a shame if such concerns about how we treat our own citizens were, say, being used merely as a cynical and temporary cover for being unwilling to help the global poor, wouldn’t it?

But Europe

Many who are willing to accept the above arguments still can point to one bit of hard data — based on the chaos we’ve seen in Europe, in terms of ethnic ghettos and terrorist attacks, aren’t the prospects for assimilation rather dire? While this is a reasonable concern, there are many reasons to think that Europe’s experience is not and will not be America’s.

First, Europe is much more accessible by foot and small boat traffic than America is, so the refugee populations that end up in Europe are far less likely to be vetted well or at all, as many are essentially illegal immigrants. Second, it is difficult to separate refugees from regular immigrants, so when we refer to the problems of muslim integration in Europe, it is very difficult to tell whether we are discussing actual refugees at all, or just migrants generally. Refugees, again, tend to look quite different from migrants in terms of education, global outlook, and ability to be culturally assimilated. Third, Europe’s problems reflect decades under a completely different immigration system than what we have in the United States. In the years following WWII, when much of Europe was in shambles, migrants from other countries were encouraged to come in and fill the many jobs now left empty by wartime deaths. In most countries, this process was expected to be temporary, and as such, the European countries did very little to plan for a long-term assimilation of those migrants. As such, migrant populations in Europe have tended to cluster in urban ghettos that have, over decades, developed quite separately from the rest of the city, and create an enormous amount of resentment between citizens and non-citizens. In the United States, on the other hand, we have not had this recent need to bring in enormous amounts of migrant laborers from the Middle East and Africa, and as such, our immigration policy is designed much more carefully for long-term assimilation. Even groups that have entered the United States and tended to form ethnic urban clusters, like Mexicans, tend to assimilate quite well long term. Finally, because the United States is so large, refugees can easily be spread out across any number of high population cities, minimizing the likelihood that they will end up falling into an ethnic ghetto where the chances of assimilation are low.


I hope that I have shown that the refugee program is something that should be championed not just by liberals, but by all American citizens. The program has enormous upsides for the refugee population, and extremely small downsides for the American population, besides the cost of resettlement and support, which adds up to a couple cups of coffee a year for the average American. Refugees tend to assimilate well, commit few crimes, and contribute to the local economy as would any normal citizen. Trump’s decision to decrease the number of refugees allowed in by 60,000 will result in the preventable deaths of hundreds — there just won’t be a single big event to turn those deaths into news. Instead, these citizens will just quietly and anonymously be hunted, raped, tortured, and slaughtered when they could have simply been given a new home. If anything, the program should be expanded far beyond 110,000 admittees. While I am not arguing we need to bring our levels as high as the Scandinavian countries, we can do better. I urge you to write and call your representatives, and tell them that we are not afraid of the refugees, and that we approve of low-cost efforts to save high-risk individuals from death and destruction.

Like what you read? Give Sean Freeder a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.