What It Would Take to Keep America White

A Lot More Than You Might Think, But Also Not So Much

A tweet thread made the rounds over the weekend discussing how to keep America (non-Hispanic) white. It was shared by the alt right crowd in particular, and from a user whose handle is “Wasp Nationalism.” Here’s the thread:

“Our situation” here is:

Let me start this post with a few notes.

1. “Wasp Nationalism” clearly advocates for policies in favor of white, Anglo-Saxon Protestants. He describes himself as a “Dixie expat,” and tweets under a pseudonym from a Tom Wolfe novel: the character is a WASPy bond trader falsely accused to maliciously murdering a black teenager. This set of facts should make clear that this tweet thread is a neat encapsulation of the white nationalist right’s concerns about demographics.
2. I intend to treat these concerns seriously and with as much neutrality as is possible. This is partly because I think they have a grain of truth, as I think the left takes integration and national identity questions too flippantly, but is 99% because I think the “insult and disparage people” strategy of persuasion is not very effective. I think that the white nationalist right is incorrect on a number of fact-based questions, and further that the values-based assessments they make are inconsistent, or would have different empirical effects than they expect. I will be treated their arguments as serious, effectively mainstream views to be debated on the grounds of fact, not name-calling.
3. My personal views on race, American identity, and U.S. history are emphatically in favor of racial mixing, pluralist integrationism, and blood-and-iron-Sherman-was-a-hero-Union-at-any-cost-ism. I am not treating Wasp Nationalism’s views this way as a closet apology for them, but because, fundamentally, I think that to persuade someone they are incorrect, you have to be willing to engage in herculean feats of epistemic charity.
4. I have an ulterior motive here: I am doggedly pro-natalist, as anyone who’s read my work knows. I’m proud to say I’ve had a hand in nudging the consensus political view towards pro-natalism as well. But the biggest threat to the victory of pro-natalist policy is white nationalism. Pro-natalist has been plagued throughout its history around the world by association with race-nationalist movements or other extremist visions of nationalism. I believe that association can be broken and should be broken. As such, my hope is, in this post, to explain why (1) race-specific pronatalist policy will fail, (2) more broadly pro-natalist policy may not fail, (3) pro-natalism has no necessary connection to race nationalism, and indeed can defang such movements.

How Should We Forecast Population?

Lots of people forecast population. The last official Census forecast of population is from 2014. It is very wrong, and has greatly overestimated population growth since 2014 due to overestimation of births, underestimation of mortality, overestimation of immigration, and underestimation of emigration. As such, I will not chain myself to Census estimates. Rather, I will build my own estimates entirely. I calculate out age-specific fertility rates (Total Fertility Rate) for nonhispanic whites and for “other” out to 2060. Here are my estimates:

And here’s crude birth rates:

As you can see, I forecast convergence in TFR. You can quibble with these figures, but, fear not, I will do sensitivity testing below.

I also develop group-specific mortality estimates. Here are crude mortality rates, but fear not, the underlying calculations use age-specific mortality rates forecast forward:

These estimates hold life expectancy constant at estimated 2016 levels, but, again, I’ll show different sensitivity possibilities.

I also forecast immigration and emigration. The underlying calculations forecast Immigration/Emigration for Male/Female Nonhispanic White/Other by Age. That ends up being 800 cells per year just for migration forecasts, feeding into overall migration figures. But that’s boring, so here’s crude net migration rates per year out to 2060.

Note that my 2017 net migration rate is not precisely calibrated to 2016; I could do that, but I actually do expect 2017 net migration to be substantially below 2016.

As a result of all these components, I derive the following estimate of total population, compared to Census estimates from 2014:

As best I can tell, Wasp Nation is calibrating to these 2014 forecasts, so our calculations will vary to some extent simply because I have calibrated to more updated population and change components than he has. And as you can see, this makes a big difference. Whereas Census forecasts continuing population growth at a fairly robust rate, I forecast a national population decline by the 2040s. This is part of the reason I urgently advocate for pro-natalist policies: national population decline is likely to set in within my lifetime, and I am afraid of the consequences.

We can also look at our group-specific population forecasts.

As you can see, Census and I agree on nonhispanic white (NHW) decline, though I am substantially more aggressive. Census, meanwhile, anticipates a continuing geometric growth in the other race/ethnicity group (OTH) population, whereas I expect a declining growth rate. Again, this is because I am using more updated figures than Census, and have a few more pessimistic assumptions as well. One assumption I make that Census has no calibration for is that I assume a small fraction of people born to OTH parents end up identifying as NHW. This is an empirical fact, but it is hard to estimate how prominent it is. I estimate the net shift into NHW is about 0.3–0.6% of OTH births per year, with no other shifts. This is probably on the low side of plausibility, but I will show sensitivity tests below.

So this is my basic forecast. Wasp Nation does not show a core population forecast in his tweets, so I cannot compare to him. Here you can see my estimate of NHW population share to 2060 vs. Census’ from 2014.

As you can see, we both agree on decline, we both agree whites go below 50%, but, because I have far more pessimistic assumptions about fertility convergence than Census does and because I have far more pessimistic assumptions about immigration than Census does (my assumptions more closely reflect observed post-2014 reality than theirs, FYI), I end up with 47% white share instead of 43%. As such, merely updating Census forecasts for more recent data and more realistic assumptions adds about 4% to the 2060 NHW share of the population. As you’ll see below, that makes just adding 2–3 more years of data equivalent to a large policy change in its impact.

What Impacts Might Policy Have?

Wasp Nation goes through several different policy proposals and analyzes their impact on demographic change. Some I can analyze, some I cannot. I’ll do the best I can.

Implementing the RAISE Act Would Delay White Minority

Wasp Nation suggests the RAISE Act would delay white minority because it would bias immigration towards Asians and Europeans who have lower fertility than Hispanics. He is correct that the RAISE Act would bias green card issuance towards these groups, a topic I’ve discussed in length as it pertains to religion (RAISE Act would lower Christian and Muslim immigration will increasing Hindu, Jain, and Sikh immigration, and to a lesser extent Jewish immigration).

But what about this fertility question? Well, here’s a comparison of the total fertility rate today for foreign-born Hispanics vs. foreign-born Asians.

As you can see, Wasp Nation is again correct here: foreign-born Hispanics have higher fertility than other groups, and are the only group with above-replacement-rate fertility here. Among native-born, all groups have below-replacement fertility, with Asians the lowest and Hispanics the highest. So it’s true, swapping Asian and European immigrants for Hispanic immigrants would, both long-run and short-run, result in NHW individuals having a higher overall share of births.

But that’s not all the RAISE Act does. It also reduces total inflows. And it turns out that very few groups have higher immigration than before. So we need to model a direct decrease in immigration, and a gradual decrease in “Other” fertility vs. my baseline.

But that’s not all! The RAISE Act also changes the educational composition of immigrants. It will bias toward immigrants who have a better chance of integration, which should result in reduced emigration for those groups. More broadly, it’s widely demonstrated that tightening rules for inflows also reduces outflows. So emigration should go down. Finally, the RAISE Act could have two different impacts on the likelihood of currently nonwhite persons re-identifying themselves as NHW: if it attracts more integration-prone people, the share could rise. But if it is paired with discriminatory policy that sets nonwhites apart, that ratio could fall. However, I already have that ratio very low to be conservative, so I’ll assume it rises from 0.5% in 2017 to 1.5%.

Now, when I go in to try and input these changes, I encounter a weird thing. Immigration is already way too big. Inflows into the US turn out to be *way* bigger than green card issuances. Why?

Simply put, because about half of green cards go to people already here, and the vast majority of inflows from abroad don’t need green cards! They come as students, H1-B holders, and any number of other visas. That is to say, the RAISE Act does not regulate entrance by immigrants, but rather their residence.

So that actually changes how we would calculate this. The average immigrant will have a shorter duration here alongside less likelihood of integrating. In other words, we should slightly reduce immigration for all groups, substantially increase emigration for OTH, and reduce the pace at which OTH become NHW slightly. We should slightly reduce fertility, but the effects become a lot more ambiguous when we consider the RAISE Act’s actual structure. Ultimately, I have the RAISE Act slashing total inflows by about a fifth, while outflows are raised about a fourth. Here’s total population:

As you can see, the RAISE Act possibly yields a diminishment of American population and, by extension, total economic output, military capacity, and global power. I have not included any creative dynamic factors like “maybe low immigration creates tight labor markets which pushes up wages which boosts family income which boosts fertility” because every chain of that story is extremely speculative.

In terms of the NHW share of the population, here’s the output:

As you can see, the RAISE Act probably would result in a net “whitening” of the population. I have the 2016 NHW share clocked in at about 48%, up from the 47% in my baseline, and up from the 43% in Census’ baseline.

Wasp Nation says:

Factcheck: Mostly true. The RAISE Act might plausibly delay NHW minority until as late as 2052. But the mechanism should be crystal clear: we maintain white population share by having less population, less GDP, and less global power and reach.

Deporting 1.2 Million Illegal Immigrants Would Postpone White Minority Even Further

Wasp Nation mulls over deporting 1.2 million illegal immigrants every year between 2020 and 2030, making for 12 million deportations:

He says that this, combined with the RAISE Act, keeps the US white-majority out to 2060. Is he right?

Well, he ignores the possibility of higher outflows of non-deportees due to deportations; I’ll include it: deportee family members likely accompany. Furthermore, such a wave of deportations almost certainly reduces immigration as well. It would also almost certainly harden the lines between NHW and OTH, reducing integration further. Meanwhile, such mass deportations would almost certainly be socially disruptive, so we’ll put in slight penalties for mortality and fertility.

Here’s total population:

As you can see, aggressive deportations would meaningfully reduce national population, indeed even more than the RAISE Act would. Surprise surprise, yanking 12 million people out of their houses has a demographic impact. Again, this is a further decrease in long-run GDP, national strength, and global power: without manpower, there is no military.

But when we turn to the white share of the population, the impact is striking:

As you can see, deportations have a very big impact on the white population share. It basically stabilizes through the 2020s, and then declines, reaching about 51.5% in 2060, so higher that Wasp Nationalism suggested.

But again, think about how we did this: population growth was suppressed. This means persistent economic headwinds for the United States. The cost of maintaining the white population share is, to some extent, once again, lost national greatness and power.

Letting in More White Immigrants Would Boost Whites Even More

Next up, Wasp Nation suggests that we absorb 100% of the white population of South Africa.

I’m not kidding.

Let’s make a first sanity check. Say we implemented a policy announcing that we would accept any white South African who wanted to come to the US, how many would come? Would we get 100% of them?

Well, consider that when the areas around Syria had open doors to Syrian asylum seekers and those peoples’ homes were literally being destroyed, less than 1/4 of the population has fled the country after several years. So to be blunt: the idea that open immigration policy would enable a massive flood of white South Africans is nonsense.

Let’s assume something more plausible: an open-door policy for any white residing in Africa yields something like 100,000–300,000 more NHW immigrants per year during the 2020s. That will boost immigration and slightly boost emigration. I assume this is on top of the RAISE Act caps as well. This has no impact on immigrant integration, and I assume no impact on mortality. But what about fertility? What is the fertility rate among white South Africans? Using the 2011 Census of South Africa, we can check! And it turns out that it is about 1.5… so incredibly low. In other words, adding white South Africans to the US will slightly reduce NHW fertility slightly.

When we plug this all in, here’s what we get for population:

The biggest imaginable new white inflows I can cram in result in… barely any difference in long-run population. The reasons are several, but boil down to these new immigrants having critically low fertility, being dumped into an already-low-fertility population, and, furthermore, some of them are offset by higher outflows.

What happens to the white population share?

I won’t show it as a graph because it’s boring: it nudges ever so slightly upwards, but no big change.

Instituting Racial Quotas Would Collapse Population

The next proposal is to simply reinstate racial/national immigration quotas.

So, okay, this is no longer additive. We are swapping out the RAISE Act for a quota law which requires 80% NHW immigrants. Now, this is a tricky one to model. Will we get to 80% White immigrants by reducing nonwhite inflows, or boosting white inflows? Boosting white inflows seems hard… so we are gonna do this 75% by reducing nonwhite inflows. This will reduce fertility a bit more, because the USA has one of the highest Non-Hispanic-White fertility rates in the world. But by shifting the composition of immigrants and restricting big-sending-countries, quotas might boost the integration rate. I’ll set it at 1%.

Total population is reduced still further, following that low green line. By 2060, the U.S. has barely over 300 million people. We are deep in the throws of a half-century-long secular decline in demand for goods and services, at least half of all metro areas have stagnant or negative population growth, and politics have become quite literally zero-sum. Yay us.

But at least the white share is high!

Lookee there! Nearly a 55% white share! And all we had to do to get it was deport 12 million people, throw the floodgates open for white African immigrants without any vetting or restriction, tank all recruitment of students or workers from outside of Europe and Australia, and shrink out population by about 8–10%! And now that we’re enduring a century of decline, a rapidly rising median age, and an inability to recruit soldiers, China has annexed Taiwan and established an ASEAN “protectorate,” even as Russia has re-established its sphere of influence over the Baltics and Ukraine. Oh, and the fastest-rising powers in the world are all African countries. When Ethiopia moves to annex Eritrea in 2057, igniting a regional conflict that kills nearly a million people, the United States can do nothing but look on helplessly, as we our deployable military has shrunk, and our flagging economy can support only 2 carrier groups at a time.

This is fantasy, of course. But so are the policies we’re discussing. And the point is, the absolute size of the American demographic, economic, and tax base has global consequences. If you want liberty defended from her many foes, you need a growing population. Consider the victory of the Union in the civil war. Was it strategy? No. Leadership? Ha! Technology? Hardly. Superior logistics? Somewhat. Superior industrial base? That helped.

But in the end you know what it was?

Manpower.

There will be a Great Power conflict in this century. I know my more liberal readers are rolling their eyes here; but that’s fine, you’re already on the same page as me re: the badness of immigration restrictions. But for my more rightist leaders, consider your own worldview. With a declining America, how long before the fundamental anarchy of the international arena rears its ugly head?

Luckily, Wasp Nationalist has considered this issue.

Increasing Fertility for Whites Can Whiten the Country

But it is not possible.

This is the grand plan Wasp Nationalism offers:

Double. Fertility. Oh boy.

I’ve done some thinking about fertility politics. This is sort of my jam. And let me tell you: not under any sane policy regime does fertility double. The biggest fertility jump in history was in Ceaușescu’s Romania in 1967, which increased fertility from 1.9 kids per woman to 3.7. That is almost a doubling. But it only lasted for about 2 years, then declined steadily. And of course, that required a totalitarian state able to absolutely control the availability of abortion, contraceptives, etc. That’s not only, well, evil, but also impossible in the United States.

Look, even using policy levers to get the U.S. to just replacement-rate fertility would be hard. Getting us to a new baby boom would be… well it’s unimaginable and, as you saw above, I can imagine a lot of crazy stuff!

But let’s play this game anyways. Let’s say we shell out the $250-$950 billion per year needed to get a big fertility boom. What are some facts we know?

1. It’s easier to increase fertility among poorer people
2. It’s easier to increase fertility among people with higher desired fertility
3. Minorities in the US are poorer and have higher desired fertility

So if we launched a big pro-natal push, it would almost certainly have a bigger impact on minorities than on whites!

So you’d have to have an explicitly racist policy: subsidies for whites, not for others. Okay. But would it work?

Well, Hispanics can easily “become white.” So can multiracial people. Race in the US today is self-identified with relatively little policing. Implementing a Non-Hispanic-White policy would require an aggressive expansion of the administrative, bureaucratic state, taking blood samples of private citizens and storing the data on government servers for law enforcement purposes. If you’re okay with that, then you’re not paranoid enough about the government.

But let’s say we’re okay with that; what then?

Well, whites have lower desired fertility and higher income, and rapidly rising parent-age mortality, so you’d need much bigger per-child subsidies to engineer a baby boom. In other words, it’d be more expensive.

So here’s what TFR looks like after all of those other policies above had their impact:

Let’s say we abandon one of our carrier battle groups, shrink our airforce, and redirect money away from taking care of old people, and so we drop $850 billion per year on promoting childbearing among white people. With a plausible set of elasticities, what effect might that have? Well, using the same elasticities I used for this study, my guess is that $850 billion per year boosts white fertility somewhere between 0.35 and 0.75 children per woman. So let’s have that impact phase in.

Let’s make other reasonable assumptions: sharp rises in fertility create some echo-effect on mortality. Sharp rises in fertility among whites will also partially bleed over through mixed-race couples or mis-reporting couples, or peer effects. Special benefits for whites probably also increase re-identification as white.

Here’s our new population chart:

So big birth incentives result in population stability. 2060 population ends up being comparable to what it was with just the RAISE Act implemented; so pronatal policy mostly offset the deportations and immigration quotas.

The result is that white share stabilizes by the late 2050s.

Mission accomplished for white nationalists! We’ve got a policy blueprint!

Let’s review how we got here.

To stabilize the white population share by the late 2050s (still lower white share than today!), would take:

1. The implementation of a strict racial quota system for immigrants, requiring 80% of inflows to be nonhispanic white. This would cripple the United States’ biggest export industry, higher education, and would also cripple many other businesses, resulting in a service-led recession.
2. The deportation of 100% of the illegal immigrant population, or about 12 million people over 10 years. This would cripple numerous U.S. industries in the short run, create massive demands on the foster care system, provoke international outcry and response, and almost immediately shrink the economic base of the United States, resulting in a recession.
3. $850 billion per year in pro-natal policies for whites. The entire military budget is just $600 billion per year, so financing this from discretionary money would require that we become a pacifist nation. Even if we cut 100% of discretionary spending on the environment, international affairs, foreign aid, science, ag, transportation, education, etc, it would raise less than half the necessary funds. The entire Medicare budget is about $980 billion; Social Security is $1.25 trillion. In other words, this pro-natal policy would be the 3rd biggest line item in the United States budget, and require raising taxes by 15–25%. This would, wait for it, almost certainly trigger a recession.

So in exchange for a smaller nation with less global influence, weakened security, decades of recessions and secular stagnation, we can get… flat population and a flat white share of the population.

Maybe that’s worth it to you. But it isn’t to me. As such, rather than promoting a white nation, I consider it efficient (and also, of course, moral) to promote policy that is integrative. A steady clip of deportations is useful for many reasons for encouraging immigrant integration. As is a steady inflow of refugees. As is a policy of requiring that immigrants learn English. As is the guarantee that immigrants will be free from harassment, legal or otherwise. I’m well aware that this view of positive, pluralist nationalism is as anathema to extremists on the right as on the left. But, if we don’t go this route, things will end badly, one way or another.

Check out my Podcast about the history of American migration.

If you like this post and want to see more research like it, I’d love for you to share it on Twitter or Facebook. Or, just as valuable for me, you can click the recommend button at the bottom of the page. Thanks!

Follow me on Twitter to keep up with what I’m writing and reading. Follow my Medium Collection at In a State of Migration if you want updates when I write new posts. And if you’re writing about migration too, feel free to submit a post to the collection!

I’m a native of Wilmore, Kentucky, a graduate of Transylvania University, and also the George Washington University’s Elliott School. My real job is as an economist at USDA’s Foreign Agricultural Service, where I analyze and forecast cotton market conditions. I’m married to a kickass Kentucky woman named Ruth.

My posts are not endorsed by and do not in any way represent the opinions of the United States government or any branch, department, agency, or division of it. My writing represents exclusively my own opinions. I did not receive any financial support or remuneration from any party for this research.