A Brief Population History of Scandinavia
Because The Data Exists
I’ve written a lot about the Nordic countries, and whether they are comparable to the U.S. But as I wrote those pieces, I realized that I had a large gap in my actual knowledge about Nordic population history. So I recently decided to take the first step in correcting that gap, and explore what data was available. And lo and behold, Nordic statistical agencies produce really easy-to-use time-series formatted spreadsheets of all their national vital statistics and migration records. Huzzah!
This post won’t have a big takeaway in terms of policy. Rather, it’s just some information I collected that I thought other people might enjoy. So, with that, let’s dive into the Nordic countries!
Population Growth
We’ll start with population growth. Below, I’ll show the 10-year cumulative population growth rate for the U.S., Norway, Sweden, Finland, Iceland, and Denmark, as well as for all the Nordics combined, back to 1700. The U.S. population estimates are my own imputations and calculations from Census data.
The chart above shows that U.S. growth rates from 1700 until 1860 were well above Nordic levels, but began to slide until about 1940, when U.S. growth converged with the Nordics. However, in the post-WWII period, U.S. growth has generally exceeded Nordic growth, with the exception of Iceland. Interestingly, however, Nordic growth has accelerated in recent years even as U.S. growth has slumped, so that now U.S. and Nordic growth have more-or-less converged, with several Nordic countries having greater recent growth rates than the U.S.
But, of course, scale matters too. In terms of raw growth numbers, the last time any Nordic country added more people than the US in a given year was, well, never; at least, not since 1700. Combining all the Nordics together, the last time their combined growth exceeded U.S. growth was also, er, never, at least not since 1770 when the combined Nordic data can be calculated.
We can also look at raw population totals. Of course the U.S. is vastly bigger now, but when did we become bigger, in terms of population?
By my estimates, we were already more populous than Norway, Iceland, Finland, or Denmark when the earliest data on those countries is available. Indeed, I actually have one data point for Iceland showing 1703 population of 50,358: versus U.S. of 286,146. So U.S. population surpassed most Nordics sometime before 1750. I’ll admit that this finding shocked me. I would never in the world have guessed than in 1735, there were already more Americans than Norwegians. But chalk this up to the bad pre-Revolutionary history kids get. People often forget that more time elapsed between Christopher Columbus’ discoveries and the American Revolution than between the American Revolution and today. The Americas had a long time to get populous.
The biggest Nordic country, Sweden, was surpassed in the 1760s sometime and, by 1800, there were more Western settlers in the United States than there were people in all of Scandinavia. I specify “Western” here because my estimates exclude the native American population.
So that’s population. Let’s move next to births and deaths!
Let There Be Babies
My opinion on fertility is well-known. We should have more babies. But how many babies have we had in the past? This question turns out to be surprisingly difficult in the U.S. Universal collection and assembly of birth records didn’t occur until the 1940s, and even partial collection didn’t really start until the 1900s at the Census-level. So our time series here is really short, sadly. We do have a few decennial Censuses that reported births, but nothing compared to the Nordic records.
Why are Nordic birth records so historically extensive? Oh I’m so glad you asked. Because the answer is Lutheranism and guys I am not joking here.
To this very day, in most Nordic countries, the Lutheran churches, which means the historic established state churches, report all baptismal and funeral records to the government. They have done so since the 1700s. In a society where 99%+ of the population was all Lutheran in the state church, this mean that the government basically had a complete Census of birth information, as well as even family connection information. This data has become invaluable for modern researchers, but I want to note again how it came about: Lutheranism! Or, rather, religious uniformity and a close church-state relationship. In other words, a certain unique institutional-demographic nexus created policy and governance options that are not available to other countries, such as those with less religious uniformity, no established church, or even an established church that doesn’t practice infant baptism. But since Lutheranism of the past was (1) amenable to co-optation by the state, (2) suuuuuuper into infant baptism, like it’s a very big deal to us, (3) nearly-uniformly shared in Nordic countries, we get good data.
So. Birth rates:
First of all, note those U.S. birth rates early on. They’re crazy high, and way above Nordic birth rates. But even after the U.S. largely converges, we nonetheless remained moderately above most of the Nordics, and remain so today. Around the 1900s, Finland was a bit above us, and you’ll note that Iceland has had a higher birth rate for a long time. But for all Nordics combined, we are substantially higher.
Death in the Northlands
Having done births, obviously, we’ll do deaths next. So, here goes:
I know full-page images are a little obnoxious but this one needed it, because otherwise it’s pretty busy.
First of all, notice that dark red dot at 1850. That’s the U.S. for the first Census I could find that enumerated mortality. As you can see, we had a very low death rate compared to the Nordics.
Second, notice all the spikes. Those spikes correspond to exciting events in Nordic early-modern history like famines, devastating months-long impassable snowdrifts, volcanic eruptions, and all manner of nasty occurrences. Keep in mind, that 1850 mortality rate for the U.S. includes the mortality impact of brutal slavery in the south. Nordic climate has a larger country-level mortality impact in the pre-modern period than U.S. region- and race-based slavery (though of course slavery probably has a worse impact on the enslaved people).
But by the mid-1900s, US and Nordic mortality was quite similar, and remained so until the 1960s. Since then, the U.S. has had lower mortality than the Nordics, except for Iceland, thanks to higher immigration and a younger population on average.
About Those Immigrants…
Ah yes, immigrants! About them. Below I present immigration data for the Nordics and the U.S. The Nordic data is all quite straightforward, just what their stats agencies report. The U.S. data is far more artisanal. In order to make it “immigration” in the population-balance-sheet sense, I must include all inflows-resulting-in-residence as immigrants. Thus, I include legal immigrants, but also estimates of illegal inflows. I also include estimates of enslaved migration as immigrants. If you don’t like that, check your freakin’ priors about what it means to be an “immigrant.” It means moving across borders, generally establishing residency. It doesn’t mean voluntary.
Now in this case, my U.S. time series is actually the oldest one, which is nice. Let’s note a few things:
- Current U.S. immigration rates are probably a bit on the low end, historically speaking.
- Current U.S. immigration rates are lower than current Nordic immigration rates; much lower, with only Finland even comparable.
- Peak U.S. international inflows came in 1945. Obviously, that is demobilization. These immigrants don’t show up on any immigrant roster; but any population balance sheet of the U.S. has to count them, because otherwise explaining the sudden jump up in population in 1945–1947 is really, really hard.
- Immigration into the U.S. is falling, even as Nordic immigration is rising.
- Nordic immigration is at historically totally unprecedented levels.
I won’t spend time exploring the implications of these data features. Consider yourself what they may mean.
And About Those Emigrants…
The Nordic countries all pretty aggressively track emigration as well, which is super cool. The U.S. does not formally track emigration. I define emigration as simply the residual growth of my other 3 components. So any error in the data is lumped into emigration. Treat these numbers with a grain of salt.
Here you’ll notice something fairly remarkable. The late 1800s saw a wave of Scandinavian migration to the United States, which should show up as high emigration. And yet, even for specific Nordic countries where the data is complete, Nordic emigration rates are no more severe than U.S. emigration rates were in the 1900s and 1910s. How can this possibly be true!?
Well, a few things to take note of. First of all, the “Nordics” line reflects reported emigration from only countries reporting at the time, but divides by population for all Nordics. Especially if Iceland, Denmark, or Finland had equally- or more-severe outflows as Sweden or Norway, which seems entirely plausible, the all-Nordics line could be higher.
Next, my U.S. emigration estimate is an imputation from flawed data. If I’ve over-estimated immigration or births, then emigration will be over-estimated as well. If I’ve underestimated deaths, then emigration will be over-estimated. Death-reporting was incomplete in the early 1900s; I have tried to correct for this as best as I am able, but it’s possible I am nonetheless under-reporting deaths.
So the Nordic line may be higher and the U.S. emigration line may be lower.
But the truth is, the U.S. did have lots of emigration. In some periods, scholars have estimated that, for certain groups, emigration may have offset more than 50% of immigration: many immigrants ended up not stickin’ around! So high immigration actually should yield high emigration.
But to see how this all really shakes out, we need to look at net flows.
Here it becomes clear what’s going on. Yes, the U.S. had some high gross emigration: but gross immigration was even higher! Meanwhile, the Nordics had large outflows with few or no inflows during the 1800s. For most of the post-war period, the U.S. has had positive net migration, with just a few exceptions. In most years, we had net migration rates well in excess of the Nordics. However, that hasn’t been true in recent years as U.S. net migration has fallen and Nordic net migration has hit unprecedented highs.
I also want to point out some episodes of negative net U.S. migration. First of all, we can see WWI and WWII deployments very clearly. But we can also see emigration in the 1930s. U.S. immigration laws cut off new inflows, but they didn’t cut off outflows, with the result that the U.S. had a migrant-leakage and actually faced net negative international migration. We can also see a sharp move downwards in net migration around Korean War deployment times.
In recent years, we can see a spike in Nordic net migration. That spike is driven by crisis migrants from the Middle East and Africa, but also by migrants from the new eastern EU states. Crucially, we can see that the Nordics have had persistently positive net migration rates for decades. No welfare-death-spiral in sight here.
What Is a Border, Really?
However, the Nordics cheat on migration data. If a person moves from Sweden to Finland, they show up as an immigrant for Finland, an emigrant for Sweden. Yes, the Nordic-wide net migration stat controls for this effect, but gross migration rates remain inflated due to intra-Nordic migrations comparable in distance to U.S. interstate migration.
But this is a classic unit-of-analysis problem in migration. The smaller you define your units, the higher the typical rate of migration across unit-boundaries. For example, we also have internal migration data for the Nordics. Here’s internal migration data for the U.S. and for the Nordics, as well as international migration data, grouped by size of the unit being assessed:
Here’s another look at the scale of different regional definitions:
As you can see, Nordic regions don’t have good comparators to U.S. regions. US counties are 2–8 times the scale of Nordic municipalities, but just 1/3 to 1/20th the size of Nordic counties or regions.
Luckily, Nordic countries and U.S. states are similar scales. So we can compare U.S. gross interstate migration to Nordic migration. But here’s the trick: our Nordic data includes movements among Nordics, and from outside, and it double-counts Nordic moves. So to get a comparable U.S. estimate, we need to take U.S. interstate migration, double it, then add in international migration.
The result is this:
Viewed on a similar population-geographic scale, Americans have historically been vastly more likely to have recently experienced significant residential mobility than Nordic residences. However, even as migration has become more and more prevalent for the Nordics, it has become less and less so for the United States. The U.S. migration-propensity-premium is eroding when we compare similarly-scaled units with similarly-defined terms.
But what if we ignore the basic problem of unit-scale, and just look at gross rates? Yeah, let’s do that, it’ll be fun!
First of all, let me say, kudos to Finland! They’ve been consistently tracking the same annual estimator of domestic migration since 1878. That’s kind of incredible.
Second of all, hah, yeah, I know, you can’t read what anything is. That’s intentional. There are tons of different estimators; too many for me to really address in detail. Instead, I’ve highlighted for each Nordic country the indicator most similar to U.S. intercounty migration in terms of the scale of the geographic unit being measured. So now, I’ll do a second graph, of just those lines:
That should be a bit more readable. It should also look a bit different; U.S. migration is estimated higher. That’s because I combined interstate and intercounty migration here, to get all cross-county migration in one estimator, to compare to approximately similar estimators for other countries.
As you can see, some countries have experienced similar migration trends as the U.S. Iceland’s migration has fallen, and the very short time series for comparable Denmark migration also suggests a similar trend. If you sort of tilt your head and squint, you can also see some recent resonance between CPS migration estimates and Finnish migration estimates.
But Sweden and Norway both show rising migration! That’s really interesting! To the extent that declining-migration explanations are about technological or economic trends shared with the Nordics, this different trend is genuinely remarkable. I should note that, for Sweden, the large-region migration rate, intercounty, has been more stable, though still up slightly since the 1980s. For Norway, migration rates have been stable for the largest regions, but have risen somewhat for a slightly smaller region.
However you cut it though, the U.S. went from having a CPS-measured migration rate much higher than the Nordics, to a CPS-measured migration rate about even with Nordic migration, even on the low-ish side. I specify CPS because, if you use ACS or older IRS estimates, migration remains on the high-end vs. the Nordics, though the most recent methodology changes for IRS seem to put the IRS estimator nearer CPS than ACS.
Conclusion
As I said at the beginning, I don’t have a huge takeaway point here. I just wanted to explore the data. I leave to others the task of saying what much of it means, but I do want to note the change in migration. The U.S.’ historic position as a high-internal-and-external migration country is eroding. Immigration and domestic migration are declining in the U.S. even as they continue to rise in some other countries. Theories that presume to explain the decline in U.S. migration by appeal to broader economic forces should seriously contend with the experiences of countries where migration continues to rise, and try to explain why their experience is different.
Nonetheless, by the best metrics, such as a harmonized Nordic/US interstate/International migration estimate for consistent geographic units, or such as ACS-measured gross migration, it seems that the U.S. remains a relatively high-migration country, just not nearly as much as it used to be.
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.
