They are Doomed!!!

I've been exposed to an ever increasing stream of hyperbolic pontification regarding European demographics. I'm not sure if it's because there's actually an increase in the amount of alarmist writing on the subject or I'm just being exposed to it more often. Fine, fine examples include:

The fertility bust: Very low birth rates in Europe may be here to stay; and
Is “Old Europe” Doomed?

Doomed, I tell you, Doomed!!! The ideas being put forth include the following:

1. The native populations of Europe (and Japan) are shrinking;
2. Those populations are shrinking because they aren't very religious; and
3. The trend will continue until they shrink away to nothing.

Hmmmm. Where have I seen this argument before. Oh wait, I remember, except it was exactly backwards (except for the religion part). When Paul Ehrlich and the population bombsters (would be a good name for a rock band, don't ya think?) predicted doom for the planet a few decades back, they based their prediction on population growth trends that had been trending for thousands of years. The bombsters turned out to be somewhere between wrong and absurdly wrong. These new pundits are making predictions based on trends that are a few decades old. I think that the idea that these trends will continue long term is likely to turn out to be as misguided as the bombsters' assumption.

Birthrates in the developed world are strongly correlated with population density. As an example, the following chart shows some of the more populated countries in the developed world, their population density, and their birthrates (defined here as births per year per 1000 people). These numbers are all derived from the CIA's world factbook.

The correlation coefficient for this data is -.73 which gives a R² of over 0.5 which means that over half the variance in birth rate is explained by the variance in population density.

So how important is religion when it comes to birth rates? This data doesn't say anything about it directly, only that it at most explains less than half of the variance in birth rates. If one thought religion was important for predicting birthrates, one might guess that the more religious countries would fall above the yellow trendline (in other words, even more births than predicted for the level of population density). For sure, the United States is religious and falls above the trendline, but I'd be surprised if France and Japan, which fall above the trendline, are really all that much more religious than Catholic Spain and Italy or even Protestant Germany. So I'm currently pretty skeptical about the correlation of religion and birth rates.

So why are birthrates low in Europe (and not so high in the United States either)? Well, I don't know, but here's an idea. A population adjusts its fertility rates over time to push the its size towards an equilibrium point. As the population grows and gets more crowded, birth rates fall. As the population falls and more space becomes available, birth rates increase. Other mammals, such as coyotes, vary their litter sizes hugely to achieve exactly this effect. So why not humans?

So how did the populations of Europe and Japan get so large that it's causing birthrates to fall? Again, I'm not sure, but it may be that the rapid success in raising life expectency caused overcrowding and now the people in Europe and Japan are adapting. The basic concept is that if you double life expectency, you also double the population and population density in short order, all else being equal. As an illustration, I've run a population model that's illustrated in the following graph:

The population (shown on the y-axis) in the model is steady at 1 million people for the first 25 years. In other words, the birth rate and death rate are in equilibrium during that period.

At 25 years the life expectency is instantaneously doubled. While nothing happens instantaneously in the real world, we did defeat disease and hunger amazingly rapidly during the 19th and 20th centuries. In this simple model, doubling the life expectency basically means nobody dies for a while.

The population soars and the birth rate begins to fall. Eventually, people start dying again (but at a slower rate since they live twice as long) and the population plummets as the system tries to get back to an equilibrium of 1 million people. As is common in real world systems, the population actually overshoots the other way and the population drops below 1 million, before the birth rates increase enough to stop the fall in population size. Over the next couple of hundred years the population oscillates about and slowly converges towards the equilibrium population of 1 million.

Europe and Japan may simply be in the middle of this sort of change in population. In the above graph, Europe and Japan might be at the equivalent point of somewhere around year 50. I can believe that it's possible that their populations may decline for a lot of years and they may end up with a far smaller population than they have now. But given how crowded Europe and Japan are, that may be a good thing in the long run.

At any rate, I don't think they are doomed. Unless they are doomed to get ever richer at a somewhat slower pace than other countries. Doomed to a comfortable life, riding on the coat tails of the hard working and innovative Americans. Doomed to having their security provided by others. Doomed to eat cheese and drink fine wine during ever increasing leisure. Doomed, I tell you, DOOOOOMMMED!