Does the South beat out Chicago because of better weather?

Here is a quick summary of research looking at how weather might affect population changes in Chicago and similarly cold-weather places versus the warmer weather of the South:

Renn pointed me towards the work of Edward Glaeser, the Harvard economist and U. of C. grad who’s become one of the most prominent analysts of the American city in the 21st century. And he thinks there’s a strong correlation. He’s got a whole paper on it, in fact: “Smart Growth: Education, Skilled Workers, & the Future of Cold Weather Cities.”

Cities with average January temperatures under 30 degrees Fahrenheit grew in population only one-third as quickly from 1960 to 1990 as did cities with average January temperatures above 50 degrees. The shift of population toward the Sun Belt can also be seen at the state level: while population in the colder 25 states grew 95 percent between 1920 and 1980, the warmer 25 states saw their average population grow 309 percent…

In short, there’s something of a chicken-and-egg question with the air-conditioning solution that Glaeser cites. Adoption of air conditioning, across the South, was slower than you might expect from the weather. Its availability is a well-established boon to the South, but so is being able to power and afford air conditioning.

It’s significant that Enrico Moretti, like Glaeser an economist interested in how knowledge workers cluster in cities and regions, has most recently turned his focus back to the Tennessee Valley Authority. That’s a government project so ingrained in Southern culture that, as a kid, I thought that the Tennessee River was just called the TVA (emphasis mine):

We find that the TVA’s direct productivity effects were substantial. The investments in productive infrastructure resulted in a large increase in local manufacturing productivity, which in turn led to a 0.3% increase in national manufacturing productivity. By contrast, the indirect effects of the TVA on manufacturing productivity were limited. While we do find strong evidence of localized agglomeration economies in the manufacturing sector, our empirical analysis clearly points to a constant agglomeration elasticity. When the elasticity of agglomeration is the same everywhere in the country, spatially reallocating economic activity has no aggregate effects, as the benefits in the areas that gain activity are identical to the costs in areas that lose it. Thus, we estimate that the spillovers in the TVA region were fully offset by the losses in the rest of the country.

The intensely regressive economic (and cultural) practices of the South damned up potential across its old borders; once they began to fall, it created a flood, draining Yankee knowledge, technology, and workers.

While everyone wants to talk about the weather, it isn’t the only factor nor the most important factor in population and economic growth. To suggest this is the case is to rely on strong ecological arguments, perhaps like those made by Jared Diamond in his more popular books. Yes, air conditioning matters but humans were able to live in both the warmer South and colder North before air conditioning or central heating. More broadly, factors like electricity and water (see the recent troubles in the Southwest) matter more and are essential to even having air conditioning in the first place. Thus, the twist of invoking the TVA, an important adaptation to nature, makes the matter all the much more complex: essential infrastructure makes all sorts of other things possible.

Some big cities only made possible by air conditioning?

This seems pertinent with the recent heatwave in the Midwest and East Coast: how many of the major cities of the world wouldn’t exist without air conditioning?

It wasn’t until the beginning of World War II that homes in southern U.S. cities began using air conditioning units. By 1955, one in every 22 American homes had air conditioning. In the South, that number was about 1 in 10, according to the historian Raymond Arsenault [PDF]. Since this increase in air conditioning use, many of these Southern cities experienced a population boom.

I took a look at the metro areas in the U.S. with more than 1 million people and found which have historically been the hottest, based on the number of cooling degree days per year — a statistic used to measure how much and how many days the outside temperature in a certain location is above 65 degrees. Using numbers from NOAA, I found that between 1971-2000, six big cities in the South had an average of at least 3,000 cooling degree days. I also compared the 1940 metro population (when available) to the metro population in 2010. From the time just before air conditioning became popular in the South to today, population growth in the region has skyrocketed. This raises the question: would these hot Southern cities be around, at least in their present form, if air conditioning hadn’t been invented?

But, of course, there are bigger, hotter cities across the globe. In fact, seven of the largest metros in the world have an average high temperature above 90 degrees Fahrenheit.

Not surprisingly, all of these cities are found in developing countries. As Michael Sivak, a professor at the University of Michigan notes, only two of the warmest 30 global metros can be found in developed countries. With the middle class growing in warm metros in countries like India, demand for air conditioning is increasing. A recent New York Times article reported that sales of air conditioning units in India and China are growing 20 percent per year and are fast becoming a middle-class status symbol. Last year, 55 percent of new air conditioners were sold in the Asia Pacific region.

Is there some sort of giant control group we could use to figure this out? Over the weekend, I was in a 150 year old church with no air conditioning. It was hot though I think this was primarily because there was no air movement; indeed, when we walked outside afterward, it felt more pleasant as there was a slight breeze. Before air conditioning, people obviously survived in such temperatures (and also survived in the winters without central heating as we know it today).

So this seems to be the real question: could we expect that there would be major changes in population distributions if there was no air conditioning whatsoever? Would Florida really have few people and post-World War II Sunbelt expansion not taken place? The best solution to all of this would be to have people move to more temperate climates where it doesn’t get too hot in the summer or too cold in the winter. This generally requires consistent breezes, usually off major bodies of water. Of course, not everyone can live in places like Hawaii which only has a record high temperature of 100 degrees. Did the Mediterranean climate help give rise to empires like Greece and Rome (though it makes it difficult to then explain the Sumerian, Assyrian, Babylonian, and Persian empires which must have adapted to desert climates)?

More broadly, we could discuss the influence of ecology on population growth and state building. I remember studying the mysterious decline of the Maya in southeastern Mexico/Guatemala. More recent scholars have suggested some kind of ecological explanation, perhaps a drought, that led to increased contentious competition for dwindling resources.