Metropolitan Math is No Myth

Apr 29, 2020

In an opinion piece appearing last year at the site Market Watch, George C. Edwards III (a professor of political science at Texas A&M and author of Why the Electoral College is Bad for America) purports to debunk one of the “myths” that supporters of the Electoral College often rely on. A closer examination, however, suggests that the “myth” actually stands up quite well.

According to Edwards, the “biggest argument for keeping the Electoral College” is that it “protects rural states against the dominance of large states and big cities.”

I’d certainly quibble that this is the “biggest argument” for retaining the state-by-state election of the president, though it’s certainly true that the Electoral College does provide some advantages for less-populated states as well as residents of rural areas in both large and small states. But the fundamental argument he seeks to make, which appears to be that the population of big cities and large states is not sufficient to dominate in any system of directly electing the president, is hard to sustain once all the facts are considered.

Take, for example, the following assertion, towards the end of the piece:

The principal reason for cities’ lack of sway is their size. They are simply too small to wield control of politics. The combined population of the 20 largest cities in the U.S. in 2016 was 34.1 million out of a total population of 323.1 million, representing only 11% of the population.

The problem here is that Edwards adopts something of a narrow (though technically accurate) definition of the term “cities,” counting only those who live within the city limits and ignoring the suburbs that comprise the greater metropolitan area. The argument made by myself and others isn’t that candidates won’t campaign outside the city limits under a direct election system, it’s that candidates will spend the vast majority of their time in (and also focus their policies on the interests of) the major metropolitan areas of the country. And when you look at the number of Americans living in major metropolitan areas, the population soars well beyond what Edwards claims.

Using U.S. Census data, the three largest metropolitan areas alone (New York, Los Angeles, and Chicago) have a combined population of nearly 43 million people. The twenty largest combined metropolitan areas have a total of about 125 million people, and expanding that to include every metropolitan area with 1 million or more people (there are 53 such areas in the U.S.) adds up to almost 184 million people.

I’m not suggesting that, under the National Popular Vote Interstate Compact (NPV) or some other form of direct elections the candidates would only spend time and money (and policy attention) in those 53 one-million plus population centers. At a certain point, a smart campaign manager is going to realize that there are also lots of votes to be had in the 54th-largest metro area in the country (Fresno, CA, pop. 994,400), and even the 62nd (Bakersfield, CA, pop. 896,764) 67th (Oxnard-Thousand Oaks-Ventura, CA, pop. 850, 967) 77th (Stockton-Lodi, CA, pop. 752,660), and 101st (Modesto, CA, pop. 549,815) largest metropolitan areas in the country.

All of them, conveniently enough, in a single state that is also home to the second (Los Angeles), twelfth (San Francisco), thirteenth (Riverside), seventeenth (San Diego), twenty-sixth (Sacramento), and thirty-fifth (San Jose) largest metropolitan areas in the country. In fact, all told there are about 223 million Americans in metropolitan areas of half a million people or more.

The point is obvious – candidates will focus on large metropolitan areas because that’s where the most votes are and there are significant economies of scale when it comes to voters who live in densely-populated areas. Under the Electoral College as it presently operates, on the other hand, candidates have to try to win entire states, which means visits to places much further down on the list of population centers, and with much greater connection to rural and exurban voters.

In 2016, for example, presidential campaign visits in Pennsylvania included Gettysburg (about 7,700 people in the “city” and 102,000 in the whole county), Altoona (46,000 or so people in the city, 127,000 in the county) and Johnstown (about 21,000 people, 143,000 in the county). And, of course, plenty of visits to the Philadelphia and Pittsburgh areas. A quick glance at other states such as Maine, Michigan, Ohio, and Wisconsin also showed visits to areas that at least give rural and exurban voters a chance to have presidential candidates pay attention to their issues.

There’s no system for electing the President that would force candidates to literally give equal weight and attention to every voter – decisions have to be made on how to allocate scarce time and funds, and which issues to focus on. Right now, the state-by-state election process ensures that urban and suburban voters get plenty of attention, but also helps to ensure that candidates need to take into account the interests of those outside of the major metropolitan areas.

As I explain to people as a shorthand way of making this point: It’s a good thing that the 2020 Democratic candidate will almost certainly be visiting Wisconsin, and not just Milwaukee and perhaps Madison. Smaller communities like Appleton, Beloit, Chilton, La Crosse, Sheboygan, that are more closely connected to rural residents and are outside of major metropolitan areas, are among the sorts of communities that are likely to pop up on the 2020 presidential campaign travel itineraries in not only Wisconsin but in other states as well because of the Electoral College. NPV or some other form of direct election would almost certainly end or sharply reduce this, and it would be a shame to leave these voters behind.