How Mega-Regions Would Change How We Vote
Shortly after the election, this Mega-Regions map made the rounds on the Twitter-sphere, which used commuter sheds to illustrate the extent to which the US’s geography is organized around cities. In this past election, the Electoral College itself incidentally revealed how arbitrary and outdated state boundaries are. For starters, flipping just three border counties to neighboring states would have flipped the election result. States come pre-gerrymandered. As something of a curiosity, I wanted to test how redrawing state lines around these Mega-Regions could have changed the outcome of the election, if at all.
It’s very much not novel to point out that Democratic voters are concentrated to urban centers — Democrats rely upon their ability to run up the score in urban areas, and when Democrats win states, it is because cities turn out to vote. However, most of the maps I’ve seen on the geographic sorting of political parties fail to communicate just how significant the urban areas are in terms of pure sum of voters. I made the following map to show where margins of victory were especially concentrated relative to the national vote count.
In the map, the margin of victory in each county is divided by the total number of votes nationally; while Trump won many counties by 50 points, many of those high-margin counties have fewer than 10,000 votes cast. By contrast, Hillary Clinton winning LA County by 55 points reflects a surplus of 1.27 million votes, reflected by the darker blue. This is why Hillary Clinton can garner nearly 3 million votes more than Trump, and it still appears that only a small portion of the country voted for her.
Cities are woefully, even comically undervalued by the Electoral College (to say nothing of voter suppression that targets people living in cities). And because voter geography has come to cleave so neatly along party lines, both parties are locked in a game of whether or not urban voters are successfully suppressed. So the question becomes, “if state lines were realigned around mega-regions, would the resulting Electoral College better represent voter preference?” If we apply the existing Electoral College structure to the Mega-Region map, we do see a markedly different outcome:
It is rough, since it is based on existing counties, and relies on my best estimation of how counties would be reassigned to new mega-region ‘states.’ Additionally, the Mega-Regions map creates 55 states rather than fifty. I assumed the same apportionment methodology that the US currently relies upon (number of electors determined by total legislators), so each mega-region receives two “senators,” and 428 electors are divvied proportionally by population. The 270 to win rule remains in place.
Mega-Regions reveal a few interesting lessons about our current Electoral College.
First, cities are not the sole beneficiaries of more urban-centric political boundaries. Large rural areas are made more coherent. For instance, New York’s 29 electoral votes are always solidly blue because New York City’s 2.5 million voters overpower the more rural parts of upstate New York. Smaller population centers like Buffalo, Rochester, and Syracuse are coupled with the Finger Lakes and Adirondacks wilderness to cast their own votes, which actually flips their outcome. In California, the northern and inland areas that are more agricultural actually gather more voting power and break from the state’s strong blue bent. In fact, the mega-region map would leach quite a few votes from large urban centers, allowing the rural regions to split off and cast their own votes.
Small states and ultra-rural regions still have an outsize influence. This redrawing does not eliminate the inherent “representation gaps” that benefit small states — these are the gaps between what a state’s electoral vote count would be divvied solely by population, as opposed to number of legislators. The below map illustrates how many electors each mega-region “state” gains or loses by traditional apportionment, rather than purely population-based apportionment. As seen below, apportioning electors based on legislators rather than population would have netted 13 electors to the GOP, almost exclusively from rural regions:
If we compare how mega-regions are electorally over- or under-represented with the current map:
Realigning around mega-regions would have smoothed out the starker representation gaps (e.g. California & Texas), with an explicit benefit to right-leaning places. Apportioning based on population would not have flipped the election in the current system, but it would have in a mega-regions election, which reveals the inherent bias of providing favor to small, less populous states. There is no defensible reason that one set of voters should have their votes count for more than another set, regardless of rural vs. urban or small state vs. big state.
Ultimately, the question we have to answer is whether our electoral system promotes equity.
What this exercise really proves is that there is no system of electoral college that can be equitable while allocating all the votes in a state to one party or the other. Equity in voting is simple: one person gets one vote.
Mega-regions as a concept may make sense for legislative representation. In the republic, equity means that all people can have a voice through representation, and institutions create opportunities for those who would be marginalized to form a plurality that gives them a voice. It could enable regions with distinct identities and economic interests to represent themselves, and reduce some of the internal competing tensions that exist in current states. One imagines a region around New York City, for example, that could build regional infrastructure in a less fragmented manner. Cities like St. Louis could be given a stronger voice in state politics, rather than being split and cannibalized by two different states. Conversely, in more rural places attached to blue cities, like eastern Washington/Oregon or most of Illinois, setting policy or making spending decisions with the same assumptions as Seattle or Chicago makes little sense. Redrawing state lines around economic centers and activities could be a very powerful tool for giving voice to more and more diverse concerns.
But for the presidential election, ultimately, one person-one vote is the only structure that passes muster, and any system that suppresses or devalues one set of voters relative to another is a system that fails at being democratic.
