Free Markets, Free People

representatives


Should House districts cross state lines?

That’s the solution that came to my mind when I read this piece in the New York Times.

I don’t think my suggestion would violate the important aspects of our constitutional design.

As attractive as the idea of having fewer constituents represented by each Representative may be, increasing the number of seats to around 1,000 would make the House unwieldy.  Dunbar’s number reflects the difficulty of becoming familiar with large numbers of other people, so in very large bodies, it becomes difficult for one “side” to get to know the other.  That increases the tendency toward misunderstanding and factionalism, with negotiations handled entirely by a relatively small number of leaders, whips, and committee chairs.

Then there are logistical issues involved with more than doubling the size of the House (where will they all sit?), and — this might be a minor issue, but — do we want to pay 1,000 Congressmen and their staffs?  Do we expect that Congress will produce better legislation with 1100 members than it does with 538?

But the status quo does seem flawed.  The Senate may be designed to give some people more representation than others, but that’s because the Senate traditionally was supposed to be the great protector of the states.  The House was intended from the start to represent the people directly rather than the people as represented by their states, so for one legislator to represent 958,000 people (Montana) while another represents 527,000 (Rhode Island) doesn’t seem quite right.

There are a number of places where it strikes me as natural that a House district would cross state lines, because the people on either side of the border have more in common with each other than they do with other people in their state.

If an agreeable method of choosing where those lines are drawn can be devised, I see only one major difficulty with this idea.  That is: how to treat electors for the Electoral College.  If a district straddles two states that vote differently for president, the solution I see is this:

  • Each state delivers its 2 base electoral votes to whoever wins the state.
  • Any district which doesn’t cross a state border delivers its elector to whoever won the state.
  • If a district straddles a state border where the states voted differently, its elector votes for whoever won the district.

That might actually improve the Electoral College.

But perhaps I’m missing some other important snag here.  Your thoughts?