I’ve noticed in the comments on some of my previous diaries that people are curious about my assertion that the US House of Representatives is too small. I have previously written about this on my old blog, so I’ve decided to repost that article here with a few modifications.
The image above shows which states gained or lost seats after the 2010 Census. It’s clear that the population is continuing to move from the northeastern corner of the country to the Sun Belt in the south and to the West Coast. California did not gain any representatives this cycle, but its population has exploded over the past 50 or 60 years to the point that roughly 1 in 8 Americans is a Californian.
It seems only fair that states that lose population should lose representation in the House, but there is a small problem with this reasoning: Only one state actually lost population over the ten years between Census 2000 and Census 2010.
As you can see, only Michigan lost population is absolute terms. Everywhere else, the difference between gaining and losing House membership is down to relative population. This is because Congress decided to fix the size of the House at 435 voting members, for political reasons, at its 1910 membership. Back then, there were fewer than 100,000,000 Americans. Today, there are over 300,000,000. So in 1910 there was one House member for every 200,000 people, whereas each member today represents on average nearly 710,000!
Historically, Congress has recognized a need to increase the size of the House to limit the population/representative ratio.
The cube root observation is supported by political science research, notably by Rein Taagepera in the 1970s. Taagepera observed that national legislatures tend to have a membership close to the cube root (that number which, multiplied by itself twice equals the first number) of their populations. The 2010 Census found the US population to be 308,745,538, so our House should have something close to 675 or 676 members. Our 435 voting members are far short of that number.
Interestingly, the British House of Commons is quite a bit larger than the rule would predict. The 2011 UK Census put the population at 63,181,775. The cube root rule predicts a house size of 398, but the House of Commons has 650 MPs! Perhaps we should switch.
The small size of the US House with respect to its population leads to a number of problems. States can gain population but lose representation, as shown above. Dividing the small number of representatives between the states leads to absurd differences in population per representative between states. Montana and Rhode Island have similar populations, 989,415 and 1,052,567, respectively, yet Montana has only one Representative while Rhode Island has two. The smallest state, Wyoming, also has one Representative who only has to speak and vote for 563,626 people. Few states have enough representatives to make reasonably proportional representation possible. Running the House apportionment algorithm with 650 Representatives, however, gives every state larger than West Virginia at least five members, and every state bigger than Rhode Island would have at least three. Both Montana and Rhode Island get two Representatives.
The only thing between Americans and a more representative House is the statute fixing the House at 435 members, so it would only take an act of Congress to change it, not a constitutional amendment. Formalizing the cube root rule is one option, but even a fix so that states cannot lose House members without actually losing population seems like a much easier sell. If we do nothing, it is inevitable that the ideal of one person one vote will be strained to the breaking point with huge and wildly uneven populations per representative.