Should We Expand the U.S. House of Representatives?

Law professor Steve Vladeck recently made some noise when he proposed expanding the U.S. House of Representatives in a recent op-ed piece, where he argued in favor of increasing the number of elected representatives from today’s 435, where each member represents about 710,000 Americans, up to 650, where each would represent aroung 500,000.

… unlike changing the equal representation of the Senate, changing the size of the House can be done simply by statute; the change to the Electoral College would follow automatically. As for how many seats to add without creating an unmanageable legislative body, a good starting point might be to reduce the average size of House districts by one-third — to 500,000 residents per representative.

Given estimates that the U.S. population today is roughly 325 million, that would mean expanding the House to roughly 650 seats — coincidentally, the exact size of Britain’s House of Commons. (Another proposal is the so-called “Wyoming rule,” which would peg the average size of House districts to the total population of the least-populous state, i.e., 563,626, under the 2010 Census.)

The United Kingdom has a population of about 67 million people today, about one-fifth the population figure he cites for the United States, so that’s not exactly a compelling argument for improving representative democracy so much as it is an exercise in mystic numerology.

But since Vladeck claims that having a lower people-per-representative ratio would lead to an “unmanageable legislative body”, let’s start with the following argument he made in support of his position to find out how well founded it is:

There are three major problems with Congress’s refusal to increase the size of the House. First, and most obviously, it is difficult — if not impossible — for any one person adequately to represent the interests of three-quarters of a million people. A member of Congress who somehow finds a way to meet with 1,000 different constituents every single day would still not see everyone they represent over the course of a two-year term.

We wondered how big of a problem that personal interaction with an elected representative really might be, so we extracted historical population data from each U.S. Census from 1790 through 2010, the most recent year the decennial Census has been conducted and also the number of representatives that have been set as a result so we could calculate the changing population-to-representative ratio over time. We also recognized that if a representative were unable to engage with their constituents to the degree established at the nation’s founding, that they would be likely to hire personal staff to assist them for that purpose, so we collected data for the size of the personal staff for each representative’s office over these years.

The first chart below shows what we found for the total numbers of elected representatives and their paid personal staff members who would be available to engage with members of their congressional districts for the census years of 1790 through 2010, which would cover the total through this point of time in 2018.

This chart presents an interesting point. Before 1893, elected members of the U.S. House of Representatives were not permitted to hire people to serve on their personal staffs using taxpayer dollars. After 1893, there were allowed to hire one person to fill that role, which lasted until 1940, when they could expand their staffs to hire three people. After World War 2, that doubled to six, which lasted until 1966, when the number of personal staff allowed per representative doubled again to 12. In 1975, that figure rose to 18 full time employees and in 1979, that number was again augmented by allowing the hiring of an additional four part time workers, which we’ve treated as the equivalent of two full time workers in our chart.

How does that affect the ratio of constituent contact with their elected representative’s office? That answer is shown in our next chart, where we’ve indicated the number of a district’s population that is served per member of their elected representative’s office, as made up by the elected representative and their paid personal staff, as well as by the representative himself, if he were completely unassisted.

In 1790, we calculate that a single elected member of the U.S. House of Representatives would have served approximately 37,421 Americans from their home congressional district, without the aid of any paid personal staff. Through the 2010 census year, we find that a single employee of an elected U.S. Representative’s office would serve about 33,798 Americans each, nearly 10% fewer than what an unassisted Representative could achieve on their own in 1790. It is also far less than the ratio of 177,000 to 1 of population-to-representative that was reached in 1890, prior to the decision of the U.S. House of Representatives to allow each member to hire personal staff to assist them in serving their congressional district’s needs.

If we ignore the contribution of the members’ personal staffs in assisting them, then the ratio has grown from 37,421 per representative in 1790 to 709,760 per representative through the 2010 census year. Focusing on that number however is really a bit of misdirection on Vladeck’s part, because of the role that representatives’ personal staffs have in facilitating communication between the elected representative and their constituents.

Closing out Vladeck’s other two perceived “major” problems, the second one has do with the unequal distribution of the U.S. population among individual states which makes it difficult to exactly slice and dice congressional districts so that each cover identically sized populations, although they are currently approximately equal within each state regardless of its relative population to other states. Vladeck’s third problem is more about his being upset that states with small populations end up with a disproportionate say in the Electoral College under the geographic affirmative action program established by the authors of the U.S. Constitution to reduce discrimination against dispersed rural populations, where he argues in favor of expanding the political power of more urbanized, higher population regions to influence the selection of the U.S. President by expanding the House of Representatives to their benefit.

We’ve ambivalent on whether increasing the number of members of the House of Representatives is a worthwhile consideration, but if the point of the move is to improve the ability of congressional district residents to interact with their representative’s office, there’s not much to be gained from the proposal that Vladeck has put forward. We might be more positively inclined to support such a change if the pay of both elected representatives and their personal staffs were slashed proportionately according to their reduction in number of residents served, along with the number of paid personal staff members, since there would clearly be less work for them to do. Plus, we know exactly where they could meet in Washington D.C. if the U.S. Congress ever got really serious about expanding the number of representatives to match the original intent of having no fewer than 30,000 Americans for each U.S. Representative to represent were to be truly realized….

Data Sources

U.S. Census Bureau. Decennial Census of Population and Housing. [Online Database]. Accessed 20 October 2018.

Wikipedia. United States Congressional Apportionment. [Online Article]. Accessed 20 October 2018.

Sourcewatch. Congressional offices and staff. [Online Article]. Accessed 30 May 2015.