The decline of representation in our representative democracy

The effort by Texas Republicans to wring five more GOP House seats from their state triggered a range of responses. For example: This doesn’t seem like something that a party confident in its electoral chances would do.

Perhaps the most useful responses extrapolated from this attempt to game the rules to the rules themselves. Redrawing maps to divvy up Republican voters as usefully as possible to swing five of 435 total seats relies on there being 435 legislators representing distinct areas. Neither of those things has to be the case, though they are and have been for some time.

For about a century, in fact, a period during which the population of the United States more than tripled. At the risk of being painfully on-brand, the number of representatives hasn’t been increased since before the massive, post-war boom in babies that reshaped the country.

What that means is that the country’s representatives are much less able to represent their constituencies than they used to be.

Consider the president. In 1920, about 26.8 million people cast ballots for the chief executive — about a third of the votes cast last year for Donald Trump alone. Last year, there were 15.8 million votes cast just in California.

The winner that year would represent about 105 million people. Trump currently represents about 342 million.

A few weeks ago, I wrote about how bad we are at comprehending such large numbers. But let’s set that aside for now and just consider the extent to which we are effectively represented by the legislators elected to represent us federally.

In the Senate, for example, the average number of people represented by each senator went from about 204,000 in 1820 to 1.1 million in 1920 to 3.3 million in 2020.

In the House, the average number of U.S. residents rose from 45,000 in 1820 to 243,000 in 1920 to 762,000 in 2020.

You’ll notice that we’re using slightly different metrics for evaluating the House and the Senate. You’ll see why in a moment.

The distribution of seats in the House is adjusted after (nearly) every census (and, Texas Republicans hope, occasionally at the mid-point between censuses, too). The population is divvied up and seats assigned by state, with the effect that, even though each seat is meant to represent a consistent number of people, there is instead a range — currently a fairly wide range — between the two.

On average, though, the math works out. Dividing up the population by the number of seats generally yields a figure close to the average number of people represented in each House seat.

You’ll notice that on the graph above I indicated the point at which the number of House seats was frozen. If the number of seats had continued to grow at the rate seen prior to 1930, the chart above might look like the one below.

Anyway, that didn’t happen. And, compared to the other side of the Capitol, the allocation of power in the House is relatively fair.

There, too, there’s a range between the senators with the largest and smallest constituencies. It’s a much larger range, though — so large that it can be hard to distinguish from the original chart of Senate representation.

As you are no doubt aware, there is a massive gap between the number of people represented by senators from Wyoming (the least populous state) and California (the most populous). For what it’s worth, the ratio between the largest and smallest states used to be worse. In 1900, for example, New York had 172 times as many residents as Nevada (7.3 million vs. 43,000). The ratio of people represented by each state’s two senators was just as large.

But the difference in actual residents was fall smaller than the current gap. That was a gap of about 7 million, compared to the 39-million-resident gap that current exists between the most- and least-populous states.

The average Californian has to compete against 19.8 million other residents (in theory) in order to be heard. The average New Yorker in 1900 had to compete against only (“only”) 3.6 million.

The lopsidedness of the Senate can be viewed another way. In 1900, the population of New York was larger than that of the 17 least populous states — meaning that New Yorkers had two senators for the 34 senators representing those 17 smaller states.

Now, California’s population is larger than the 21 smallest states, meaning that there are 42 senators representing a population that’s smaller than the population represented by California’s two senators.

California is more populous than these 21 states, to be specific.

We don’t here need to go into the debate over whether the allocation of power in the Senate is justified or necessary. What we can do, though, is note that that allocation is much more unequal than it used to be.

At the same time, we can acknowledge that the representative democracy created under the Constitution is far less representative of individual Americans than it was or than it could be. Even before Texas games the system.

Photo: The Senate Armed Services Committee in 1949. There’s a future president in the mix. (National Archives)