Skip to main content
How The House Got Stuck At 435 Seats

Apportionment, or the process of determining the number of seats each state has in the U.S. House of Representatives, happens like clockwork at this point. Every 10 years, the Census Bureau counts how many people each state has and then uses that number to calculate how many representatives each state gets out of the 435 seats.1 In April, for instance, we learned from the reapportionment process that California would lose a seat for the very first time while Texas would gain two.

But despite some states losing seats while others pick them up, the reapportionment process is itself now fairly mundane. That wasn’t always the case, though.

“The first presidential veto was used on the apportionment law, so it’s been a hot issue from the very, very beginning,” said Margo Anderson, a professor emerita at University of Wisconsin-Milwaukee who studies the social and political history of the census. In fact, until the House was capped at 435 seats2 by the 1929 Permanent Apportionment Act, each apportionment period was regularly accompanied by clashes over how to best divvy up political power in Congress — including the size of the House.

On the one hand, it’s probably a good thing that Congress is no longer debating the size of the House every 10 years. After all, the reason we have the 1929 Permanent Apportionment Act in the first place is that Congress was unable to reach an agreement on how to reapportion the House for nearly a decade.

Why has Biden’s approval rating fallen? | FiveThirtyEight Politics Podcast

On the other hand, the fact that the size of the House hasn’t increased in more than a century is a real problem for our democracy. For starters, there is an ever wider gulf between Americans and their representatives, as the average number of people represented in a district has more than tripled, from about 210,000 in 1910 to about 760,000 in 2020.3 Moreover, some states are severely over- and underrepresented as a result.

Increasing the size of the House would not resolve all the challenges facing the U.S., as any expansion would involve trade-offs. For instance, adding representatives could decrease day-to-day legislative efficiency, and it would undoubtedly increase the size of the federal government. Yet expanding the House is one of the more straightforward reforms that leaders in Washington could pursue in our era of polarized politics. The size of the House is determined by statute, not the Constitution, meaning Congress could pass (and the president could sign) a law to change it. 

It’s worth exploring, then, whether 435 is still an appropriate number of House members to represent our sprawling, diverse nation. Whether Congress will take up this issue anytime soon is another question entirely, but here’s how we got stuck at 435 in the first place — and what it would mean if we increased that number.

Why 435?

There have been 435 seats in the House for so long now that it might seem as if the Founding Fathers had foreseen it as a natural ceiling for the chamber’s size. But that isn’t the case: 435 is entirely arbitrary. The House arrived at that number because of political expediency — and it has stayed there because of it, too.

Up until 1910, when the chamber expanded from 391 to 435 seats,4 the size of the House had experienced a mostly unchecked pattern of growth. Only once, after the 1840 census, did the number of seats in the House not increase; 1910, however, marked the last time the House grew, even though the U.S. population has more than tripled since then, from over 90 million in 1910 to over 330 million today.

The 1920 census is when things broke down. For the first time, a majority of the population lived in “urban” areas. And although the Census Bureau’s definition was broad — it included any place with at least 2,500 people — the finding reflected America’s power center was moving away from rural areas toward urban ones due to industrialization and high levels of immigration from Southern and Eastern Europe. This made the apportionment process particularly challenging, as Congress had to navigate two competing concerns: first, the worry that greater urban power would lead to rural seat loss if the House didn’t expand, and second, a growing belief among many members that the House was already too crowded and that an increase in seats would make it truly unwieldy.

Nevertheless, the Republican chair of the House Census Committee put forward legislation in 1921 to increase the size of the House by 48 seats — 483 in total. Once again, this would have prevented any state from losing a seat, a politically attractive option.5 But this time both parties were deeply divided over expanding the House, with arguments that adding seats would be too expensive or hinder legislative functions. 

Congress tried a number of alternatives. First, the House passed an amended bill to keep the House at 435 members. Eleven states stood to lose seats as a result, and unsurprisingly many senators from those states worked behind the scenes to keep that bill from ever getting a vote in the Senate. Next, the House tried to expand to just 460 seats instead of 483, which would have caused only two states to lose a seat, but that narrowly failed by four votes on the House floor. This left Congress at an impasse, and over the next few years, reapportionment stalled. 

Some rural legislators charged that the timing of the 1920 census presented an inaccurate picture of the country’s population, claiming for instance that many people had migrated to cities only temporarily during World War I but would soon return to rural areas. Others argued that non-citizens ought to be excluded from the counts, which would have primarily affected Northern states with large immigrant populations. Meanwhile, some Northern Republicans, upset by Democrats’ disenfranchisement of Black Americans in the South, countered that representation ought to be reduced in Southern states that suppressed voting rights. There were also arguments over which method was best for apportioning seats, as one method tended to put slightly more seats in less populous states and the other put more seats in more populous states.

The lack of consensus on how to reapportion the House meant that by the late 1920s, reapportionment had dragged on for nearly a decade and had all the makings of a constitutional crisis. “The issue began to come to a head as the 1928 election loomed,” said Anderson of the University of Wisconsin-Milwaukee. “Because the realization was, we’ve got the Electoral College apportioned on the basis of the 1910 census, and if the popular vote and the electoral vote diverge, it’s because we didn’t reapportion.”

Cuomo is resigning. Now what? | FiveThirtyEight Politics Podcast

Fortunately for electoral legitimacy, Republican Herbert Hoover won both the popular and electoral vote in the 1928 presidential election. Having served from 1921 to 1928 as secretary of commerce, which oversees the Census Bureau, he was especially cognizant of Congress’s apportionment failure. In April 1929, Hoover called a special session of Congress, where one of the main focuses was apportionment, and by June, legislation had passed both the House and Senate and was signed by Hoover. The law, the Apportionment Act of 1929, created what we know as the “automatic” reapportionment process today. It capped the number of House seats at 435 and moved the responsibility of determining the seat count from Congress to the president — an early example of Congress giving away power to the executive branch.6

But given the rancor surrounding reapportionment, the law didn’t come without serious consequences for representation. Specifically, it cut requirements that members be elected in single districts and that those districts be contiguous and compact, serving relatively equal-sized populations. This meant a state that lost seats could now draw wildly disproportionate districts to keep power in more rural parts of the state. 

“It essentially created massive malapportionment for the next 40 years,” said Anderson. But, she stressed, this was done because it made the law “politically palatable.” 

In fact, the law’s lack of a population requirement helps explain why more than half of all members from rural districts backed it, even though most of the states that lost seats were based in the rural South and Midwest.7 These representatives knew their states might lose seats, but they hedged that their slower-growing or shrinking districts might not end up on the chopping block now that the apportionment process didn’t require districts to have equal populations.

Later, to uphold the tenet of “one person, one vote,” the Supreme Court would rule that congressional districts must be approximately equal in population, but that wouldn’t happen until 1964. And even then, unequal representation in the House has persisted, largely because the size of the chamber hasn’t budged despite massive growth in the U.S. population.

The problem with being stuck at 435

In 1910, the largest state, New York, had about 9 million more people than the smallest — that is, least populous — state, Nevada. But today, the largest state, California, has nearly 39 million more people than the smallest, Wyoming. 

This staggering gap makes it far more likely for states to end up with wildly unequal district populations thanks to the Constitution’s requirement that each state have at least one congressional district. The Supreme Court requires districts to have equal populations, but this applies only to the districts within a state — not between states. So even though the average House district will have just over 760,000 people after this round of reapportionment, each state’s average district will vary quite a bit, especially as states get smaller in size.

An illustration of an elephant and a donkey playing tennis on top of a leaning stack of books. The books’ spines are labeled “Supreme Court,” “State Legislature,” “Electoral College,” “Senate,” and “House of Representatives.”

related: Advantage, GOP

Take the smallest and largest states with only one representative: Wyoming and Delaware, respectively. Wyoming, with just under 578,000 people, winds up overrepresented because it’s guaranteed a seat despite falling well short of that 760,000 national average. Conversely, Delaware has nearly 991,000 people, which leaves it underrepresented because it isn’t quite large enough to earn a second seat. Meanwhile, Montana has only about 95,000 more people than Delaware, but that’s enough for the apportionment formula to eke out a second seat, meaning Montana will have two districts to Delaware’s one and an average district size of just over 542,000, making its constituents the most represented in the country.

State lines make perfectly equal districts across the country impossible, but there’s no question that increasing the size of the House would help reduce how unequal district sizes among states have become. Expanding the House could also make districts smaller, which in turn could help with representation, as the average number of people living in a congressional district has grown by about 520,000 people from 1920 to 2020 — three times more than the total shift from 1790 to 1910.

In fact, the problem of representation in the U.S. is so bad that each member of the House represents far more people on average than legislators in most other large, developed — or developing — democracies. On the one hand, this is somewhat understandable given the U.S. has the third-largest population in the world after China and India, the latter of which also happens to be the only democracy with more people per representative than the U.S. But beyond India, other large democracies with more than 100 million people, like Brazil and Japan, offer their constituents far more representation than the U.S. Moreover, their lower legislative chambers are only somewhat bigger than the U.S. House.

The U.S. has a representation problem

Average population per seat in the lower legislative chamber or unicameral legislature in the U.S. and 30 other democracies

Country Population (millions) Seats Avg. population per seat
India 1,326.1 543 2,442,161
United States* 331.1 435 761,169
Brazil 211.7 513 412,702
Colombia 49.1 172 285,377
Japan 125.5 465 269,909
Mexico 128.6 500 257,299
Argentina 45.5 257 176,962
South Korea 51.8 300 172,784
Australia 25.5 151 168,652
Spain 50.0 350 142,902
South Africa 56.5 400 141,159
Germany† 80.2 598 134,046
France 67.8 577 117,588
Chile 18.2 155 117,334
Netherlands 17.3 150 115,203
Canada 37.7 338 111,521
United Kingdom 65.8 650 101,171
Italy 62.4 630 99,052
Poland 38.3 460 83,222
Belgium 11.7 150 78,138
Israel 8.7 120 72,296
Czech Republic 10.7 200 53,512
Hungary 9.8 199 49,105
Austria 8.9 183 48,412
Portugal 10.3 230 44,794
Switzerland 8.4 200 42,020
New Zealand 4.9 120 41,046
Greece 10.6 300 35,357
Denmark 5.9 179 32,790
Sweden 10.2 349 29,233
Finland 5.6 200 27,858

Population estimates for countries other than the U.S. are all from July 2020.

*The population of the 50 U.S. states.
†The Bundestag, Germany’s lower legislative chamber, must have a minimum of 598 seats. Due to Germany’s electoral system, which can assign additional seats, the size of the Bundestag often fluctuates. It has 709 seats now.

Sources: CIA World Factbook; U.S. Census Bureau; “A Different Democracy,” by Steven L. Taylor et al.

“Our congressional districts are just massive, there’s really nothing else like it,” said Jonathan Rodden, a political scientist at Stanford University who studies political geography. “The scale of districts in Canada, the U.K. and Australia is so much smaller … the U.S. is really an outlier in this.” 

Brian Frederick, a political scientist at Bridgewater State University, studies apportionment issues and has argued that the House should be expanded. He notes how the size of America’s districts hurts the quality of representation that voters receive. In fact, his research has found a lot of upsides for smaller districts. For instance, representatives who serve fewer people are more popular, more likely to have contact with their constituents and more likely to get higher marks for their constituent service. Moreover, they often better reflect the views and makeup of the people in their districts. “The reality is that it’s easier to represent fewer people than it is a larger number of citizens on a per-district basis,” said Frederick. 

Both he and Rodden noted that an expansion of the House could also increase the relative demographic diversity in the House. For instance, having districts with smaller populations could produce a plurality-Native American congressional district in Arizona or New Mexico, which is currently not possible given the size of the group’s population. However, Rodden warned that opportunities to expand representation for minority groups could vary, especially in the South, where Black voters are often over-concentrated in districts to ensure representation.

Adding seats to the House could have electoral benefits, too. First, a growing House would make it less likely that states lose representation in the reapportionment process. Under current conditions, states with a shrinking population often lose seats, but this is true even of states where the population is growing.8 An expansion of the House would also help reduce the Electoral College’s bias toward small states, as more populous states would pick up more representatives, and therefore electoral votes in the Electoral College.9 And finally, a larger House could theoretically help reduce partisan gerrymandering. As Rodden told me, when you add more and more seats, you converge on proportional representation at some point because the districts just become so small. Still, he cautioned that line drawers could get pretty creative, so more districts might not always result in more proportional representation.10 

Clearly, expanding the House has many potential upsides — many of them beneficial to democracy, too — but, of course, a lot hinges on just how many seats would be added. And on that point there is no easy answer.

How to expand the House

A number of ideas have emerged for how best to expand the House. Some reformers have suggested a one-time, arbitrary fix, like adding 50 seats. Others have argued for a more substantive overhaul, like resizing the House based on the population of the smallest state — often called the Wyoming rule, as Wyoming has occupied this position since 1990. 

But there’s actually a fairly straightforward solution that isn’t too far off from what America used to do before — albeit unintentionally. It’s known as the cube root law in political science, or the fact that the size of a country’s parliament often hews to the cube root of the nation’s population.

Matthew Shugart, a professor emeritus at University of California, Davis, has tried to unpack why this is often the case. After all, there is no law that says countries’ parliaments must be the cube root of their population, yet they often are, as the chart below shows. Of the 30 major democracies Shugart and his co-authors looked at alongside the U.S., a majority of them have legislatures very close to — or fairly near — the cube root of their populations.

Take Canada. Its lower legislative chamber, the House of Commons, has 338 seats, almost exactly in line with the cube root law’s expectation of 335 seats. This is in large part because Canada has frequently adjusted the chamber’s seat count to account for population growth. But other bigger democracies like Brazil and Japan also have seat counts that fall fairly close to the cube root of their respective populations. Of course, this isn’t true of every democracy Shugart and his co-authors studied. Some countries like the U.S. fall well below the cube root of its population. And countries like Australia, India and Israel are even more underrepresented than the U.S. in their legislatures.11 It’s also the case that some countries like Germany, Italy and the U.K. may actually be overrepresented in their lower chambers — for instance, the U.K.’s House of Commons has 650 seats, well more than the expected 404 seats.

An illustration of a wrecking ball breaking down a brick wall that looks like the American flag.

related: Why The Two-Party System Is Effing Up U.S. Democracy Read more. »

According to Shugart, the reason why representation in countries’ lower chambers is often so close to the cube root of their populations is that the legislators must strike a balance between communicating with one another and their constituents. “It is about finding what is the optimal size,” he said. And in many countries, that seems to be roughly the cube root of a country’s population.

In fact, it’s a pattern the U.S. used to mostly follow until the size of the House was capped at 435 seats in 1929. But as the chart below shows, the House would have to grow to 692 seats to reflect where the cube root law expects representation in the U.S. to be now.

That would make the House almost 60 percent larger than it is now, so it’s hard to imagine a one-time increase of that scope. Shugart suggested a phased expansion over the next few decades, although he also didn’t think the House necessarily had to get all the way to 692 seats — he just stressed that, according to the cube root law, where the U.S. currently falls suggests that it is dramatically underrepresented.

Regardless of the potential benefits of a bigger House, though, there would likely be steep opposition to expanding it because of some of the tradeoffs — and potential downsides — involved. For instance, a larger House would by necessity mean a bigger government and more spending. House members make $174,000 per year, and after five years of service they are also eligible for a pension. Combine that with new staff, new construction for office space, perhaps even a roomier House chamber and you’re talking about many millions or even billions of dollars

There could also be consequences for governing, too, such as more gridlock and partisanship. “By increasing the number of players who have to be satisfied in the legislative game, you make arriving at the kind of majorities — or, in most cases, supermajorities — that you need to pass legislation more difficult,” said L. Marvin Overby, a political scientist at Pennsylvania State University-Harrisburg who studies Congress and has expressed skepticism toward the promised benefits of House expansion. He also warned that a bigger House might produce fewer competitive seats thanks to partisan sorting and fewer representatives open to compromise. “You would have even less of an incentive as an individual member of Congress to try to do things on a bipartisan basis,” said Overby, “because your district would be increasingly homogeneous — increasingly Democratic or increasingly Republican.”

As such, even more elections may be effectively decided by primaries instead of general elections than they are today, which is already the case in the vast majority of House districts. And with more safe seats, incumbents would likely have an even easier time getting reelected than they currently do.

In addition, there simply isn’t public support for expansion at this point. In 2018, 51 percent of Americans told the Pew Research Center that the size of the House should stay the same, while only 28 percent wanted to expand it (another 18 percent actually wanted to shrink it). Moreover, members of Congress aren’t wild about the idea, either. Legislation introduced in February by Rep. Alcee Hastings of Florida, a Democrat who died in April, aims to establish a bipartisan commission to examine the size of the House, among other things. But the bill has only four co-sponsors and looks unlikely to go anywhere.

Clearly, there are pros and cons to increasing the size of the House, but at the very least, the idea should be more openly debated because, in terms of changes that could be made to our institutions, expanding the House is actually doable. For instance, the Senate’s small-state bias often gets a lot more attention, but any change to the Senate would require a constitutional amendment whereas the size of the House could be altered with a simple bill.

“It’s going to be difficult to increase the size of the House of Representatives; I’m under no illusions,” said Frederick of Bridgewater State University. Nevertheless, it may be time for a change given how unequal districts have become between states and how underrepresented Americans are after more than 100 years of being stuck at 435 House members. Said Frederick, “There’s no doubt that a larger House with smaller constituency population size per district would improve the representational quality that citizens receive from members of Congress.”

Anti-establishment sentiment on ballot for both parties this week

COVID-19 cases are rising. Americans’ thoughts on next steps. | FiveThirtyEight


  1. Strictly speaking, the House has 441 members: 435 are voting members from each of the 50 states, and six are nonvoting members. The District of Columbia, Guam, the U.S. Virgin Islands, the Northern Mariana Islands and American Samoa each have a delegate, while Puerto Rico has a resident commissioner.

  2. This number increased to 437 in 1959 to accommodate the statehood of Alaska and Hawaii but returned to 435 in 1963, after the reapportionment process in 1960.

  3. Throughout this piece, we’ll often refer to census years as the time when the House was reapportioned for simplicity’s sake. But historically, Congress has usually passed apportionment acts in the year or two after the census was released, but still in time for elections ahead of the next Congress.

  4. Upon statehood, Congress added two seats for New Mexico and one for Arizona in 1912 during the 62nd Congress, which had already passed the Apportionment Act of 1911. This brought the total number of House seats to 394 before the House expanded to 435 seats in the 63rd Congress.

  5. Minimizing seat loss had long been a major consideration in the apportionment process, to the point that only a handful of states lost any seats in the five apportionments from 1870 to 1910.

  6. This is why even now the secretary of commerce reports apportionment figures to the president, who then transmits that information to Congress.

  7. In total, 21 states lost seats as part of the reapportionment process, and most of them were in the South or Midwest. Meanwhile, the 11 states that gained seats were industrial states like New York and Michigan and fast-growing states mostly in the West.

  8. Michigan, Ohio and Pennsylvania all grew about 2 percent between 2010 and 2020, for instance, but all three states still lost seats in the 2020 apportionment process.

  9. A state’s electoral vote count is the total number of representatives plus a state’s two senators. (The current number of electors is also what FiveThirtyEight is named after, so if this happened, we might have to consider a name change. Perhaps SixThirtyEight?)

  10. In a 2013 paper that Rodden co-authored, he found that the more districts were added to Florida’s map, the more proportional the partisan split became across those districts, although Republicans still retained an advantage given Democrats’ overconcentration in urban areas. But he didn’t find this pattern everywhere. For instance, his forthcoming study of Pennsylvania’s map didn’t find that adding more districts led to more proportional outcomes.

  11. The lower chambers in the case of Australia and India. Israel has a unicameral legislature.

Geoffrey Skelley is an elections analyst at FiveThirtyEight.