States have different numbers of U.S. House representatives because the Constitution allocates House seats by population, with each state guaranteed at least one, and a fixed total of 435 seats is reapportioned every 10 years after the census. States that grow faster or avoid population loss gain seats; slower-growing states lose them. This is a zero-sum redistricting process, not a proportional grant of new seats.
The mechanism: fixed House size plus decennial reapportionment
The House is capped at 435 voting members by the Apportionment Act of 1911, a cap later folded into permanent law (Title 2, U.S. Code, Section 2a). Because the total is fixed, every seat a growing state gains must come from a state that is relatively shrinking. The constitutional basis sits in Article I, Section 2, which requires seats to be apportioned among the states "according to their respective Numbers," and in the Fourteenth Amendment, which requires counting "the whole number of persons in each State."
The practical measure is not raw population but the "method of equal proportions," often called the Huntington-Hill method, adopted in 1941. It assigns seats in steps. After every state receives one seat, the algorithm computes a priority number for each additional seat: population divided by the square root of n(n+1), where n is the seats a state currently holds. The next seat goes to the state with the highest priority number, and the process repeats until 435 seats are assigned.
That square-root denominator is where the intuition breaks. It deliberately favors small states in the early rounds and large states in later rounds, producing a compromise between proportional representation and protecting small-state influence. A state does not need to hit a population "threshold" to gain a seat. It needs to rank high enough in that priority sequence relative to every other state. This is why seat changes can look arbitrary: they depend on the whole national field, not on a state's own growth alone.
The Census Bureau publishes official apportionment results, and the president transmits them to Congress within one week of the opening of the first regular session. States then redraw district lines, which is a separate legal and political process.
Quantified context: what the numbers look like
Per-person representation varies widely because the 435-seat cap is spread unevenly across a growing population. As of the 2020 apportionment, California held 52 seats, Texas 38, Florida 28, and New York 26. At the other end, states such as Wyoming, Vermont, Alaska, North Dakota, South Dakota, and Delaware held one seat each.
The average district size in 2020 was roughly 761,000 people, up from about 709,000 after the 2010 census. That growth reflects a fixed House denominator against a rising national count. A single-seat state like Wyoming had a population of about 577,000 in the 2020 census, well below the national average district size, because its one guaranteed seat is protected regardless of the math.
The 2020 cycle shifted seven seats among 13 states. Texas gained two; Florida, North Carolina, Colorado, Montana, and Oregon each gained one; California, Illinois, Michigan, Ohio, Pennsylvania, West Virginia, and New York each lost one. New York's loss was decided by a margin of 89 people in the reapportionment calculation, a reminder that these allocations are sensitive to small differences.
For scale, the original ratio in 1790 was about 34,000 people per representative for a House of 105 seats. Today that ratio is roughly 22 times larger in the average district.
Timeline: how apportionment has evolved
| Year | Change | Effect on representation |
|---|---|---|
| 1787 | Article I, Section 2 sets population-based apportionment | Establishes seats tied to state population counts |
| 1790 | First census | First apportionment of 105 seats |
| 1911 | Apportionment Act caps House at 435 | Creates the fixed pool that makes seats zero-sum |
| 1929 | Permanent Apportionment Act | Makes the 435 cap automatic after each census |
| 1941 | Huntington-Hill method adopted | Uneven tie-break and priority method still in use |
| 1959 | Alaska and Hawaii admitted | Final expansion to 50 states; 435 cap unchanged |
| 1964 | Reynolds v. Sims requires equal district populations | Forces intrastate equality even as interstate gaps remain |
| 2020 | Most recent apportionment | Seven states lost seats, six gained |
The table shows a structural pattern: representation within a state was equalized by court decisions, while representation between states remained unequal because the total seat count stayed fixed.
Edge cases and the counterintuitive detail
The most important caveat is that "more representatives" is not identical to "more representation per person." A voter in Wyoming shares one representative with about 577,000 people; a voter in a large state's district shares one with closer to the national average. On a per-capita basis, small single-seat states often have more House representation per resident than large states do.
Yet that advantage is not guaranteed to persist. If a state's population grows while its seat count stays flat, its district size rises, and its per-person representation falls. The formula protects the minimum of one seat, not a fixed ratio. A state can be "overrepresented" relative to the national average at one census and "underrepresented" at the next.
Another edge case is the treatment of territories and the District of Columbia. They have delegates with limited floor privileges, not voting representatives, so the 435-seat math excludes them. Puerto Rico, for example, has a resident commissioner with committee privileges but no floor vote. This is a structural exclusion, not a population-ranking outcome.
The counterintuitive professional caveat concerns the "equal proportions" name. It does not produce equal population per district across states, and it was never designed to. It produces the smallest possible relative differences in representation ratios among states, which is a different objective. Two states can have identical populations and different seat counts in rare tie scenarios, and small population shifts at a census boundary can flip a seat.
Misconceptions that distort the debate
A common misconception is that states gain seats purely by growing. They gain only if their priority rank rises relative to other states. A state can add hundreds of thousands of residents and still lose a seat if its growth trails the national field. In the 2020 cycle, several states lost seats despite population gains.
A second misconception is that the House could simply add seats to correct disparities. It could, legally, but only by an act of Congress, and the 435 cap has held for more than a century. Expansion proposals exist but are not current law, so any claim about future seat totals is speculative.
A third misconception conflates House seats with electoral votes. Each state's electoral votes equal its House seats plus two senators, so small states get a disproportionate boost in presidential elections from the Senate component. That is a separate mechanism from House apportionment.
A fourth misconception treats district-level equality as proof of state-level fairness. Reynolds v. Sims (1964) required districts within a state to be roughly equal in population. It said nothing about equality between states. A state can have perfectly equal districts and still have a much larger average district than a neighboring state.
Practical actions: how to check your own state's position
Readers who want to test these mechanics can take a few concrete steps.
Pull your state's 2020 apportionment result from the Census Bureau's official apportionment page and compare it with 2010 and 2000. Compute your state's seats per million residents and compare it with the national average of about 1.31 seats per million people (435 seats / roughly 331 million people). Find the average district size in your state by dividing state population by House seats, then compare it with the national average of about 761,000. Track the Census Bureau's annual population estimates to see whether your state's share of the national total is rising or falling between censuses. Distinguish House seats from electoral votes when reading election coverage, because the Senate component can mask House-level changes.
A red-flag warning for interpretation: seat changes announced after a census reflect a fixed 435-seat field, not new seats created from growth. If an article reports that a state "gained representation" without noting which state lost a seat, treat the framing as incomplete.
Limitations: what the math cannot tell you
Apportionment math is deterministic given a population count, but the count itself has known error. The Census Bureau measures net undercount and overcount, and those errors can shift a close seat allocation. The 2020 cycle's 89-person margin in New York is the clearest recent example of how measurement uncertainty can decide a seat.
The framework also says nothing about how districts are drawn within a state, which is where gerrymandering, racial vote dilution, and partisan considerations enter. Those are governed by state law and federal Voting Rights Act jurisprudence, not by the apportionment formula.
Finally, the question of whether the House should remain at 435 is a policy choice, not a mathematical necessity. The Constitution sets no fixed size, and Congress has changed it before, most recently in 1911 and 1929. Any claim about the "right" number of representatives is a normative argument about representation, not a result of the apportionment algorithm.