A congressional seat opens up. Someone dies in office, resigns, gets appointed to something else. Now the state has to run a whole extra election just to fill it. Sounds simple enough. It isn't cheap, and it isn't the same price everywhere you go.
Run a special election in California and you're looking at an estimated $100 million. Run one in Wyoming and it's an estimated $4 million. Same basic task, fill a vacant seat, prices that are 25 times apart. That gap is the whole story here, and it's worth sitting with before you ever look at another special-election headline and assume the number means the same thing everywhere.
Think about what that actually means in practice. Two states, two vacant seats, two elections designed to do the exact same job: let voters pick who represents them. One costs tens of millions more than the other, and it has nothing to do with who's running or how contested the race is. It's baked into the map and the population before a single ballot gets printed.
Here's what the estimated cost of a special election looks like at the top and the bottom of the list.
The most expensive states to run one: California: $100 million New York: $60 million Texas: $50 million
The least expensive states to run one: Wyoming: $4 million North Dakota: $5 million South Dakota: $5 million
Look at that spread. California's estimated cost is 25 times Wyoming's. New York's is 15 times Wyoming's. Even Texas, the third most expensive state on this list, comes in at more than 12 times what Wyoming spends. This isn't a small rounding difference between neighboring states, it's an entirely different scale of spending to accomplish the same basic democratic task of letting voters pick a replacement.
And notice who's clustered at the bottom: Wyoming, North Dakota, South Dakota. Low-population states with relatively simple election logistics. At the top: California, New York, Texas, three of the biggest, most populous states in the country. That's not a coincidence, and it's the key to understanding this whole list.
When a member of Congress hangs on well past the point of effective service, or leaves office suddenly, someone has to pay to replace them. That cost isn't abstract. It's ballots printed, polling places staffed, poll workers paid, ads run just to tell voters an election is even happening. States with huge populations spread across huge geographies, like California and Texas, have to stand up that entire operation at massive scale. States like Wyoming, with a fraction of the population and far fewer places to set up a ballot box, can do it for a much smaller check.
This matters for how you read Too Damn Old's data. When you see a member's age sitting next to the cost of replacing them, the size of that number is partly about that member and partly about which state they happen to represent. A vacancy in California was always going to cost more to fill than a vacancy in Wyoming, no matter who's sitting in the seat or how long they've held it. Use these estimates to understand the real, tangible price of turnover, not as a state-by-state ranking of efficiency or waste.
It also reframes what's actually at stake when aging members of Congress delay decisions about their own future. Every seat held past the point of real capacity to serve is a seat that could suddenly trigger one of these price tags, on top of the deeper cost of a district or state going without full representation while the process plays out. Multiply that risk across a delegation full of aging members in a big state, and the potential bill starts looking a lot bigger than any single line item.
These are estimated costs, built from the real, documented expense of running a statewide or district-level special election, scaled to reflect each state's population, geography, and election infrastructure. States with more voters, more counties, and more polling infrastructure to activate cost more to run an election in. States with smaller populations and simpler election administration cost less. We don't invent these numbers or guess at them state by state, they're derived directly from the cost data in our live dataset, which is built and updated to reflect real-world election administration expenses.
Want the full breakdown of how Too Damn Old builds every estimate on this site, from special-election costs to the age data behind them? The complete methodology is public and always up to date at toodamnold.com/data. Go check the math yourself. That's the whole point.