About CityWage
CityWage answers one question: what does it actually cost to live comfortably in a given US city? Not the rent figure alone, not a cherry-picked index — the full salary you'd need to cover necessities, save meaningfully, and still have room for the things that aren't strictly essential.
Every figure on this site is computed from government data. Housing comes from HUD's Fair Market Rents. Food, transport, healthcare, utilities, and other necessities come from the Bureau of Labor Statistics — Consumer Price Index and Consumer Expenditure Survey data, regionalized where BLS publishes it. Median local salaries come from BLS Occupational Employment Statistics by metro area. Nothing is hardcoded and nothing is invented.
The 50/30/20 rule
We apply the widely-cited 50/30/20 budgeting framework: 50% of take-home income covers needs, 30% goes to wants, 20% to savings and investments. A "comfortable" salary in CityWage terms is one where the needs category is fully funded from real local cost data — not stretched, not roommate-dependent, not drawn from averages that bury the reality of expensive metros.
The full calculation, including which data sources feed which line item, is documented on the methodology page.
Who this is for
If you're weighing a relocation, negotiating a salary, comparing two job offers, or just curious whether your current income actually clears the bar in your city — this site exists for you. The figures are conservative where they need to be (we use 2-bedroom Fair Market Rents, not studios) and regional where the data is strongest.
What we don't do
CityWage doesn't sell data, doesn't accept payments for placement, and doesn't publish figures that can't be traced back to a named public source. When data is missing or unreliable for a city, we say so on the page rather than fill the gap with guesses.
Corrections
Found a number that doesn't match your experience? Government data lags reality, especially in fast-moving housing markets. If something looks off, check the methodology page for the data vintage we're using, then reach out.