cost of living

How to Compare Cost of Living Between Cities the Right Way

Aug 28, 202614 min read
How to Compare Cost of Living Between Cities the Right Way

A city with a lower headline index can still cost your household more. The reliable way to compare cost of living between cities is to normalize the same basket of goods and services against one base city, assign weights that reflect your spending, and keep housing, taxes, utilities, groceries, transportation, and healthcare visible as separate categories.

That approach produces a defensible shortlist rather than a generic ranking. It also explains why two households with the same income can reach opposite relocation decisions.

Table of Contents

What Comparing Cost of Living Between Cities Actually Means

A single cost-of-living index can mislead buyers and renters because it averages different households into one number. A renter with a long commute, a homeowner with no mortgage, and a family paying for childcare don't experience the same city-level price pattern. Their relevant baskets are different, so their affordable-city rankings can be different too.

A practical comparison starts by choosing one base city and one reference period. Each target city is then expressed as a ratio against that base. If housing in the target has an index of 118 and the base is 100, housing costs are represented as 18% higher than the base for the measured basket. The ratio is useful only when both cities use comparable definitions, geography, and time periods.

Numbeo provides a clear model for this normalization. Its index uses New York City as the reference point, with New York City = 100, and expresses other cities relative to that benchmark. Its Cost of Living Plus Rent Index weights local prices against the New York value, while its methodology describes estimated average expenses for a four-person family. An index of 72 therefore represents a basket interpreted as 28% cheaper than New York, while 134 represents 34% more expensive, as explained in Numbeo's cost-of-living methodology.

Three rules make the comparison defensible

  1. One base city and one reference period. Don't compare a current rent estimate in one metro with an older annual index in another.
  2. Discrete categories, not just a roll-up. A headline score can hide a housing premium or a transportation advantage.
  3. Household-specific weights. National averages provide a starting point, but your rent, commute, tax status, and healthcare needs should override them where they differ.

The comparison should also distinguish price levels from purchasing power. Numbeo's 2025 city rankings include a Local Purchasing Power Index, and its U.S. metro examples show that cities in a relatively tight cost cluster can still differ substantially in purchasing power. That supports a more useful question than “Which city is cheapest?” Ask instead, “Which city keeps my total annual spending within an acceptable band after my income, taxes, and spending mix are included?”

The Core Cost Categories You Must Measure

A useful model separates six cost families before calculating a composite. The category split matters because local prices don't move together. Housing responds heavily to supply and demand, utilities reflect climate and energy systems, and transportation depends on whether your household owns cars or uses transit.

Category Primary Metric Distortion When Isolated
Housing Rent, mortgage, property tax, insurance, or price per usable area Rent alone can misrepresent owner costs, while an overall index can hide shelter pressure
State and local taxes Income, sales, and property taxes Lower rent may be offset by tax liabilities
Utilities Electricity, heating, cooling, water, and communications Climate and grid mix can matter more than local wages
Groceries Standardized food basket and applicable food taxes Supply chains and tax rules can create category-specific premiums
Transportation Vehicle ownership, fuel, insurance, maintenance, parking, transit, and commute A low fuel price doesn't make a car-dependent city cheap
Healthcare Premiums, routine care, prescriptions, and out-of-pocket spending Provider density and insurance markets can change household exposure

Housing deserves its own model

Housing is usually the largest source of variation between cities, but “housing cost” means different things to renters and owners. A renter needs a comparable unit, neighborhood, and lease condition. An owner needs mortgage cost, property taxes, insurance, maintenance, and sometimes association fees. A median rent figure can't represent both.

The Cost of Living Index manual says composite scores are built by multiplying category indexes by their weights and adding the results. It also notes that the housing component can carry 70.9% of the housing index, which shows why shelter can dominate the housing calculation even when the headline city score looks moderate. Review the C2ER Cost of Living Index manual before comparing indices with different bases.

Taxes and essentials create hidden reversals

Separate income, sales, and property taxes. A state with higher income taxes may still produce a lower total burden for a household with modest taxable income, while a place with lower rent may impose higher property costs or sales taxes. Utilities shouldn't be inferred from wage levels, either. Heating and cooling demand can change the annual pattern substantially.

Groceries reveal another type of distortion. A city may have affordable housing but higher food prices because of regional supply chains or food-tax treatment. Transportation needs the same treatment. Compare car ownership and parking with transit fares and commute distance, not fuel prices alone.

Healthcare belongs in the model even when the household expects few medical visits. Premiums, provider availability, age, and out-of-pocket exposure can vary independently of rent. Treating all six families as one blended number is the most common way a comparison loses its practical meaning.

Building a Weighted Basket for Your Household

The weighted-basket method turns a city ranking into a household model. Start with the Bureau of Labor Statistics Consumer Expenditure Survey, which provides geographic spending data by region, division, selected states, selected Metropolitan Statistical Areas, and population size. The program measures expenditures, not local price levels, so use it to establish spending shares rather than treating it as a city price index. Its geographic framework is described in the BLS Consumer Expenditure Survey geography data.

Use this four-step workflow:

  1. Pull a default share set. Choose the BLS household or geographic grouping closest to your situation.
  2. Replace mismatched shares. Increase transportation for a long commute, housing for a high-rent target, or healthcare for a multigenerational household.
  3. Convert prices into ratios. Divide every target-city category price by the same category price in the base city.
  4. Multiply and sum. Apply each household weight to its category ratio, then add the weighted results.

An infographic showing four steps to build a weighted basket for your household budget cost analysis.

A reproducible Austin and Denver example

For illustration, assign these household weights:

Category Weight
Housing 33%
Taxes 16%
Groceries 13%
Transportation 11%
Utilities 7%
Healthcare 10%
Childcare, insurance, and discretionary spending 10%

The weights total 100%. Suppose Austin is the base at 100 for every category, and your collected Denver-to-Austin ratios are housing 109, taxes 98, groceries 105, transportation 101, utilities 108, healthcare 103, and residual spending 100. The composite is:

(109 × 0.33) + (98 × 0.16) + (105 × 0.13) + (101 × 0.11) + (108 × 0.07) + (103 × 0.10) + (100 × 0.10) = 104.02

Denver would therefore score approximately 104 against Austin's 100 in this illustrative model. The calculation is not a published Austin-Denver index. It's a transparent demonstration of how your own category ratios become a composite.

The result is highly sensitive to household structure. Changing housing from 33% to 45% can reverse a recommendation when one city has a meaningful shelter premium. That isn't a flaw in the model. It shows that the recommendation depends on the buyer's actual budget rather than an average household that may not resemble them.

Where to Source Reliable City-Level Data

No source is complete enough to carry the entire comparison. The strongest workflow layers a structured price index, spending data, current market signals, and a separate housing check.

The BEA Regional Price Parities measure metro and state price levels relative to the national average and expose differences between housing and other essentials. BEA published 2025 RPP materials in June 2025 and updated metro and state price-level data in 2026, so its Regional Price Parities materials are useful for structure and current context. The weakness is release timing and the fact that an annual measure may lag a rapidly changing rental market.

The BLS CPI geography series supports regional, city-size, and selected metropolitan comparisons. BLS also publishes two-year means for regions, divisions, selected states, selected metropolitan statistical areas, and population-size groups in its geographic CPI data. CPI is a time-series inflation measure, though, not a direct cross-city price-level index, so it shouldn't replace RPP or a market-basket comparison.

Source Base Frequency Coverage Key Weakness
BEA RPP National average Annual Metro and state price levels Can lag short-term market changes
BLS CPI geography CPI reference framework Regional and metro releases Selected categories and geographies Measures inflation over time, not full city price levels
C2ER COLI Participating urban-area average Quarterly U.S. city and metro comparisons across major categories Methodology and participation can limit comparability
Numbeo New York City = 100 in its index framework Frequently updated Broad global city categories Crowdsourced data can reflect urban and expatriate users
EIU Common-currency market basket Survey-based Global city comparisons Modeled and less granular for household-specific analysis

C2ER has published its Cost of Living Index since 1968 and maintains an archived quarterly dataset with about 40,000 records dating to 1990, with back issues available since 1980, as documented in its quarterly index archive. That history helps analysts distinguish structural differences from short-term shocks.

For current rent and grocery signals, Numbeo can supplement the structured sources. For global comparisons, the World Bank explains why surveys such as the EIU's cover 140 cities in nearly 90 countries and convert price quotes into a common currency before applying weights in its city cost-comparison paper. For U.S. shortlisting, combine those datasets with city and neighborhood comparison data, then validate housing against a federal fair-market-rent series or a carefully defined rental index.

Turning the Numbers Into a Side-by-Side Comparison

A side-by-side model needs a fixed base, category weights, and a consistent scoring direction. Select your current residence, a national reference, or another defensible base city. Then express every category in the comparison as base = 100, so a higher number always means a higher measured cost.

A ReloScore grade can provide location context, but it shouldn't replace the underlying cost calculation. Keep the overall grade separate from category sub-scores, then combine the cost categories using household weights. This prevents a strong safety or quality-of-life result from hiding a budget failure.

A four-step infographic showing how to compare cost of living between cities using weighted data analysis.

Worked category comparison

Use Austin as the base and assign these illustrative category indexes:

Category Austin Denver
Housing 100 109
Groceries 100 105
Utilities 100 108
Transportation 100 101

The figures are an example of the calculation format, not verified city estimates. To complete the model, add taxes, healthcare, and residual spending from the same reference period and source family. With the weights established earlier, multiply each index by its household share and sum the results.

A household that values utilities and groceries heavily may view the Denver premium differently from a household that owns a home outright. A remote worker may also assign less weight to commuting and more to housing, internet, or local services. The model should reveal those tradeoffs rather than suppress them.

Use filters after the math

Once the categories are normalized, use constraints to reduce the candidate list. On New York, New York on ReloMaps, for example, a reader can inspect the destination through broader location context rather than treating a citywide cost number as the whole decision. Apply a rent cap, tax bracket, household size, commute requirement, and homeownership preference before reviewing the final shortlist.

The useful output isn't “City A ranks first.” It's “These three cities meet the rent ceiling, tax assumptions, household profile, and acceptable weighted cost band.” That result is easier to audit, explain to a partner, and revise when one assumption changes.

Common Pitfalls That Distort Cost-of-Living Comparisons

Generic rankings fail in predictable ways. A self-audit should challenge the input definition before trusting the final score.

  1. Using median rent as owner cost.
    Correction: Add mortgage payments, insurance, property tax, maintenance, and fees for owners. Rent is only the relevant housing metric for renters.

  2. Comparing nominal prices without regional tax adjustment.
    Correction: Convert shelf prices into after-tax household costs and separate income, sales, and property taxes.

  3. Mixing geographic units.
    Correction: Compare metro with metro, city with city, or county with county. Don't place a city estimate beside a broad regional figure without documenting the mismatch.

  4. Treating salary as a cost measure.
    Correction: Model gross income, taxes, benefits, and purchasing power separately. A higher nominal salary doesn't automatically create more disposable income.

  5. Relying on a single-year snapshot.
    Correction: Use a longer historical series where available. C2ER's quarterly archive is valuable because it extends across decades, allowing analysts to distinguish persistent differences from temporary inflation or housing shocks.

  6. Ignoring tax divergence inside one metro.
    Correction: Check the actual residence and work jurisdictions. Two neighborhoods in the same labor market can expose a household to different local obligations.

  7. Trusting crowdsourced data during volatile quarters.
    Correction: Use Numbeo as a current signal, then validate major categories with structured public data and a local housing series.

The technical error behind many bad comparisons is normalization. The World Bank warns that converting local prices at spot foreign-exchange rates without a common basket can overstate or understate real costs. For international work, price equivalent items, convert consistently, and apply spending weights instead of treating exchange rates as purchasing-power measures.

Audit rule: If a ranking doesn't show its base city, reference period, geographic unit, category definitions, and weights, treat the score as a screening signal, not a relocation decision.

For more practical location-analysis guidance, review the ReloMaps relocation research library. The point isn't to discard rankings. It's to understand what they can and can't answer before a household commits to a move.

A Repeatable Checklist for Any City Pair

A comparison becomes reliable when another analyst can reproduce it from the same assumptions. Run these steps for two cities, then repeat the final stress tests for a third candidate.

  1. Lock the base city. Use the current residence or a clearly defined reference city. Don't change the base between categories.
  2. Choose the comparison date. Record the month or quarter for rents, groceries, utilities, and taxes.
  3. Build the household basket. Start with BLS expenditure patterns, then replace categories that don't fit your household.
  4. Source structured data. Use BEA for price-level structure, BLS for relevant geographic price movement, and C2ER for historical city comparisons.
  5. Collect local market signals. Supplement structured indexes with current rent and grocery observations, but document whether the data is official, modeled, or crowdsourced.
  6. Normalize every category. Set the base city to 100 and express the comparison city as a category ratio.
  7. Apply taxes and benefits. Calculate income, sales, and property taxes separately, then include employer benefits where the household would receive them.
  8. Calculate the weighted composite. Multiply each category ratio by the household weight and add the results.
  9. Stress-test household variants. Run single-earner, dual-income, renter, and owner versions when those scenarios are plausible.
  10. Log assumptions and validate. Save the source dates, geographic definitions, weights, exclusions, and a secondary-source check.

A ten-step repeatable checklist illustration for comparing the cost of living between two cities.

Thresholds that resolve real relocation questions

When does an income difference outweigh a cost gap? Use 12% as a screening threshold, not a universal law. If the destination produces at least a 12% improvement in after-tax household income and the weighted cost increase is smaller, income may outweigh the price difference. This is Step 7, and it must use disposable income rather than gross salary.

How much rent premium is acceptable for a comparable neighborhood? Apply a 15% rent-premium rule only when the higher-priced neighborhood materially improves the factors your household values, such as commute, safety, or school access. If the premium buys no measurable household benefit, reject it or broaden the search area. This belongs in Step 5 because neighborhood selection affects the housing input before normalization.

Which utility period should you use? Use a 90-day utility average instead of a winter peak. A short peak bill can distort a city with seasonal heating or cooling demand. Record the season and tariff assumptions in Step 10.

How should healthcare be compared? Weight premiums and out-of-pocket exposure by age band and household composition. A family with older members shouldn't use a young single adult's average as its healthcare ratio. This is a Step 3 adjustment, because the household basket determines the category's importance before city prices are compared.

Why do remote-worker calculators often fail? They can misclassify the tax location when the worker keeps a prior tax state, works across jurisdictions, or changes residency without changing payroll records. Run a tax-residency overlay in Step 7 and treat the worker's actual legal and payroll position as an explicit assumption.

What salary increase offsets a $5,000 move cost? Divide the one-time move cost by the number of months you expect to remain in the destination, then compare that monthly recovery requirement with the after-tax income difference. The exact breakeven salary depends on the household's tax rate and time horizon, so a gross salary figure alone isn't defensible. This calculation belongs between Steps 7 and 8.

How often should the model be refreshed? Refresh the market-sensitive categories quarterly when possible because C2ER publishes quarterly comparisons, while BEA RPPs update annually. Re-run the complete model after a major move in rent, employment, household size, or tax residency, rather than waiting for an annual calendar update.

The Austin and Denver verdict

The supplied scenario produces a deliberately narrow result. Austin housing costs are described as roughly 18% higher per square foot, while Denver groceries are about 6% higher and childcare about 9% higher. Those figures must be tied to the scenario's stated assumptions and shouldn't be generalized to every neighborhood or household.

The scenario also says Denver doesn't receive a state-income-tax advantage, and that the all-in index falls within 4% between the two cities. That proximity means the headline winner is less important than the housing model. Austin tilts the decision only when equity appreciation is included, which is an investment assumption rather than a pure cost-of-living measure.

The checklist therefore produces a conditional verdict: choose Austin when the household can carry the housing difference and values the modeled equity outcome; choose Denver when its housing, utilities, or household-specific constraints produce the better weighted result. A ranking alone can't show that conditional logic.

Decision standard: Don't select a city because its composite is lowest. Select it because the result survives changes to housing tenure, income structure, tax residency, commute, and the time horizon for recovering moving costs.

ReloMaps brings city and neighborhood context into the same relocation workflow, including ReloScore grades, cost-of-living indicators, rent, taxes, commute, demographics, safety, and risk factors. Build your weighted assumptions, test candidate locations, and visit ReloMaps to turn a broad city search into a documented shortlist.

Frequently Asked Questions

How can I compare cities when one household owns a home and the other rents?

Use separate housing scenarios rather than forcing both households into median rent. For the owner, include mortgage, property tax, insurance, maintenance, and fees; for the renter, use a comparable unit and neighborhood. Run both scenarios through the same non-housing basket so the result identifies whether the conclusion depends on tenure.

How do I compare cost of living for a remote worker who may owe taxes in two states?

Model tax residency and work-jurisdiction exposure as separate assumptions. A remote worker who keeps a prior tax state may not receive the apparent tax benefit of the destination, so calculate disposable income under each legally plausible filing scenario before comparing city prices.

What should I do when two neighborhoods have different apartment sizes?

Normalize housing by a comparable unit specification, not by citywide median rent. Match bedrooms, usable area, household occupancy, building type, and neighborhood function, then apply the 15% premium rule only when the more expensive location provides a clear household benefit.

How can I handle a city with extreme seasonal utility bills?

Use a 90-day average from a representative period and run a seasonal sensitivity test. Keep heating and cooling assumptions visible, because a winter peak can make a climate-related difference look like a permanent annual cost.

Should healthcare receive more weight for an older household?

Yes, healthcare should be weighted by age band and household composition. Compare premiums, expected routine care, prescriptions, provider access, and out-of-pocket exposure rather than using one citywide healthcare average for every family.

How can I test whether a higher salary actually makes a city more affordable?

Compare after-tax disposable income with the weighted annual basket. A gross income advantage matters only after taxes, benefits, commuting changes, healthcare exposure, and the one-time move cost are included in the same scenario.

Produced via Outrank

Enjoying the blog? Subscribe to ReloMaps+ for premium area intelligence — and to keep us independent.

Subscribe today