RPP vs CPI: How They Differ and When to Use Each
The two most-cited government price measures answer fundamentally different questions. Mixing them up is the single most common error in cost-of-living analysis.
CPI tells you whether things got more expensive over time in one place. RPP tells you whether things are more expensive in one place than another right now. CPI is for inflation analysis. RPP is for relocation analysis. Confusing them produces real planning errors, a city with high inflation can still be cheap to live in, and a city with low inflation can still be punishingly expensive. Use the right tool for the question you are actually asking.
The One Sentence That Clears Up the Confusion
CPI is a clock. RPP is a map.
The Consumer Price Index, published monthly by the Bureau of Labor Statistics, is a time-series index. It tracks how the price of a representative basket of goods and services changes from month to month and year to year. Its job is to tell you whether prices are rising, falling, or staying flat over time.
Regional Price Parities, published annually by the Bureau of Economic Analysis, are a cross-sectional index. They measure how the price level in one geographic area compares to another at a single point in time. Their job is to tell you whether one place is more expensive than another, holding time constant.
These are different questions. They require different methodologies, different data sources, and different interpretations. A reader who confuses them ends up making decisions on the wrong evidence.
What CPI Actually Does
The BLS publishes the Consumer Price Index based on monthly price surveys conducted in roughly 75 urban areas. Each month, BLS price collectors record the prices of thousands of items, groceries, rent, gasoline, healthcare services, recreation, apparel, and feed those observations into a weighted index.
The index is anchored to a reference period. Currently, the U.S. CPI uses 1982-84 as its base, set to 100. A CPI value of 320 today means consumer prices have roughly tripled since the early 1980s. When commentators say "inflation is 3%," they mean the CPI rose 3% year over year.
BLS publishes several CPI variants, CPI-U (urban consumers, the headline figure), CPI-W (urban wage earners, used to adjust Social Security), and Chained CPI-U (which accounts for substitution between goods). For most public policy and personal finance questions, CPI-U is the relevant series.
Regional CPIs exist. BLS publishes CPI for the four U.S. Census regions (Northeast, Midwest, South, West) and for 23 metropolitan areas. But these regional CPIs are indexed to their own local base periods, not to a national average. They tell you how Chicago prices changed relative to Chicago prices in 1982, they do not tell you whether Chicago is cheaper or more expensive than Atlanta.
What RPP Actually Does
Regional Price Parities take a different approach. Instead of tracking prices over time within a region, they fix time and compare prices across geography.
The BEA constructs RPP by combining three data streams. For rents, the BEA uses gross rent data from the Census Bureau's American Community Survey, which covers all metropolitan and micropolitan statistical areas. For goods and services, the BEA uses BLS price data adjusted for geographic differences in consumption patterns and unit prices. The composite all-items RPP is a weighted average of these components.
Critically, RPP is normalized to the national average. The U.S. national price level is set to 100. Every state and metro receives an index value that is directly comparable to every other one. An RPP of 115 means prices are 15% above the national average. An RPP of 88 means prices are 12% below it.
This normalization is what makes RPP useful for cross-metro comparisons in a way that CPI is not. The two indexes are answering different questions, and the BEA designed RPP specifically to fill the gap that CPI does not address.
A Worked Example: Same City, Two Stories
Consider Phoenix, Arizona. Suppose CPI for the Phoenix metro rose 4.2% over the past year, while the national CPI rose only 3.1%. A casual reader might conclude Phoenix is becoming an expensive place to live.
Now layer on RPP. Phoenix's all-items RPP for the same year reads about 99, essentially identical to the national average. Despite faster recent inflation, Phoenix is still a roughly average-priced metro on an absolute basis.
Both observations are correct simultaneously. Phoenix prices rose faster than the national rate (a CPI fact), but Phoenix prices remain near the national average level (an RPP fact). The two measures do not disagree because they are not measuring the same thing.
The practical implication: if you are deciding whether to take a job in Phoenix, RPP tells you what you need to know about how your dollar will stretch there. If you are negotiating a cost-of-living adjustment for an employee already in Phoenix, CPI tells you how much their existing budget has been eroded.
When CPI Is the Right Tool
Use CPI when the question is about change over time:
- Inflation tracking. How fast is the cost of living rising overall?
- Indexed contracts. Many leases, alimony agreements, and pension benefits adjust based on CPI. The contract specifies the index version and base year.
- Real wage analysis. Has my paycheck kept up with prices over the last decade? Divide nominal income by the CPI ratio between then and now.
- Social Security cost-of-living adjustments. The annual COLA is calculated from CPI-W.
- Federal poverty thresholds. The Census Bureau adjusts the official poverty line each year using CPI.
- Real interest rate calculation. Subtract CPI inflation from nominal interest rates to estimate real returns.
When RPP Is the Right Tool
Use RPP when the question is about comparison across places:
- Relocation decisions. Will my standard of living rise or fall if I move?
- Salary benchmarking across markets. Is this offer in Denver competitive against my current Boston salary?
- Real personal income comparisons. Which states have the highest real living standards after adjusting for local prices?
- Federal budget allocations. Some federal programs use RPP-derived adjustments to set local payment rates.
- Academic research on regional inequality. Studies of place-based prosperity rely on RPP to compare outcomes across metros.
- Cost-of-living-adjusted poverty rates. The Census supplemental poverty measure uses RPP-derived geographic adjustments.
Where the Two Measures Overlap
CPI and RPP are not entirely independent. The BEA uses BLS retail price data, the same data BLS uses to compute CPI, as one input to the goods and services components of RPP. Where they diverge is in the framing: BLS measures price change; BEA measures price level.
BEA also publishes a "real personal income" series that combines BLS personal income data with RPP geographic deflators. This is a hybrid measure that adjusts incomes for both inflation and geographic price differences, a more complete picture than either CPI or RPP alone provides. For long-term analysis of how American living standards have evolved across regions, the real personal income series is the right starting point.
The Three Most Common Mistakes
Mistake 1: Using CPI to rank cities by cost
Headlines that say "Phoenix is the third most expensive city based on CPI growth" are wrong. CPI growth tells you about inflation in Phoenix relative to Phoenix's own past. To rank cities by cost of living, you need RPP.
Mistake 2: Using RPP to argue about inflation
RPP cannot tell you whether prices are rising. The annual RPP releases reflect the year's price level relative to that year's national average. An RPP that stays flat year over year does not mean prices are stable, both the local and national price levels could have risen by 5%, with the ratio unchanged.
Mistake 3: Mixing different vintage years
BEA publishes RPP with about a two-year lag. If you compare your current nominal salary against a 2022 RPP, you are implicitly assuming the metro's relative price level has not shifted. For most metros that assumption is reasonable; for fast-growing or fast-declining markets it can introduce meaningful error. Always check the vintage of the RPP data you are using.
How PlainCost Uses Both
PlainCost is built around RPP because our core use case, relocation, salary benchmarking, and metro comparison, is a cross-sectional question. The metro and state pages on this site display all-items, goods, services, and rents RPP for every published metro and state. The salary calculator converts your nominal income to a national-average-equivalent using RPP.
For inflation-tracking questions, we cite CPI inline where relevant, for example, in our discussion of how rents have shifted between 2018 and 2024, where the temporal dimension matters. But the headline data on this site is RPP, because that is the right tool for the question most readers are asking.
Use the cost of living calculator to convert salaries between metros, browse metro-level RPP data to see current price levels, or read the methodology page for a deeper look at how the underlying data is constructed.
Sources and Further Reading
For the BEA RPP program documentation, see the BEA's Regional Price Parities by State and Metro Area reference page. For the BLS CPI program, see the Consumer Price Index home page. The BEA's real personal income data, which combines both measures, is available on the BEA regional data portal.
Frequently asked questions
Is RPP the same as inflation?
No. RPP measures how prices in one place compare to another at a single point in time. Inflation, captured by the Consumer Price Index (CPI), measures how prices in one place change over time. RPP is a spatial measure; CPI is a temporal one.
Can I use CPI to compare cost of living between cities?
No. CPI is not designed to compare price levels across cities. The BLS publishes regional CPIs, but they are indexed to a base period for that region and cannot be cross-compared directly. To compare price levels across metros, use the BEA's Regional Price Parities.
Why do RPP and CPI sometimes disagree about a city?
They cannot disagree because they measure different things. CPI tells you whether Boston is more expensive this year than last year. RPP tells you whether Boston is more expensive than Cleveland right now. A metro can show high inflation (CPI rising) and still be relatively affordable (low RPP) compared to other cities, or vice versa.
Which is more reliable for budgeting?
Each is reliable for what it measures. For setting next year's budget, CPI tells you how much more your current basket will cost. For deciding whether to take a job in a different city, RPP tells you how your purchasing power will change. Most personal-finance decisions need both.
Are RPP and CPI built from the same underlying data?
They share some inputs. The BEA uses BLS price data, the same surveys that feed CPI, when constructing the goods and services components of RPP. But the BEA layers on rent data from the Census American Community Survey and applies its own geographic adjustments. The two measures are related but methodologically distinct.
Sources: U.S. Bureau of Economic Analysis, Regional Price Parities; U.S. Bureau of Labor Statistics, Consumer Price Index.
Last updated: May 2026
