Price Adjustment Using Regional Price Parities

New RPPs for 2022 and Revisions Back to 2008

In December 2023, the U.S. Bureau of Economic Analysis (BEA) released new 2022 estimates of regional price parities (RPPs) for states, state portions, and metropolitan areas. RPPs for these geographies were revised from 2008 to 2021. The revised estimates incorporated improvements to the methodology for estimating housing services.

The release included two price-adjusted series that were estimated using the RPPs: real personal consumption expenditures (PCE) for states and real personal income for states, state portions, and metropolitan areas. For each, new estimates were released for 2022 and revised estimates were released for 2008 to 2021.

This article describes the RPPs and their use in the price adjustment of regional PCE and personal income. Per capita results for 2022 are presented, including RPP-adjusted and real estimates.1 Recent methods improvements are discussed, and their impacts on the RPPs are measured.

RPPs are spatial price indexes that measure price level differences across U.S. regions for one period. They are estimated annually and express each region’s price level relative to the U.S. average, which serves as the reference and is equal to 100. Each set of annual RPPs is independently estimated and cannot be compared across years.

RPPs are derived by directly comparing detailed price and expenditure microdata across all regions and product categories.2 The RPP for all items covers all consumption goods and services. RPPs are also estimated for four subcategories: goods, housing rents, utilities, and other services.

Another example of spatial price indexes is purchasing power parities (PPPs), which measure differences in price levels across countries for a given period and can be used to convert estimates of per capita gross domestic product into comparable levels in a common currency. BEA’s RPPs compare regions within the United States, without the need for currency conversion.

State RPP results for 2022

In 2022, state RPPs ranged from 112.5 in California to 86.6 in Arkansas (table 1). The District of Columbia’s RPP was 112.8. These results show that California’s price level was 12.5 percent higher than the U.S. average and 29.9 percent higher than in Arkansas.

States in the top RPP quintile were generally located in the Northeast and West and had relatively high housing rents and utilities RPPs (charts 1 and 2). California had the highest housing rents RPP across states at 160.2. The District of Columbia’s housing rents RPP was 177.2. Hawaii had the highest utilities RPP across states at 201.2, more than double the national average.

States in the bottom RPP quintile were all in the South and Midwest and had relatively low housing rents RPPs. West Virginia and Arkansas had the lowest housing rents RPPs at 53.9 and 55.8, respectively.

BEA’s regional estimation of real PCE and real personal income uses two sets of price indexes. Current-dollar estimates are first adjusted by the RPPs. These results are then converted to real, constant-dollar estimates using the U.S. PCE price index.

RPP-adjusted PCE and personal income

RPP-adjusted results are obtained by dividing each region’s current dollar estimates by its RPP. This controls for geographic price level differences within the period and allows for more meaningful regional comparisons of expenditures and income.

For example, in 2022, per capita personal income for the District of Columbia was $95,970, more than double that of Mississippi, which was $46,370 (table 1). However, the District of Columbia’s RPP of 112.8 was higher than that of any state, while Mississippi’s was the second lowest, at 87.3. After adjusting with the RPPs, the difference in their per capita estimates narrows, and the District of Columbia’s result ($85,044) was only about 60 percent higher than Mississippi’s ($53,099).

State RPP results show a positive relationship between the price levels and income: in general, as RPP-adjusted per capita income increases so do the corresponding RPPs (chart 3). Similar results are also seen in international results obtained using PPPs.3

In 2022, per capita PCE for California was $60,272, more than 40 percent higher than for Arkansas, which was $42,245. California’s RPP of 112.5 was the highest across states, while Arkansas’ was the lowest, at 86.6. After adjusting with the RPPs, the difference in their per capita PCE estimates narrows, and California’s result ($53,589) was only about 10 percent higher than Arkansas' ($48,783).

States with higher per capita PCE tend to have higher RPPs and are adjusted downwards, while those with lower per capita PCE tend to have lower RPPs and are adjusted upwards, narrowing the range. This is seen in the per capita PCE results, where the range across states drops from $46,055 before adjustment to $30,537 afterwards.

Real PCE and personal income

The RPP-adjusted estimates are converted to constant dollars by division with the U.S. PCE price index. The real results can also be obtained by dividing current-dollar estimates by BEA’s implicit regional price deflator (IRPD), the product of each region’s RPP, and the U.S. PCE price index.4 The IRPD links the independent sets of annual RPPs across years with a reference equal to the U.S. price level in 2017. The year-to-year change in the IRPDs is a measure of regional price change over time.

Within each period, the IRPD retains the RPPs’ directly measured geographic relationships. However, the IRPD’s time-to-time measurements are indirect, depending on comparisons of regional IRPDs linked across years through the U.S. PCE price index. They are not based on detailed comparisons of regional price and expenditure data across years.5 In 2022, the year-over-year percent change in the IRPD increased across all states, ranging from 11.8 percent in New Hampshire to 3.6 percent in Alaska (table 2).

In 2022, real per capita PCE increased in 42 states and District of Columbia (chart 4). Alaska had the highest growth rate at 5.8 percent, and New Hampshire had the largest decline at 3.7 percent. These states had the lowest and highest percent change in their IRPDs, respectively. In general, higher growth rates in the IRPDs are associated with lower annual percent change for real per capita PCE (chart 5).

Real per capita personal income declined in 46 states and the District of Columbia (chart 6). North Dakota had the highest growth rate at 2.8 percent, and Rhode Island had the largest decline at 8.7 percent. North Dakota’s IRPD change was 3.8 percent, and Rhode Island’s was 9.2 percent.

The real growth rates for PCE and personal income are impacted by the growth of the current-dollar estimates and the IRPD.6 In 2022, current-dollar PCE grew 9.2 percent, faster than the 6.4 percent growth rate in the U.S. PCE price index, and most states had increasing growth in their real estimates.7 On the other hand, current-dollar personal income only grew 2.0 percent in 2022, slower than the U.S. price growth, and most states had declining rates of change in their real estimates.

The RPPs released in December 2023 incorporate an improvement to the estimation of housing rent and utility expenditures. Since 2021, BEA has used the American Community Survey’s public use microdata sample to create a streamlined and integrated approach to measuring housing services across all BEA’s data programs. That approach includes an adjustment to tenant rents observations if the reported rent includes utility costs. For these observations, the included utility costs are imputed and subtracted from the rent expenditure.

The improvement more precisely imputes these utility costs. Previously, the imputation was based on reported utility payments for similar housing units, without regard to how many utilities were consumed. The new method bases the imputed cost on the specific combination of utilities consumed. For example, if the tenant rent for a housing unit includes electricity and water costs, the imputation is based on reported utility payments in similar housing units that consume only those two utilities.

Revisions to RPPs from 2008 to 2016 have a larger range than those from 2017 to 2021 (chart 7). All years incorporate the improvement described above, but the earlier period also carries back revisions originally introduced in December 2022.8 The revisions from 2017 to 2021 primarily reflect the utility imputation improvement. These are small and do not exceed 1.0 percent in any year (table 3).


Footnotes

  1. RPP-adjusted estimates are presented here for methods illustration and are not available on BEA’s website.
  2. For more information on the estimation of RPPs, see “Regional Price Parities, Real Personal Consumption Expenditures, and Real Personal Income” on BEA’s website.
  3. For an analysis of the relationship between PPPs and income, see “What are PPP adjustments and why do we need them?” on the Our World in Data website.
  4. The price adjustment of personal income requires a small rebalancing. See “Regional Price Parities, Real Personal Consumption Expenditures, and Real Personal Income” on BEA’s website.
  5. The growth rate of the IRPDs will not necessarily equal the region or metropolitan area price deflators published by the Consumer Price Index (CPI) program at the U.S. Bureau of Labor Statistics. These deflators are calculated directly from extensive CPI price and expenditure microdata. The IRPDs are based on indirect comparisons of regional price levels that incorporate data from the American Community Survey, the Energy Information Administration, and BEA’s PCE by state results, in addition to the CPI microdata.
  6. The implicit price deflator for the United States is equal to the U.S. PCE price index, with a base of 2017.
  7. For more information on the state estimates of current-dollar PCE and personal income, see BEA’s Interactive Data Application and select “Regional Estimates.”
  8. This is the second of two improvements to the imputation. For information on the first, see Bettina Aten and Eric Figueroa, “Revisions to Source Data for Regional Price Parities: Updates Back to 2017,” Survey of Current Business (April 20, 2023).