1 Introduction

In April 2014, the RPPs and the price-adjusted estimates of regional personal income became official statistics of the United States Bureau of Economic Analysis (BEA), and will be published annually on a regular basis (Aten and Figueroa 2015). The estimates are based on Consumer Price Index micro data which are not publicly available and on rent data from the Census that are accessible to users.Footnote 1 In addition, the methodology adopted BEA’s national personal consumption expenditure (PCE) price index as the measure of aggregate annual price change for the U.S. An aggregate price index for the U.S. is necessary in order to express the current dollar RPP-adjusted incomes in each region in constant dollars. The results shown here are for the latest year, 2013, and highlight RPPs for several expenditure categories and real per capita income growth and levels.

There have been many studies on measuring or controlling for regional price differentials within countries, see for example, the comprehensive literature review in Deaton and Dupriez (2011) and Biggeri and Laureti (Biggeri et al. 2015). Geographically large countries such as India (Deaton 2003; Deaton and Tarozzi 2005), Brazil (Deaton and Dupriez 2011, Aten 1999), Indonesia (Arndt and Sundrum 1975), the United States (Koo et al. Koo 2000), and Australia (Wingfield et al. 2005; Waschka et al. 2003) as well as geographically smaller ones like the United Kingdom (Fenwick and O’Donogue 2003), Germany (Roos 2006), the Philippines (McCarthy 2010), and Italy (Biggeri and Laureti . 2009; Biggeri et al. 2015), appear to have significant sub-national differences.

One of the major challenges both across and within countries is to obtain consistent and representative price data from multiple sources and outlets, and to match these observations with weights that reflect actual consumption patterns. There are services that are particularly difficult to compare, such as construction, housing, education, medical and government services, and the documentation and research on handling these categories of consumption is extensive, dating back to the origins of the ICP (Kravis et al. 1982).

The Consumer Price Index calculated by most national statistical offices is the most obvious source of price data, but it is often collected in a manner that makes it difficult for estimating spatial price differences. Researchers have turned to private survey companies, to scanner data and internet scraping projects, but their data are not necessarily representative of the entire country or of the average consumer. Deaton (2006), Deaton and Dupriez (2011), and more recently the World Bank (2015), have matched detailed consumer expenditure household surveys with ICP product specifications, thus tracking both the spending patterns of particular populations and their relative price levels.

Another motivation for obtaining accurate measures of regional price levels is their relevance to assessing poverty levels and income inequality indicators within countries. The former has a long history in the U.S., beginning with landmark study in the 1960s (Orshansky 1969). More recently, following a report by the National Academy Panel on Poverty and Family Assistance (Citro and Kalton 2000), the Census Bureau began publishing research and estimates on alternative poverty measures that take into account geographic price differences (Short and O’Hara 2008, Renwick 2011, 2015).

The BEA, in a joint project with the Bureau of Labor Statistics (BLS), first estimated regional price parities for consumption goods and services for 38 metropolitan and urban areas of the U.S. for 2003 and 2004 (Aten 2005, 2006). These areas, for which BLS produces the Consumer Price Index (CPI), represent about 87 % of the total population. The method was expanded to cover the remaining nonmetropolitan portions of each state. Estimates for 2005 and 2006 were reported in the Survey of Current Business in November 2008 (Aten Aten2008, Aten and D’Souza 2008). Experimental estimates for 2007 incorporate the multi-year American Community Survey (ACS) from the Census Bureau, as do official estimates for 2008 forward.Footnote 2

The methods and results differ from versions prior to 2014 because they include a two-stage, rolling average estimation process. The first stage estimates annual multilateral price level indexes for CPI areasFootnote 3 and for several consumption expenditure classes such as apparel, food and transportation. In the second stage, the price levels and expenditure weightsFootnote 4 are allocated from CPI areas to all counties in the United States.Footnote 5 They are then recombined for regions, such as states and metropolitan areas, and merged with data on rents from the Census Bureau’s American Community Survey (ACS). The ACS provides a broader geographic coverage than the CPI areas, including county-level data, thus allowing us to augment the allocated CPI price levels with observed housing observations.

The final RPPs are obtained by stacking the 5 years of the first-stage results, plus the annual rent indexes, and calculating the multilateral aggregate price index for all goods and services and rents. For example, the 2010 RPP is a 5-year average of the 2008–2012 CPI-derived price indexes for goods and services excepting rents, plus the 2010 rent indexes from the ACS.

The following sections describe in more detail the use of the price levels and expenditure data from the CPI and the housing data from the ACS, how their geographies are reconciled, and how the overall indexes are computed.

2 Price Levels for CPI Areas

CPI price data cover a wide array of consumer goods and services, ranging from high-expenditure goods, such as new automobiles, to low-expenditure services, such as haircuts. Over a million price quotes are collected each year and are classified into more than 200 item strata, each consisting of detailed Entry Level Items (ELIs), which may be further divided into Clusters. The item strata can be combined into nine expenditure groups: apparel, education, food, housing, medical, recreation, rents, transportation and other goods and services.Footnote 6

Because the CPI was not designed to measure geographic price level differences, items with identical characteristics are not always priced in all areas. Therefore, for the ELIs in the 75 highest item strata (accounting for roughly 85 percent of expenditure weights), we estimate hedonic regressions which take into account the variation in the characteristics of the sampled items.

For the “Women’s Tops Excluding Active and Outerwear” Cluster, for example, we use a hedonic price model to adjust for the type of clothing (jacket, sweater or blouse), the fiber content, the length of the sleeves, the closure type, the size range, the brand category (exclusive/luxury, national or private), country of origin, and the type of outlet where it was sold. An example of an item-specific hedonic regression may be found in Aten (2006).

For the remaining item strata, we use a shortcut approach consisting of a weighted regression on areas and ELIs (and Clusters when available) as independent variables. This regression is a modified version of the weighted Country-Product-Dummy (CPD-W) approach as originally described in Summers (1973), and more recently in Rao (2004), Sergeev (2004), and Diewert (2005). Overall results do not differ greatly whether detailed hedonic regressions are run on all item strata, or on only the top 75 in combination with this shortcut approach (Aten 2006).

After the ELI price levels are estimated, they are aggregated to yield item strata price levels, again using the CPD-W approach, with weights corresponding to the importance of the ELIs within the item strata. Both the ELI and the item strata price levels undergo an outlier checking process described in detail in Aten et al. (2011). It is modeled after the Quaranta tables first used in the OECD-Eurostat basic heading level price comparisons.Footnote 7 We flag observations that are (1) either very large or small relative to the mean in that area and ELI; (2) that are either large or small relative to the variance of the ELI observations; or (3) are large or small once they have been adjusted for the relative price level of the area. It is an iterative process that looks at the raw price data as well as the relative prices after the hedonic adjustment.

The outcome of this process—hedonic and CPD-W regressions - is a matrix of 38 by 200-plus area-item price levels, with the weighted geometric average of the 38 areas in each item strata indexed to one. These are price relatives, not prices, for each area-item combination. Any of the 38 areas could be the numeraire, but their geometric average provides a more neutral interpretation.

In order to obtain overall price and quantity indexes that combine the 38 by 200-plus area-item price levels with their corresponding expenditure weights, there are several aggregation methods that could be used. The most well-known are the EKS-Törnqvist and Fisher indexes, the CPD-W approach, the GAIA index and the Geary index. The advantages and disadvantages of these methods and their corresponding statistical and economic properties have been thoroughly reviewed elsewhere, see Kravis et al. (1982), Diewert (1999) and Balk (2009), for example. Aten and Marshall (2010) have shown that the Geary RPPs for the 38 BLS index areas do not deviate significantly from the other possible indexes.Footnote 8

The Geary indexes, in addition to being base-invariant and transitive, are additive, or matrix consistent. The latter means that the values for any item will be directly comparable between areas, and the values for any area will be directly comparable across items. In other words, an overall index for Food can be obtained directly from the values for lesser aggregations within Food, such as Meat, Vegetables and Dairy, and so forth. For the RPPs, we aggregate the 200-plus items to 16 expenditure classes, consisting of Food, Apparel, Transport, Housing, Education, Recreation, Medical, Other and Rents. Each of the classes is broken down into Services and Goods, with the exception of Apparel, with only a Goods component, and Rents, with only a Services component.

The Geary multilateral price level index, P Geary , is given by:

$$P_{Geary}^{c} = \frac{{\mathop \sum \nolimits_{n = 1}^{N} \left( {pq} \right)_{n}^{c} }}{{\mathop \sum \nolimits_{n = 1}^{N} \pi_{n} q_{n}^{c} }}$$
$$\pi_{n} = \mathop \sum \limits_{c = 1}^{M} \frac{{\left( {pq} \right)_{n}^{c} }}{{P_{Geary}^{c} \mathop \sum \nolimits_{d = 1}^{M} q_{n}^{d} }}$$

where: p is the relative price of the item stratum or expenditure class; π is the national average price of the item stratum or expenditure class; q is the notional quantity equal to (pq)/p; c and d are the BLS areas, which take a value of 1 through M = 38; n is the item stratum, which takes a value of 1 through N = 200 +.

In this first stage, we estimate a P Geary , for each of the 38 areas and for each of the 16 expenditure classes.Footnote 9 In the next stage, we combine these price level indexes with expenditure weights, including weights that represent rural regions not in the CPI.

3 Regional Price Parities for States and Metropolitan Areas

The second stage consists of expanding the Geary indexes from the geography of the 38 BLS areas to other more commonly used geographies, such as states and metropolitan areas. It is also here that the 16 price levels for the expenditure classes are augmented by more detailed Rent price levels and expenditure data from the Census Bureau, and where an overall index combining all 16 classes into one ‘All Item’ RPP index is calculated.

The process begins with the allocation of price levels and expenditure weights from CPI areas to counties. Price levels for each county are assumed to be those of the CPI sampling area in which the county is located. For example, counties in Pennsylvania are assigned price levels from either the Philadelphia or Pittsburgh areas or form the Northeast small metropolitan area. Rural counties are not included in any of the 38 urban areas for which stage one price levels are estimated, therefore these counties are assigned price levels of the urban area that (1) is located in the same region and (2) has the lowest population threshold.Footnote 10

Expenditure weights in the second stage include CPI data for rural regions, and thus in combination with the 38 urban areas, cover all U.S. counties.Footnote 11 Weights are allocated from each CPI area and rural region to the component counties in proportion to household income.Footnote 12 The county-level allocations undergo two adjustments. First, rent weights are replaced with directly observed rent expenditures from the 5-year ACS file plus imputed owner-equivalent rent expenditures. The latter are obtained as follows:

  1. 1.

    The ratios of monthly tenant rents to owner-equivalent rents in the BLS CPI housing file are estimated for several types of housing units, from studio apartments to detached houses with three or more bedrooms. The components of these ratios, that is, tenant rents and owner-equivalent rents for each housing type, are the weighted geometric means of all the observations in the CPI.

  2. 2.

    The ratios are applied to the observed unit rents in the ACS, resulting in an estimated monthly owner-equivalent rent value for each housing type, by county;Footnote 13, Footnote 14

  3. 3.

    This estimated owner-equivalent rent value is multiplied by twelve and by the number of owner-occupied housing units in each type, resulting in an annual estimate of owner-occupied housing expenditures, by county.

Note that the ratio of tenant rents to owner-equivalent rents is across all BLS sampling areas, that is, there is only one vector of ratios for each housing type. The same ratio is applied to different geographies in the ACS file, with only the distribution of rents and number of units varying across geographies.Footnote 15 Total expenditures by tenants and owners are equal to the sum of the observed annual rent expenditures and the estimated owner-occupied expenditures from step 3 above.

The second adjustment to the county level weights is to control the national shares of the 16 expenditure classes to BEA’s personal consumption expenditure shares. This yields weights consistent with BEA’s national accounts.Footnote 16 The adjustment shifts the distribution of weights across expenditure classes, notably reducing the share of rents expenditures from total consumption in the United States from 30.2 to 20.6 %. Conversely, the shares of Food, Recreation, Education and Apparel and Other goods and services are higher in BEA’s national accounts.

Once the price levels and expenditure weights have been obtained for each class, year and county, we stack the 5 years and then obtain their weighted geometric means.Footnote 17 Rents are treated separately: instead of using the BLS-CPI rents, we insert a new set of Owner Equivalent Rent expenditures, described above, as well as Rent price levels, described below.

Rent price levels are estimated directly from tenant rent observations in the ACS: annually for states, and across 3 years for metropolitan areas.Footnote 18 No imputation of owner-occupied rents is used in the price levels, instead we use rent price levels for both renters and owners.Footnote 19, Footnote 20 The rent price level s are quality-adjusted estimates using a hedonic model that controls for basic unit characteristics such as the type of structure, the number of bedrooms, the total number of rooms, when the structure was built, whether it resides in an urban or rural location, and if utilities are included in the monthly rent.Footnote 21 Additional research comparing rent price levels using the ACS and CPI Housing surveys is available in Martin et al. (2011).

The second multilateral aggregation uses the 5-year weighted geometric mean of the 15 expenditure classes derived from the BLS CPI, together with the 1-year state level rents and 3-year metropolitan area rents from the Census ACS to estimate the final all items RPPs. This implies that for the reference year 2011, for example, final state-level RPPs are composed of rent price levels in 2011 plus an average of the price levels for goods and services other than rents between 2009 and 2013. Table 1 below illustrates the rolling average combinations of CPI and ACS data.

Table 1 Reference years and source data
Full size table

The second stage Geary multilateral is analogous to the first stage, but instead of M = 38 areas and N = 200-plus item strata as inputs and Geary price indexes for 16 expenditure classes as outputs, the inputs are 51 states or 383 metropolitan areas, and N = 16 expenditure classes. The resulting P Geary , is a single ‘All Items’ price index for each state and each metropolitan area. However, because of the additive properties of the Geary, we are able to obtain the sub-aggregates of the 16 expenditure classes very easily, and do so for several groups: Goods, Services, Rents, as well as eight subgroups—Food, Apparel, Transport, Housing (excluding Rents), Education, Recreation, Medical and all Other goods and services.

4 Using RPPs to Estimate Real Personal Income

An important application of the RPPs is to control for price level differences across regions when measuring economic activity such as income levels. The price level differences measured by the RPPs are specific to one point in time. At BEA, we make an additional adjustment to convert the regional current dollar values to constant values, resulting in price-adjusted regional incomes at chained dollars, which we call “real” personal income.Footnote 22

Real personal income in chained (2009) dollars for a region is the current-dollar personal income divided by its RPP for a given year (equal to current dollar income in regional prices), divided by the U.S. PCE price index, which converts the current dollar value to 2009 chained dollars. For the U.S., the nominal and real personal income totals will equal each other in 2009, while the regional nominal and real personal incomes will differ only by the RPP of each region.

An Implicit Regional Price Deflator (IRPD) is derived as the product of the RPP times the national PCE price index. It is equal to personal income divided by real personal income. The growth rate of the IRPD is the change in RPPs between 2 years times the change in the U.S. PCE price index (see Box 1).

Box 1 Implicit price growth rates
Full size table

5 Selected Results and Conclusions

The results discussed below are constrained to the most recent years for which we have estimates, 2012 and 2013, and to the 51 states.Footnote 23 Additional geographies (Metropolitan Statistical Areas or MSAs, and the non-metropolitan and metropolitan portions of states) and additional years (2008–2011) are available.Footnote 24

5.1 Regional Price Parities

Table 2 shows the overall or All Items RPPs for 2013 for each state. The All Items RPP is subdivided into three groupings: Goods, Rents, and Services other than Rents, followed by eight additional subgroups: Food, Apparel, Transport, Housing (exclusive of Rents), Education, Recreation, Medical, and Other goods and services.

Table 2 Regional price parities by State, 2013
Full size table

The overall range between lowest and highest RPP is 30.9 between Mississippi (86.8) and the District of Columbia (117.7). Hawaii (116.2) and New York (115.3) are the next highest RPPs. The range for Rents is much wider: 95.8 between Hawaii (158.7) and Arkansas (62.9). With over 20 % of the average expenditures, Rents have the greatest impact on the overall RPP, but Services, excluding Rents, also has a relatively wide spread of 24.8, between South Dakota (91.1) and New Jersey (115.9). Medical and Education also have a large spread, which may reflect the variance in some of the comparison-resistant services, such as Hospital Services and Education Tuition and Fees.

Some of the predominantly rural states, or states with just one or two larger metropolitan areas, such as Alabama, Arkansas, Georgia, Idaho, Indiana, Iowa, Kansas, Kentucky, Louisiana, Missouri, Montana, Nebraska, New Mexico, Ohio, Oklahoma, South Carolina, Tennessee, Wyoming and West Virginia have low rent price levels, and thus also relatively low All Items RPPs, and relatively higher RPPs for Goods compared to Services.

However, the RPPs for the subgroups or RPPs with the exception of Rents are five-year averages of CPI price levels. Thus year to year differences during this five-year period (2009–2013), will only reflect shifts in the expenditure weights, and not actual price level differences of goods and services other than Rents.Footnote 25 Each of the subgroups corresponds to an expenditure class defined by BLS, and reflects prices and weights from a survey sample that distinguishes only 38 metropolitan and urban areas in the CPI. They are then allocated to counties and recombined to states and MSAs (see Sects. 2, 3). The components in each expenditure group consist of a market basket of goods and services and are described in the CPI Handbook of Methods.Footnote 26

Table 3 summarizes the RPPs by dividing the 50 states and the District of Columbia, into thirds based on a ranking of their All Items RPPs. Each tercile contains 17 states, and the RPPs shown in this table are the unweighted geometric means of their 2013 RPPs.

Table 3 Summary of state RPPs 2013
Full size table

The first row of Table 3 shows the breakdown of the mean RPPs for the top 17 states, and these are all above 100, with mean Rents RPPs reaching 122.9, much higher than the All Items RPP of 107.2. This is not true of the second and third rows, showing the middle and bottom terciles, where the Rents RPPs are lower than the All Items RPPs. If we look at the ratio of the Goods RPPs to Other Services RPPs, it is less than one for the top tercile and above one for the other two groups, a pattern that is also found in the international price comparisons literature: the price level of services tends to move in the same direction as that of rents, whereas goods will generally become relatively less expensive as the overall price level increases.

5.2 Total Personal Income

Table 4 shows the impact of applying RPPs to current dollar nominal incomes for the five highest and five lowest states in terms of total current dollar incomes.Footnote 27 The first three columns of Table 4 are the nominal personal incomes for 2012 and 2013 in millions of current dollars, and the percent change. The middle columns are the real incomes in constant 2009 dollars, together with the corresponding real percent growth for each state. The implicit regional price deflators and growth are shown in the next columns, followed by the RPPs in each year. The U.S. price index rose 1.2 % between 2012 and 2013, while total personal incomes grew by 0.8 % in real terms and 2.0 % in nominal terms for the United States.

Table 4 Real personal income and implicit regional price deflators by State, 2012 and 2013
Full size table

California, Texas, New York, Florida and Illinois have the largest total incomes in both current dollars and in real, price adjusted 2009 dollars. The smallest total incomes in current dollars are in Vermont, Wyoming, Alaska, North and South Dakota, but in real terms, Delaware and the District of Columbia supplant the Dakotas, in large part because the RPPs for Delaware and the District of Columbia are above 100 (101.3 and 117.6 in 2013 respectively). In contrast, South Dakota has an RPP of 87.6 and North Dakota 91.3.

The relationship between real income growth and price growth can be seen in Fig. 1, where real personal income growth is plotted on the vertical axis and the implicit price growth is on the horizontal axis. There is a downward trend, as higher price growth is correlated with lower real personal income growth. The axes are centered on the U.S. price growth of 1.2 % and real income growth of 0.8 %.

Fig. 1
figure 1

State total real personal income growth and implicit price growth 2013

Full size image

States with a high implicit price growth tend to have a low or negative real personal income growth. North Dakota is a prime example, a rural state which has had a surge in rents following a boom in natural gas exploration. Although the Rents RPP is still lower (91.4) in North Dakota than the U.S. average, it has seen a high increase compared to its neighbors, such as South Dakota where Rents RPPs are at 66.6. This means that a decline in nominal income translates to a much larger decline in real incomes. North Dakota has also seen a large population growth, and thus the decline in per capita terms is even greater.

The states on the upper right quadrant of Fig. 1, shown as red squares, have above average real income growth and relatively high implicit price growth rates. They are Florida, Nebraska, Georgia, Washington, Oregon, Utah, Texas, Kansas and Colorado.

North Dakota, on the lower right quadrant, has experienced a large decline in real income growth (< − 4 % between 2012 and 2013), in part because of its high price growth. The decline is greater than in other areas because its nominal income also declined in 2013.

5.3 Per Capita Personal Income

Table 5 shows the states with the five highest and lowest nominal and real per capita personal incomes. The range of RPPs means that the range of current dollar incomes is reduced when adjusted by the RPPs. DC has the highest nominal per capita income at US$ 75,329 but it drops to US$ 59,688 in real (2009 dollar) terms. Mississippi has a per capita nominal income of U$ 33,913, the lowest of the states, but it increases to US$ 36,441, and is no longer among the lowest five states in rank.

Table 5 Real per capita personal income by state, 2012 and 2013
Full size table

New Jersey and New York are among the five highest in nominal terms, but are replaced by North Dakota and Wyoming in real per capita terms. Conversely, South Carolina, West Virginia and Mississippi are among the five lowest per capita incomes in nominal dollars, but when converted to real terms, Hawaii, Arizona and Utah take their place.

Figure 2 shows the relationship between the RPPs and per capita personal incomes for 2013. The RPPs are plotted on the vertical axis against the real per capita personal incomes. The RPPs are in natural logs to more easily interpret the regression coefficients on the trend lines (the U.S. with an RPP = 100 is also plotted on the horizontal axis).Footnote 28

Fig. 2
figure 2

State RPPs and real per capita personal income

Full size image

6 Concluding Remarks

The RPPs currently reflect differences in the price levels of consumer goods and services. They are constrained by the price data available from the CPI survey conducted by the BLS and by the rent data in the ACS from the Census Bureau. A major constraint of the CPI survey is the sampling design that is limited to 38 urban and metropolitan index areas and specifically targeting time-to-time consistency rather than spatial consistency. Thus the robustness of the RPPs would benefit from a place-to-place survey of the goods and services sampled by the CPI, particularly for hard to measure items such as education, food and medical services, and in geographic areas that are sparsely populated and less well-represented in the national surveys, such as rural areas. Augmenting the price observations, possibly by web-scraping and using third-party sources of information might also provide additional consistency checks on the RPPs.

The RPPs reflect the common perception of the relatively high price level in states with large metropolitan areas, such as California, Illinois, Maryland, Massachusetts, New Jersey, New York, Virginia and the District of Columbia. The smaller states in the east and west coasts also have higher rents, for example, New Hampshire, Vermont, as well as Hawaii and Alaska (small only in terms of population, not size). Hawaii and Alaska additionally have Goods price levels which are above the U.S. average, due to their geographical isolation and the added transportation and distribution costs. In the case of Illinois, although rents in Chicago are high, there are many rural counties where the rents are lower, and this is reflected in the Goods RPP being slightly higher than the Services RPP. As shown in Table 3, Fig. 2, the relative price level of Services tends to increase with increasing incomes and with higher Rents, and mirrors the tendency in international comparisons of real incomes and PPPs.

An important extension of this work is to explore the development of RPPs that reflect more than consumption goods and services, such as investment and government price differences. In international comparisons, the price level of consumption is often a good approximation for GDP price levels from the expenditure side. This is because the relative prices of investment and government change systematically in opposite directions when measured across per capita incomes. It is not clear whether this pattern would be found across states or metro areas within a country, but it seems worth examining. One approach to this would be to see if there is a geographic pattern in the prices of inputs and outputs related to construction, producers’ durable equipment and government compensation.

A separate issue with respect to Rents is how to reconcile the Personal Consumption Expenditure (PCE) weights in the national accounts with the expenditure weights in the Consumer Expenditure (CE) survey. A partial concordance was done by Blair in 2012, but it would be helpful to produce a full and updated mapping for use in the RPPs. Figueroa et al. (2014) have compared the sensitivity of the RPPs to different relative weights, and because the national share of rents out of total expenditures is significantly lower in the PCE than in the CE, this has a significant impact on the RPPs.

Another important research area is the treatment of owner-occupied housing. There are no observed owner-occupied housing observations in the CPI, only imputed values derived from rental housing observations and adjustments for utility cost. These imputed values reflect the shelter flow-of-costs, a concept that has been extensively documented and explained elsewhere (Poole et al. 2005; BLS 2011).

We have attempted to augment rental observations with the Census’ ACS information on housing owner-costs, but the information of housing values is self-reported and extremely volatile. It is possible to obtain selected monthly owner costs that include mortgages and taxes and insurance, for example, but without knowing land prices, mortgage terms and rates, it is difficult to obtain house value estimates (Garner and Short 2009). Several online real estate companies do collect actual transaction values in many markets in the US, and if these values could be linked to broad geographies such as states and metropolitan areas, it might be possible to estimate actual owner-occupied housing price levels. At the very least, we would be able to track both the BLS and ACS rental price levels with owner-occupied housing estimates, and better understand and explain sources of differences.

Lastly, it is not clear whether prices in rural areas for items other than rents are higher or lower than in urban areas, but we currently assume they are the same. The expenditure weights vary, but the trade-off between for example, transport costs and rents, are not included in this analysis. Aten and Marshall (2010) looked at alternative estimates of RPPs using a demand-based model to allow for some substitution across expenditures goods, such as transportation and rents, but the theoretical gains in precision of such a model are offset by the need for broad assumptions about consumer behavior. More data on the prices of goods and services (other than rents) in rural or nonmetropolitan areas would add to the robustness of our current estimates.

7 Data Availability

Real personal income data, regional price parities, and implicit regional price deflators are available through the BEA website. Data are available for 2008 to 2013 for states, state metropolitan and nonmetropolitan portions, and metropolitan areas at www.bea.gov.

To access the data, select the “Interactive Data” tab at the top of the homepage. At the next screen, select “GDP and Personal Income” under Regional Data. Data are available in two formats through these links:

  • Begin using the data: interactive tables where users specify data type, region and time period.

  • Download complete data sets: flat files accessed through Real Personal Income and Regional Price Parities menus.

For further information about these data, email the Regional Prices Branch at rpp@bea.gov.