Skip to main content
FAQs

FHFA House Price Index® Frequently Asked Questions

Page last updated: October 7, 2026

The FHFA House Price Index® (FHFA HPI®) is a broad measure of the single-family house price movements in the United States. FHFA produces a suite of indexes covering different geographies and reporting frequencies. All indexes are constructed using the same technical methodology and are collectively referred to as the “FHFA HPI”.

The flagship FHFA HPI is the Purchase-Only index, which is constructed from purchase transaction data received from Fannie Mae or Freddie Mac (the Enterprises) and is seasonally adjusted. It is the most frequently referenced index in FHFA press releases, news stories, and social media. FHFA created additional indexes to address questions about house price changes in other market segments, such as the All-Transactions index, which incorporates refinance data, and the Expanded-Data index, which incorporates purchase data covering the broader single-family housing market.

The U.S. Patent and Trademark Office approved two federally registered trademarks: “FHFA House Price Index®” and “FHFA HPI®.” These trademarks cover the entire suite of indexes, and FHFA intends to protect the Agency’s branding, usage, and intellectual property. 

All indexes are available for download at https://www.fhfa.gov/data/hpi/datasets.
 

The FHFA HPI measures average price changes for sales or refinancings of the same properties over time. FHFA derives the HPI from transaction information on single-family properties with mortgages purchased or securitized by Fannie Mae or Freddie Mac. As the Enterprises purchase or securitize additional mortgages, FHFA incorporates the newly acquired mortgage data into the HPI. These new mortgage acquisitions provide additional observations that are used to identify repeat transactions for the most recent period and to update the historical index for each subsequent period back to 1975.
 

The Agency constructs the FHFA HPI suite from more than 100 million home sale and refinance transactions involving mortgages purchased or securitized by Fannie Mae and Freddie Mac. The indexes provide measures of house price changes at the national, census division, state, metropolitan area, county, ZIP code, and census tract levels. The FHFA HPI uses a fully transparent, weighted repeat-sales methodology to analyze these transaction data (https://www.fhfa.gov/research/papers/house-price-indexes-hpi-technical-description). By comparing price changes for the same properties over time and applying weights that account for the characteristics of the transactions, the methodology helps control for differences in the quality and composition of the properties represented in the sample. For this reason, the FHFA HPI is commonly described as a “constant-quality” house price index.

This distinction is important when comparing the FHFA HPI with summary statistics such as mean or median home prices. Those measures reflect changes in both the prices and the composition or quantity of housing, whereas the FHFA HPI is designed to isolate changes in house prices. As a result, the FHFA HPI provides a measure of average house price appreciation that is less affected by changes in the mix of properties represented in the market.

The FHFA HPI can be used for a wide range of housing market analyses, including updating the current value of residential property assets, estimating potential mortgage defaults and losses given default, predicting prepayment speeds for financial securities, and assessing differences in housing affordability across geographic areas. Researchers have also used FHFA HPI data in studies examining business cycles, demographic changes, environmental and disaster risks, labor markets, local government budgets, migration, political elections, property taxation, urban revitalization, and wealth creation.
 

FHFA releases HPI reports and data monthly. FHFA announces the release schedule for the following calendar year near the end of each summer. The current release dates are available on the FHFA HPI Release Dates page. Reports and data are posted at 9:00 a.m. Eastern Time on the scheduled release dates.
 

Each FHFA HPI public release includes a report and accompanying data.

The consolidated report highlights house price trends for the United States and selected regions, including the census divisions. Every three months, FHFA publishes a more comprehensive report that provides additional quarterly information for these areas and expands geographic coverage to include states and metropolitan areas. Statistics presented in the reports reference price changes measured by FHFA’s flagship Purchase-Only HPI.

The data released quarterly includes the monthly Purchase-Only HPI data as well as additional types of house price indexes. These indexes use the same general repeat-sales methodology as the Purchase-Only HPI but differ in the types of transactions or properties included in the underlying data:

  • All-Transactions HPI. Adds appraisal values from refinance mortgages to the Purchase-Only data sample.
  • Expanded-Data HPI. Expands the Purchase-Only data sample to include Federal Housing Administration-backed mortgage transactions and real estate sales records obtained from county recorder offices. These records include cash sales and purchases financed with non-conforming loans. FHFA uses the Expended-Data HPI annually to adjust the conforming loan limits, which establish the maximum original loan balance eligible for acquisition by Fannie Mae and Freddie Mac.
  • Distress-Free HPI. Excludes short sales and sales of bank-owned properties from the Purchase-Only data sample before estimating the index.
  • Annual HPI. Uses All-Transactions data to construct annual indexes for very small geographic areas, including counties, ZIP codes, and census tracts. This index should be considered developmental.
  • Puerto Rico HPI. Uses sales and refinance data for Puerto Rico mortgages provided by the Enterprises, the Federal Housing Administration, and the Federal Home Loan Bank of New York. This index should be considered developmental.
  • Manufactured Housing HPI. Uses data on conventional mortgages for single-family detached manufactured homes acquired by the Enterprises. Personal property loans are excluded from the dataset. This index should be considered developmental.

Data constraints prevent FHFA from producing every type of index for every geographic area. However, multiple index types are generally available for a given area. For example, several index types are available for individual states. Although the various indexes tend to exhibit similar long-term trends, differences among them can be more pronounced over shorter periods.
 

FHFA employs a modified version of the Case-Shiller® index methodology, using a geometric weighted repeat-sales procedure to construct the indexes. A detailed description of the HPI methodology is available at  https://www.fhfa.gov/research/papers/house-price-indexes-hpi-technical-description.

The repeat-sales procedure estimates price changes using repeat transactions on the same property units over time. This approach helps control for differences in the quality of the houses comprising the sample and is why we refer to the FHFA HPI as a “constant quality” index.

For the monthly index, we aggregate all transactions occurring within the same month and estimate a separate index value for each month. Similarly, for the quarterly index, we aggregate transactions within the same quarter and estimate index values based on the assigned quarters. 

Advanced data users should note that we do not include dummy or constant variables as additional time-varying controls.
 

Fannie Mae and Freddie Mac provide FHFA with the most current information on mortgage transactions each month. We combine these data sets to establish price differentials on properties with a history of more than one transaction. Each quarter, updates from additional data sources are merged in the same manner to construct additional index types such as the Expanded-Data index and the Puerto Rico index.

For information on how the indexes are revised, see FAQ #13. 
 

We base the Purchase-Only FHFA HPI on transactions for single-family properties involving conforming, conventional mortgages purchased or securitized by Fannie Mae or Freddie Mac. Information on FHFA HPI types that cover additional or different scopes of transactions can be found in the Overview section of this FAQs.

Conforming refers to a mortgage that meets the underwriting guidelines of Fannie Mae and Freddie Mac and does not exceed the conforming loan limit. For current and previously announced loan limits, you can find references such as a table of county-specific limits and a map at https://www.fhfa.gov/data/conforming-loan-limit.

Conventional mortgages are loans that Federal Housing Administration, the Department of Veteran’s Affairs, or other federal government entities neither insure nor guarantee.

The flagship FHFA HPI eliminates non-conforming and non-conventional mortgages when producing the index. The sample also excludes data on condominiums, cooperatives, multi-unit properties, and planned unit developments. We apply additional filters to remove property observations with potential recording errors like extremely high or low sales values, implausible appreciation changes, repeat sales within the same period, and incomplete date or physical property address information.

The Manufactured Housing FHFA HPI includes only real-property manufactured homes, as captured in data from Fannie Mae and Freddie Mac. 
 

The broad geographic scope and long history of the Enterprises’ mortgage operations enable the FHFA HPI to provide house price information for a wide range of geographies using data extending back to 1975. The FHFA HPI draws on more than 100 million transactions accumulated over several decades, providing a robust basis for measuring house price movements.

FHFA has access to these data through its role as the federal regulator of Fannie Mae and Freddie Mac. Chartered by Congress to support a reliable supply of mortgage financing for homebuyers, the Enterprises are among the largest mortgage finance institutions in the United States and hold or have securitized a significant share of the nation’s outstanding residential mortgages. As a result, their mortgage data provide broad coverage of the U.S. housing market, although they do not encompass all residential mortgage transactions.
 

For model estimation, we use the loan origination date as the relevant transaction date. It is important to note that this is different than the loan acquisition date, which reflects the date when Fannie Mae or Freddie Mac purchased or securitized the loan.​
 

Yes. Distressed transactions, including short sales and sales of bank-owned properties, are included in the FHFA HPI data used to estimate all but one of the FHFA HPI indexes. The exception is Distress-Free HPI, which specifically excludes these transactions.

FHFA introduced the Distress-Free HPI in 2012Q2, along with a Highlights article (https://www.fhfa.gov/document/d/hpi/2012q2_hpifocus_n508.pdf). The Distress-Free HPI is a version of the Purchase-Only HPI that removes short sales and sales of bank-owned properties from the transactions data used to estimate the index. This allows users to examine house price movements without these types of distressed transactions.

FHFA also examined the effects of distressed sales on the HPI in a Working Paper released in August 2013 (https://www.fhfa.gov/document/d/hpi/working-paper-13-1).
 

There is a delay between when a mortgage originated and when FHFA receives the corresponding data from the Enterprises. Loan originations typically take 30 to 45 days to reach Enterprise funding, followed by additional processing time. As a result, FHFA generally receives data on new originations with a lag of about two months.

These newly received data allow FHFA to estimate the HPI for the most recent period and lead to revisions of index values for previous periods. Data on mortgages purchased with longer lags, including seasoned loans, can continue to be incorporated into the HPI and generate revisions, particularly for the most recent periods.
 

FHFA revises historical FHFA HPI estimates as additional transaction data become available. There are three primary reasons for these revisions:

  • New repeat transactions. The FHFA HPI is based on repeat transactions, meaning that it measures changes in the value of the same property over time. When a property appears again in the data through a subsequent sale or refinance, the new observation provides additional information about the property’s price change since the previous transaction and can affect index estimates for both the current and earlier periods.
  • Purchases of seasoned loans. Fannie Mae or Freddie Mac purchased mortgages that originated in earlier periods. When these seasoned loans enter the Enterprise data, they provide additional information about transactions and property valuations from prior periods and can result in revisions to historical index values.
  • Timing of data availability. Because of the 30- to 45-day lag between loan origination and Enterprise funding, along with additional data processing time, FHFA generally receives data on new originations with a lag of about two months. These data are used to estimate the HPI for the new period and can also revise estimates for previous periods. Mortgages purchased with longer lags, including seasoned loans, can continue to generate revisions, particularly for the most recent periods.

Revisions therefore reflect the incorporation of additional information into the FHFA HPI as it becomes available.
 

No. The FHFA HPI is not adjusted for inflation and therefore measures house price changes in nominal terms. Users who want to measure house price changes in real, or inflation-adjusted, terms can adjust the FHFA HPI using an appropriate measure of consumer price inflation.

One option is the Consumer Price Index All Items Less Shelter series. For example, the Bureau of Labor Statistics series CUUR0000SA0L2 measures consumer prices excluding shelter and is available for periods extending back to the 1930s. The series, along with other Consumer Price Index measures, can be downloaded from the BLS: https://data.bls.gov/cgi-bin/srgate.
 

The two indexes use the same fundamental repeat-sales approach. A 2008 FHFA working paper examines the methodological and data differences between the two measures: https://www.fhfa.gov/document/d/hpi/revisiting-the-differences-new-explanations-rp.

For further reading, consult the FHFA HPI Technical Description at https://www.fhfa.gov/research/papers/house-price-indexes-hpi-technical-description and the Case-Shiller methodology.
 

We provide many of the FHFA HPIs in both seasonally adjusted and non-seasonally adjusted terms. We design the seasonal adjustment to remove recurring calendar year patterns or seasonal variation from the HPI that would otherwise partially confound month-to-month or quarter-to-quarter comparisons of appreciation.

FHFA uses the Census Bureau’s X-13 autoregressive integrated moving average (ARIMA) procedure, as implemented in the SAS software package. We employ the automated ARIMA model-selection algorithm in X-13.
 

Yes. The Expanded-Data index includes purchase-money mortgages from other sources (conforming and non-conforming) and cash sales to capture transactions that could potentially make up the complete single-family home market. We detail the approach to estimating the Expanded-Data HPI in the Highlights article published with the 2011Q2 FHFA HPI at https://www.fhfa.gov/document/d/hpi/2011q2_hpifocus_n508.pdf.

In general, the methodology is the same used to construct the standard Purchase-Only index, except we use a supplemented dataset for estimation. The data includes sales price information from Fannie Mae and Freddie Mac mortgages augmented by two additional sources: (1) transaction records for houses with mortgages endorsed by Federal Housing Administration and (2) county recorder data from a licensed provider. The county recorder data include records in most—but not all—U.S. counties. We include more information on the county recorder data coverage in the 2023 Technical Note available at https://www.fhfa.gov/document/d/hpi/source-update-for-county-recorder-data-fhfa-hpi-impact.pdf.
 

The manufactured housing FHFA HPI is formed using the same methodology that is used in FHFA’s traditional Purchase-Only and All-Transactions HPIs, except that rather than weighting index data from individual states, it directly pools all U.S. transactions data. This methodological difference is driven by sample size limitations prohibiting the formation of state-specific price indexes. This index should be considered developmental.

We detail the approach to estimating the manufactured housing HPI in the Technical Note published with the 2024Q2M07 FHFA HPI at https://www.fhfa.gov/document/d/hpi/mh-hpi-technical-note.pdf.
 

The FHFA HPI includes indexes for all nine census divisions, the 50 states and the District of Columbia, and every Metropolitan Statistical Area (MSA) in the United States. FHFA also produces developmental indexes for Puerto Rico and selected counties, census tracts, and three- and five-digit ZIP codes.

The Office of Management and Budget currently recognizes 387 MSAs, 13 of which are subdivided into a total of 37 Metropolitan Divisions. FHFA’s metro-level data files include indexes for Metropolitan Divisions where applicable, rather than a single index for the corresponding MSA.

For the All-Transactions and Purchase-Only indexes, FHFA releases a total of 410 metro-level indexes: 373 for MSAs without Metropolitan Divisions (excluding Eagle Pass, TX, which has a very limited sample size) and 37 for Metropolitan Divisions. For the Purchase-Only HPI, FHFA releases indexes for the 100 most populous MSAs or Metropolitan Divisions.

Beginning with the 2026Q2 release, FHFA also constructs and releases MSA-level index estimates for the 13 MSAs composed of multiple Metropolitan Divisions. These MSA-level estimates provide users with a single index for each of these MSAs in addition to the indexes for their component divisions.
 

A MSA consists of one or more counties containing an urban core, along with adjacent counties that have a high degree of social and economic integration with the core, as measured primarily by commuting patterns. Metropolitan Divisions, or Metropolitan Statistical Area Divisions, are subdivisions of some of the largest MSAs and have a similar geographic structure. MSAs are smaller geographic areas centered on urban clusters. 

Except for the annual indexes, the FHFA HPI suite reports at the metropolitan area or Metropolitan Division level. FHFA may also refer to these geographies as “cities,” although this terminology does not necessarily correspond to a single legal authority or incorporated municipality and may encompass multiple jurisdictions.
 

FHFA currently uses the MSA and Division delineations defined by the OMB Bulletin No. 23-01. For information about the current MSA delineations, please visit https://www.whitehouse.gov/wp-content/uploads/2023/07/OMB-Bulletin-23-01.pdf.

We routinely transition to newer delineations once an alignment among different data sources becomes possible. However, we do not assign properties into prior definitions that might have been used during the transaction year. Information for the prior delineations is posted on the FHFA HPI Downloadable Data page under the “Additional Data” section, “Utility Files and Background Information for Index Construction” subsection. 
 

In addition to the information displayed in the MSA tables, FHFA makes available MSA indexes. The data are available in American Standard Code for Information Exchange (ASCII) format and may be accessed at https://www.fhfa.gov/data/hpi/datasets?tab=quarterly-data. Several of the downloadable files have MSA-level HPIs.
 

There are technical subtleties involved with such linkages. ZIP codes sometimes overlap county boundaries as a single ZIP code can be located partially inside and outside of a Metropolitan Area. Thus, developing a crosswalk between ZIP codes and Metropolitan Areas is not a straightforward exercise. The Department of Housing and Urban Development has released a lookup table that maps ZIP codes to the Metropolitan Area(s) that they fall within. That lookup file, as well as a discussion of the underlying technical issues, can be found at https://www.huduser.gov/portal/datasets/usps_crosswalk.html.

Although FHFA has published HPIs for some ZIP codes, those indexes are annual (i.e., quarterly index values are not provided). Researchers needing quarterly values for ZIP codes may be interested in using index values for the applicable metropolitan area.
 

For many small- or medium-sized cities, the ZIP3 is almost synonymous with the MSA. Some published mortgage datasets show the “three-digit” ZIP code of the included properties, where it is merely the first three digits of the applicable five-digit code. For example, a property whose ZIP code is 91711 would be in the “917” three-digit ZIP code. Three-digit ZIP codes represent larger geographic areas than five-digit ZIP codes.

To aid modelers who have three-digit ZIP codes, FHFA has released a set of developmental HPIs for such areas. The underlying information we use for index production is the “all-transactions” dataset, which includes information on both home purchases and refinances. 

For more information, see the HPI Technical Note at https://www.fhfa.gov/document/d/hpi/2014q4_hpifocus_n508.pdf.
 

FHFA constructs the census division indexes from the indexes for their component states, as discussed in the Highlights article accompanying the 2011Q1 FHFA HPI release, available at https://www.fhfa.gov/document/d/hpi/2011q1_hpifocus_n508.pdf.

For the quarterly All-Transactions and Purchase-Only indexes, FHFA constructs census division indexes using the quarterly growth rates of the underlying state indexes. For the monthly indexes, FHFA uses the corresponding monthly growth rates. In each case, FHFA “builds up” the census division index by applying a weighted average of the growth rates for the component states.

FHFA sets each census division index equal to 100 in its relevant base period. For subsequent periods, FHFA increases or decreases the index based on the weighted average price change for the component states. FHFA calculates index values for periods preceding the base period in the same manner, working backward sequentially so that the growth rate between each pair of periods reflects the weighted average growth rate of the component states.

FHFA constructs the national FHFA HPI using an analogous approach, with the census divisions serving as the components. Because the census division indexes are themselves constructed from weighted state indexes, the national index is equivalent to an index based on state-level weights.

The United States is divided into nine census divisions as follows:

  • Pacific: Hawaii, Alaska, Washington, Oregon, California
  • Mountain: Montana, Idaho, Wyoming, Nevada, Utah, Colorado, Arizona, New Mexico
  • West North Central: North Dakota, South Dakota, Minnesota, Nebraska, Iowa, Kansas, Missouri
  • West South Central: Oklahoma, Arkansas, Texas, Louisiana
  • East North Central: Michigan, Wisconsin, Illinois, Indiana, Ohio
  • East South Central: Kentucky, Tennessee, Mississippi, Alabama
  • New England: Maine, New Hampshire, Vermont, Massachusetts, Rhode Island, Connecticut
  • Middle Atlantic: New York, New Jersey, Pennsylvania
  • South Atlantic: Delaware, Maryland, District of Columbia, Virginia, West Virginia, North Carolina, South Carolina, Georgia, Florida

Due to sample size limitations in the Manufactured Housing FHFA HPI, all U.S. transactions data is pooled together rather than weighting data from different states or census divisions.
 

The weights used to construct the FHFA HPI are estimates for the shares of one-unit detached properties in each state. For years in which decennial census data are available, we use the share from the relevant census. For intervening years, a state’s share is the weighted average of the relevant shares in the prior and subsequent censuses, where the weights change by ten percentage points each year. For example, California’s share of the housing stock for 1982 is calculated as 0.8 times its share in the 1980 census plus 0.2 times its share in the 1990 census. For 1983, the Pacific Division’s share is 0.7 times its 1980 share plus 0.3 times its 1990 share.

For years since 2000, we calculate state shares as follows:

  • For 2001-2005, shares are straight-line interpolated based on the state shares in the 2000 decennial Census and the 2005 values from the American Community Survey (ACS).
  • For 2006-2019, the estimates are from the annual ACS.
  • For 2020, the Census Bureau delayed the release of its 2020 ACS 1-year estimates because of the impact of the COVID-19 pandemic on data collection. You can find the full statement at https://www.census.gov/data/developers/data-sets/acs-1year.2020.html#list-tab-4MOHCRASG9G03KU0CO. We continued to use the 2019 ACS 1-year estimates.
  • For 2021 and forward, each year’s estimates are from the corresponding annual ACS or, if not available, from the most recent available ACS year.
  • The general rule is, if the latest available ACS is for the year YYYY, we use it to estimate weights for YYYY and use these weights for years YYYY+1 and YYYY+2. 

You can access the year-specific estimates of the state shares of U.S. detached housing stock at https://www.fhfa.gov/data/hpi/datasets?tab=additional-data.
 

For the quarterly index, “four-quarter” percentage change in home values is simply the price change relative to the same quarter one year earlier. For example, if the FHFA HPI release is for the second quarter, then the “four-quarter” price change reports the percentage change in values relative to the second quarter of the prior year. It reflects the best estimate for how much the value of a typical property increased over the four-quarter period. We use “one-year” and “annual” appreciation synonymously with “four-quarter” appreciation in the full quarterly FHFA HPI releases.

The “one-quarter” change estimates the price gains relative to the prior quarter. When estimating the quarterly price index, we pool together all observations within a given quarter, making no distinction between transactions occurring in different months. As such, the “four-quarter” and “one-quarter” changes compare typical values throughout a quarter against valuations during a prior quarter. The appreciation rates do not compare values at the end of a quarter against values at the end of a prior quarter.

Similar conventions apply to our reporting of “12-month”, “annual”, and “monthly” changes in the monthly FHFA HPI report.
 

The index values themselves—for the U.S., Census divisions, individual states, and metropolitan statistical areas (MSAs)— have little standalone significance. Rather, each index value derives its meaning from its relationship to preceding or subsequent index values. These relationships can be used to calculate appreciation rates.

To calculate appreciation between any two quarters, use the formula:
(QUARTER 2 INDEX NUMBER - QUARTER 1 INDEX NUMBER) / QUARTER 1 INDEX NUMBER

For ease, we made a calculator tool that is available at https://www.fhfa.gov/data/hpi/datasets?tab=hpi-calculator. 
 

The answer depends on geography and type of FHFA HPI. We normalize most of our index series to 100 for the first quarter of 1991. Notable exceptions are All-Transactions indexes, for which we maintain original base periods for historical consistency, All-Transactions indexes for metropolitan areas are set to 100 for the first quarter of 1995, and All-Transactions Indexes for states and divisions are set at 100 for the first quarter of 1980.
 

Neither. The FHFA HPI measures changes in home prices over time, while controlling for differences in the types or characteristics of homes being sold or refinanced. In other words, it is intended to capture the change in the price of housing, rather than simply the change in the average or median sale price. In 2010, FHFA published ad hoc research that describes a technique for producing state and national median and average home price statistics. The working paper is available at https://www.fhfa.gov/research/papers/approach-calculating-reliable-state-and-national-house-price-statistics.

In 2024, FHFA started publishing a quarterly median price series for the Manufactured Housing FHFA HPI, under “Summary Statistics for House Prices”, on our downloadable data page. The associated Technical Note is available at https://www.fhfa.gov/document/d/hpi/mh-medians-technical-note.pdf.

The U.S. Census Bureau, the National Association of REALTORS®, and other alternative sources, provide data on current house price levels, such as median and mean statistics.
 

No. The FHFA HPI measures average price changes in sales or refinancings on the same properties. The U.S. Bureau of Labor Statistics, which provides an owners’ equivalent rent, or private companies offer rent data across property types and sizes but those underlying sampled units may be less comparable with the homes tracked by our indexes.
 

Absolutely! The FHFA HPI data are freely available for download at https://www.fhfa.gov/data/hpi. To cite the index in an article or story, we suggest at least an attribution like “Source:  FHFA HPI®” or “Source:  FHFA House Price Index®.” Additional clarifications could be helpful to denote the type of index (Purchase-Only, All-Transactions, Expanded-Data) and whether the data are adjusted for seasonality. A more detailed citation might be “Source:  FHFA HPI® (Purchase-Only, seasonally adjusted, nominal).”