Why the FHFA index differs from online home value estimates
Property websites show a dollar figure for a specific address. This site shows an index for an area. They answer different questions, from different data, and it is normal for them to disagree.
Different question
An automated valuation asks: what would this house sell for today? A repeat-sales index asks: on average, how much have prices changed for the same homes in this area between two dates? The first is a level for one property; the second is a rate of change for many. A ZIP code index of +45% over ten years does not mean any particular house rose 45%, only that the sampled repeat transactions in the ZIP did on average.
Different data
- The FHFA index uses sales prices and refinance appraisals on homes with conforming mortgages bought or guaranteed by Fannie Mae or Freddie Mac. Cash purchases, jumbo loans and most FHA and VA loans are outside the standard sample.
- Automated valuations typically use public records, listings, tax assessments and a model of how features (size, beds, lot, age) relate to price, updated as listings and sales arrive.
Different timing
FHFA publishes quarterly indexes about two months after the quarter and the annual ZIP, county and metro indexes once a year, and revises them. An online estimate can change daily. A comparison of the two at any moment compares different reference dates.
Different geography
ZIP codes are postal routes, counties are administrative units and metro areas are groups of counties; none of them is a housing market by design. FHFA's small-area indexes also thin out where transactions are few, which is why some ZIP series stop early or skip years. An estimate for a single address can draw on comparable sales across those lines.
What each is good for
An index is the better tool for history and comparison: how an area moved through 2008, how it compares with its county over the same decade, how far it fell from a peak and when it regained it. An estimate is a starting point for one property, to be tested against an appraisal or a market analysis. This site publishes the first and does not attempt the second.