What Is a Real Estate Market Price Index?
What a property price index measures, why it exists when medians are already available, and how to read index values and their confidence labels.
If you already have a median price, why would you need an index? The answer is the single most important idea in property market measurement.
The problem an index solves
A median price describes the middle transaction that actually occurred in a period. That makes it an accurate description of what sold — and an unreliable measure of whether prices moved.
Suppose that this quarter, a higher proportion of large units transacted than last quarter. The median rises. No individual property became more expensive; the mix of what sold changed. This is called a mix effect, and it is pervasive in property data because the "basket" of goods changes every period.
What an index measures instead
An index tracks the price level of a defined market segment relative to a base period, which is conventionally set to 100. If the index reads 112, the price level in that segment is 12% above the base period.
The point of index construction is to control for the changing mix, so that the resulting series reflects price movement rather than composition. That is why a median and an index can move in different directions in the same month, and why neither is wrong when they do.
Base periods and why they matter
An index value is meaningless without its base period. "The index is 112" says nothing until you know 100 refers to a specific month or year.
Comparing index values from two different sources is therefore invalid unless they share both a base period and a methodology. This is one of the most common errors in market commentary.
Confidence and coverage labels
Index values for narrow segments rest on fewer records. Where a segment is thin, a published value may be technically computable but analytically weak.
A well-constructed index therefore publishes coverage or confidence information alongside the value, so a reader can tell whether a 2% movement in a small segment is a signal or noise. Treat an index value without that context with caution.
This article explains method rather than forecasting the market. For definitions and how measures are computed, see our data methodology and data coverage.