Housing

The median house price: what it measures and what it omits

What a median house price actually measures, why it moves when the mix of dwellings sold changes, and how to read it alongside a mix-adjusted index.

Ask a news bulletin what happened to house prices and you will usually be handed a median: the middle sale in a period's list of transactions, half above and half below. It is a useful, honest summary of one thing — the price at which a typical home actually changed hands — and it is routinely asked to do a job it cannot do. This note sets out what a median house price measures, why it moves for reasons that have nothing to do with any individual home, and how to read it alongside an index.

What the median is and what it ignores

A median is a positional statistic. Line up every recorded sale in a place and a period from cheapest to dearest, take the middle one, and its price is the median. Unlike a mean, it is not dragged around by a handful of very expensive sales, which is why property commentators prefer it. But it is a fact about the set of dwellings that sold, not about the housing stock, and still less about any one house.

That distinction carries the omissions. The median says nothing directly about quality: if the properties sold this quarter are newer, larger or better located than those sold a year ago, the median can rise even if not a single home has become more valuable. It is silent on the mix of dwellings — a market that shifts toward apartments will report a lower median without any apartment being worth less than before. It ignores incentives bundled into a transaction, such as developer rebates or chattels, which can move the recorded price while the underlying economics do not move. And it is an aggregate: a country can post a rising median because its largest city is transacting more expensive homes while smaller centres are flat or falling.

Why medians move (mix effects)

Economists call the first group of distortions composition or mix effects, and they are the main reason a median can drift away from the trend in values. A single quarter's median is the outcome of two things moving at once: the price of a representative dwelling, and the mix of dwellings that happened to sell. Only the first is what most readers think they are being told.

Seasonality adds a predictable version of the same problem. Sales in some months skew toward particular types of property, so month-to-month comparisons of a raw median can reflect the calendar more than the market. The safest use of a median is therefore a comparison of like with like — the same area, the same broad property type, the same period a year apart — which is harder than it sounds and is exactly why statistical agencies build price indexes in the first place.

How to use it alongside an index

An index such as the OECD analytical house-price series that Younivi republishes is designed to strip out mix. It tracks the change in prices for a consistently defined basket, expressed on a fixed base. The price-to-income series below is set to 100 in 2015, so 101.3 for New Zealand in December 2025 means prices were a little above their 2015 relationship with incomes, while Spain's 132.5 at March 2026 means a much larger move. Read a median for the level at which the market is clearing, and an index for the direction and size of the change; neither replaces the other.

House price-to-income index, 100 = 2015 House price-to-income index, 100 = 2015 OECD analytical house price series. 2015=100 basis; higher means less affordable. New Zealand 101.3 United States 125.1 Australia 125.1 United Kingdom 104.5 index, 2015=100
House price-to-income index (2015=100), latest OECD observations. Source: OECD analytical house price series, republished by Younivi. Full series on the House Price desk.

The same discipline applies to the ratio shown. A price-to-income index is itself a ratio of two aggregates, so it inherits the limits of both: it uses average incomes and average prices, and it can be quiet about distribution and about borrowing costs. The value of pairing it with a median is that one is a transaction level and the other is a mix-adjusted relationship — when they tell the same story the reading is more robust, and when they diverge the gap is usually composition.

Where this fits in the wider picture

House prices sit at the centre of household balance sheets, so the temptation to read one number as a verdict on the whole market is strong. The median is where that temptation does the most damage, because it is easy to compute, easy to quote and easy to misunderstand. A rising median is often reported as if every owner became richer by that amount, when in fact it describes the middle transaction and can move on the mix of what was sold.

Indexes exist because the underlying question — how much more expensive has housing become? — is genuinely hard. They trade a little simplicity for comparability, and they make cross-country and long-run reading possible on a common base. That is why a reference desk carries a level-style series alongside ratio and change measures rather than a single headline. Reading them as a set is the honest way to describe a market that no single statistic can capture.

How to read this on Younivi

The House Price desk publishes the index, the year-over-year change and the price-to-income ratio for each economy, with the source citation on the page, and the all-countries view ranks every licensed economy. Companion notes on this site cover the index base and the price-to-income ratio in more detail. Figures are republished official data, may lag the producer, and are not financial advice about any property decision.