Statistical Insights: Blowing bubbles? Developments in house prices
Few economic indicators make the newspapers’ front pages. One that often does though is house prices. This is because, as witnessed during the financial crisis, movements in house prices can have a direct impact on households’ wealth and economic growth. At the same time, fluctuations in economic activity can also drive trends in house prices. House price indicators are therefore among the indicators that are closely monitored by policymakers.
However, despite their importance, until recently, largely reflecting a variety of conceptual and measurement differences across countries, no harmonised internationally comparable measure of house prices existed. In 2013, a new statistical handbook on house price indices was endorsed by several international organisations, and since then the OECD has been working with countries to develop a new internationally comparable database on house prices.
The new data confirm the positive association between house prices and economic activity but they also reveal significant differences in the strength of the link across countries, especially in the wake of the recent financial crisis.
There is a positive correlation between fluctuations in house prices and in economic activity…
As Figure 1 shows, fluctuations in real house prices (i.e. adjusted for general inflation) and economic activity in the OECD are positively related (with a correlation coefficient of around 0.6 over the period 1971 to 2015). This relationship reflects three drivers, that may differ in intensity and over time: a leading component, as the wealth effects from increases in real estate prices can boost final consumption of home owners, through re-mortgaging for example; a lagging component, as stronger economic growth may boost house prices; and a co-incident component as both house prices and economic activity may be explained by the same underlying factors, such as credit market conditions and population growth.
Note: The real house price index for the OECD area is computed from real house price indices for the 35 OECD countries, weighted using their nominal GDP weights in PPP terms. This real house price index is sourced from the OECD Analytical house price indicators dataset and real GDP from the OECD Quarterly national accounts database.
…but the relationship may have weakened over time…
Understanding the dynamic contribution these drivers make over time is clearly of interest, especially as they provide insights on the potential build-up of vulnerabilities stemming from strong household spending growth driven by rising leverage and inflated asset prices.
The latest edition of the OECD’s Economic Outlook provides evidence of a post-crisis weakening of the relationship between house price growth and the propensity to consume, in part reflecting the changes in financial regulations and credit standards introduced after the crisis which have reduced the ability of households to use rising housing values as collateral for additional borrowing to fund current spending.
…and it differs across countries.
A number of factors influence house price movements, including real household incomes, real interest rates, mortgage market regulations and supervision, lending patterns (at fixed rate versus variable rates), tax relief on mortgage debt financing, and transaction costs such as stamp duty. Therefore, differences in institutional arrangements combined with differences in economic activity may explain heterogeneity in housing markets across countries.
The internationally comparable house price indices from the OECD database show that the relationship between house prices and economic activity is indeed stronger in some countries than others. For example, in countries like Finland, Ireland, Japan and the United Kingdom, fluctuations in house prices and economic activity are closely related, with a correlation coefficient of around 0.7 from 1971 to 2015, whereas it is much weaker in countries like France, Italy and Norway, with a correlation coefficient lower than 0.3 over the same period.
Similarly, while house prices dropped in many countries at the time of the crisis, factors that affect house price developments differ markedly across countries. Figure 2 shows how real house price developments have diverged significantly across countries since 2005. Notwithstanding some differences within each group, four broad groups of OECD countries can be distinguished:
- An initial fall in real house prices followed by a subsequent rebound in New Zealand, the United Kingdom and the United States.
- A continuous increase in real house prices pre and post crisis in Australia, Mexico and Sweden.
- A severe and prolonged fall in house prices post the crisis with only recent signs of stabilisation in Greece and Spain.
- Relatively stable house prices since 2005 in Belgium and Korea.
In respect of the above, it should also be noted that there may be significant differences across local housing markets within countries. Unfortunately internationally comparable data at sub-national level are typically not available.
Note: House price indices for individual countries are sourced from the OECD RPPI – Headline indicators dataset and deflated by the Consumer Price Indices (CPIs) for all items. The real house price index for the OECD area is sourced from the OECD Analytical house price indicators dataset.
The measure explained
House price indices, also called Residential Property Prices Indices (RPPIs), are index numbers measuring the rate at which the prices of all residential properties (flats, detached houses, terraced houses, etc.) purchased by households are changing over time. Both new and existing dwellings are covered if available, independently of their final use and their previous owners. Only market prices are considered. They include the price of the land on which residential buildings are located.
Where to find the underlying data
The OECD database on house prices is available on OECD.STAT and includes the three following datasets:
- Residential Property Price Indices – Headline Indicators: This dataset covers OECD members and some non-member countries. For each country, the RPPI that is available at the most aggregate level has been singled out in this dataset, but due to data availability, headline indicators are country specific. For example, the RPPI at the most aggregate level for the United States only covers single-family dwellings and not all types of dwellings as is the case for most other OECD countries.
- Residential Property Price Indices – Complete database: This dataset contains nominal house price indices with various breakdowns for OECD members and some non-member countries. Headline indicators are a subset of this complete dataset.
- Analytical house price indicators: This dataset contains, in addition to nominal RPPIs, information on real house prices, rental prices and the ratios of nominal prices to rents and to disposable household income per capita. It should be noted that for Brazil, Canada, China, Germany, the United States and the Euro area, the datasets “Analytical house price indicators” and “Residential Property Price Indices (RPPIs) – Headline Indicators” do not refer to the same nominal price indices. These differences are further documented in country-specific metadata that are attached to this dataset.
In the future, the OECD database on house prices will include other housing-related indicators in order to provide a more comprehensive picture of real-estate markets.
ILO, IMF, OECD, UNECE, Eurostat, World Bank (eds.), (2013), Handbook on Residential Property Price Indices, Eurostat, Luxembourg