The Residential Property Price Index (RPPI) is designed to measure the change in the average level of prices paid by households for residential properties sold in Ireland. The RPPI specifically excludes non-household purchases, non-market purchases and self-builds (i.e. where the land is purchased separately). The index is mix-adjusted to allow for the fact that different types of property are sold in different months.
The RPPI is compiled in accordance with Regulation (EU) 2016/792 on harmonised indices of consumer prices and the house price index and Regulation (EU) 2023/1470 on the methodological and technical specifications as regards the house price index and the owner-occupied housing price index, and forms part of Ireland’s international obligations to provide harmonised house price indices to the European Statistical System (ESS).
The RPPI is compiled from a variety of data sources. The principal data source is stamp duty returns made to the Revenue Commissioners. All transfers of ownership of residential properties in the State must be referred to the Revenue Commissioners for stamp duty assessment under the Stamp Duties Consolidation Act (SDCA) 1999. The data collected includes the address of the property and the sales price.
These data are matched by the CSO to Building Energy Rating (BER) data, compiled by the Sustainable Energy Authority of Ireland (SEAI). Under Statutory Instrument (S.I.) No. 243 of 2012, all residential property for sale must disclose their BER assessment (with some very minor exceptions). The BER data includes the property address, the total floor area (m2) and the dwelling type (apartment, detached house, semi-detached house, etc.). The BER data are matched to the stamp duty data using an address matching algorithm (or Eircodes, where available on both datasets).
The stamp duty returns are also matched by the CSO to the GeoDirectory (again using an address matching algorithm or Eircodes). The GeoDirectory provides the Small Area code of the property, which is used to link the property to the Haase-Pobal (HP) deprivation index. The HP deprivation index measures the relative social advantage (or disadvantage) of each Small Area and serves as a useful locational characteristic in the RPPI price model.
Only stamp duty returns where a satisfactory match is made to both a BER and the GeoDirectory (currently 75% of all returns) are used in the compilation of the RPPI.
Residential properties are heterogeneous, meaning that no two houses or apartments are exactly identical. This poses a challenge when trying to construct a price index as there is a need to separate pure price change from differences in the quality and mix of the products being bought over time. Typically, this is done by comparing the prices of exactly the same products, time after time. For example, this is the method used in the Consumer Price Index, where a fixed basket of consumer goods is re-priced every month. However, in the case of residential properties, price is determined by many characteristics (location, size, dwelling type etc.) which make direct price comparisons difficult. Furthermore, only a small portion of the total housing stock is sold in any given month. The combination of these factors means that the price comparison process that would typically be used to calculate a price index cannot be used in the case of houses and apartments.
The hedonic method is the prevalent statistical process for the measurement of change of residential property price. In this method, transactions over two or more successive periods are pooled and the characteristics which influence price (dwelling type, dwelling size, geographical location and neighbourhood quality) are analysed and their relative contributions to the overall price are estimated. By excluding the price change determined by these characteristics independently, we are left with a pure price change for a consistent set of characteristics from one time period to another - or more simply - a residential property price index. This index uses a rolling 12-month hedonic regression model.
In order to mitigate short-term volatility in the series and highlight longer-term trends the published indices are smoothed using a double-exponential data smoothing technique. However, care should still be taken when interpreting monthly changes which may indicate residual short-term volatility rather than underlying change in longer-term price trends.
Double-exponential smoothing is only applied to elementary sub-indices. The aggregates are calculated from the smoothed elementary indices using chain-linked Laspeyres method, and therefore, are smoothed only indirectly.
Weights are calculated at the beginning of each year based on the total value of transactions during the previous year as given by the stamp duty data for each component index. This approach produces an annually chain-linked Laspeyres-type price index.
The national RPPI for all dwellings is split into two higher level geographical areas: Dublin and Ireland excluding Dublin (or Rest of Ireland). These regional aggregates are further broken down by type of dwelling, i.e. houses and apartments.
In addition, house price indices are provided for each of the four Dublin administrative areas:
The volume of transactions of apartments in Dublin is not sufficient to provide the same breakdown.
House price indices are also provided for seven regions outside of Dublin:
The volume of apartments transacted outside of Dublin does not permit a regional breakdown.
These house price indices by region and area of Dublin are based on relatively low levels of transactions and can exhibit significant volatility. Short-term trends for these sub-indices may be unreliable.
In addition to regional breakdown, price indices for new and existing dwellings are produced. A new dwelling is defined as a dwelling that has not previously been inhabited. The status of the dwelling is clearly indicated on the stamp duty dataset.
The indices for new and existing dwellings are computed using the same quality-adjustment and smoothing methodologies as the regional sub-indices, but on quarterly, rather than monthly, basis and at national level only. This is due to the low transaction numbers for new dwellings, which would result in a highly volatile monthly index. Pooling transactions over a quarter, rather than a month, allows for the calculation of more robust estimates of price change.
The base period for the RPPI is year 2015. Therefore, the RPPI answers the question: how much would it cost on average to purchase the same set of dwellings sold in 2015 in any given month? The annual index for the year 2015 is set to 100 and all preceding and subsequent price movements are expressed relative to this base.
The national RPPI and regional sub-indices are compiled and published on a monthly basis. Price indices for new and existing dwellings are compiled, published, and transmitted to Eurostat quarterly. Annual price indices are derived from an average of the 12 calendar monthly indices.
The movement of the RPPI is expressed as percentage change, rather than a change in index points, because index point changes are affected by the level of the index in relation to its base period, whereas percentage changes are not.
The example below illustrates the computation of a percentage change:
| Percentage Change Calculation | |
| Current Index | 79.0 |
| Less Previous Index | 80.4 |
| Equals in index points | -1.4 |
| Divided by the previous index | 80.4 |
| Equals | -0.0174 |
| Results multiplied by 100 | -0.0174*100 |
| Equals percentage change | -1.7% |
The RPPI can be used to estimate the updated value of an individual dwelling provided that a prior value is known subsequent to January 2005. Simply multiply the sale or valuation price by the current relevant index and divide by the index at the date of sale/valuation.
For example, consider a house sold in South Dublin in June 2010 for €220,000.
| South Dublin house price index June 2010: | 88.9 |
| South Dublin house price index June 2022: | 144.0 |
| Estimated value in June 2022: | €220,000 x (144.0 / 88.9) = €356,355 |
This estimate provides an approximate value only, based on aggregate price movements of all dwellings. It presupposes that there has been no material change in either the dwelling or its neighbourhood in the period concerned. The CSO takes no responsibility for calculation of value based on the RPPI.
The legal deadline for submitting a stamp duty return is within 44 days of the date of execution (in some cases, returns are submitted later than this). In practice, therefore, only a fraction of the transactions executed in any given month are available for index compilation the following month. Rather than delay the RPPI, preliminary results are prepared based on the early returns. These preliminary results are updated the following three months as new transaction data for the reference month becomes available. Therefore, the RPPI results for the latest three months are provisional.
The RPPI is a relatively new statistical product compiled from third party data sources. Whilst every endeavour will be made to retain the consistency of the RPPI, there is always a risk that changes in the availability of the source data at some future point may impact the RPPI methodology. New data sources may also become available, enabling qualitative improvements in the RPPI methodology. In these exceptional cases, a full revision of the historic RPPI series may be necessary. Any such revisions will be signalled at least 3 months in advance and the fullest information possible on the nature of the revision will be made available to users.
The RPPI is accompanied by an extensive range of additional statistical information on the residential property market. Four principal statistics are provided; volume, value, mean price, and median price.
Volume is the number of dwellings transacted (note than more than one dwelling can be purchased in a single transaction).
Value is the total value of all dwellings transacted (in millions of euro).
Mean price is the value divided by the volume.
Median price is the price threshold separating the most expensive half of transaction prices from the least expensive half of transaction prices. Median prices are obtained by ranking all transactions from the most expensive to the least expensive. The price that ranks exactly in the middle is the median price.
These four indicators are available for both household and non-household buyers, for market and non-market sales.
Mean prices are generally higher than median prices. This is because mean prices are often inflated by the sale of individual high value properties. For example, consider a neighbourhood where just ten houses are sold in a particular month. Say, nine of these houses are sold for €100,000 and the tenth is sold for €1,100,000. The median price of these transactions will be €100,000. However, the mean price will be substantially higher at €200,000.
Mean price and median price provide answers to two different questions.
Mean price answers the question: What would be the common price of a group of properties if their value was shared equally amongst them?
Median price answers the question: What price is paid by the typical buyer purchasing property in this group?
The additional indicators for households are further available broken down by household buyer type: first-time buyer owner-occupiers, former owner-occupiers, and other household buyers (non-occupiers, for example buy-to-let).
The additional indicators are available broken down by county. In addition, a more detailed geographical breakdown is available for household market transactions by Eircode Routing Key (the first three characters of the Eircode). Note that the descriptors applied to these Eircodes are not official labels. However, they generally follow pre-existing postal town names.
A breakdown by dwelling type (house or apartment) is also available for household market transactions. In approximately one in every four cases, the dwelling type has been imputed based on its other known characteristics. Therefore, these statistics should be seen as estimates rather than definitive results.
The compilation of the additional indicators is affected by the progressive availability of stamp duty returns. Accordingly, two sets of indicators are presented, executions and filings:
Execution statistics are the definitive guide to residential property sales. However, they are necessarily incomplete for a period (until all the relevant stamp duty returns have been filed). Execution statistics are preliminary and subject to revision until 12 months have elapsed.
Filing statistics only represent administrative activity. However, filing statistics are not subject to revision and therefore may serve as useful lead indicators for broader developments in the residential property sector.
There is no specific data field in the stamp duty data which indicates whether the purchaser is a household or a non-household. However, there are two data fields that capture the name of the purchaser, a first name field and an 'entity' name field (which captures either a surname or a company/institution name). The recording of a first name is taken as indicating a household purchase (i.e. one or more private individuals is involved in the purchase). The absence of a first name is taken as indicating a non-household purchase.
Non-household transactions are residential dwelling transactions made by private companies, charitable organisations, and state institutions. The annual Non-Household Sector chapter of the RPPI release was turned into a stand-alone release as of 2023, entitled Residential Property Transactions by Non-Households. The latest release, as well as all previous versions, can be found in the Residential Property Transactions by Non-Households Previous Releases section. The Residential Property Transactions by Non-Households release provides a breakdown of non-household transactions by NACE sector and territory of the participating organisations. Further, it contains a breakdown of transactions between and within the household and non-household residential property sectors.
Explore the average property price, breakdown of buyers and trend of sales over time
Learn about our data and confidentiality safeguards, and the steps we take to produce statistics that can be trusted by all.