Methodology

The results presented in this publication are based on a number of data sources:

• The Person Income Register
• Residential Property Price Index dataset 
• Stamp Duty participant file
• DSP Child Benefit dataset 

The linkage and analysis was undertaken by the Central Statistics Office (CSO) for statistical purposes in line with the Statistics Act, 1993 and the CSO Data Protocol [1] .

All data sources are pseudonymised prior to linking. The Personal Public Service Number (PPSN) is a unique number that enables individuals to access social welfare benefits, personal taxation and other public services in Ireland. The CSO removes the PPSN and replaces it with a Protected Identifier Key (PIK). The PIK is an encrypted and randomised number used by the CSO to enable linking of records across data sources and over time, while at the same time preserving privacy.

All results are in the form of statistical aggregates which do not identify any individuals.

Data sources 

Person Income Register (PIR)

The PIR is an income register held internally within the CSO. It contains information on income received by individuals relating to employment, self-employment and social transfers. It is derived from pseudonymised versions of administrative data holdings held by the Revenue Commissioners and Department of Social Protection. The PIR provides a near complete picture on individual level income, for a calendar year. 

Residential Property Price Index (RPPI)

The RPPI dataset covers all market purchases of houses and apartments by households, both cash and mortgage-based transactions. Non-market transactions (e.g. family transfers) and non-household purchases (e.g. purchases by private companies or institutions) are specifically excluded from the RPPI index and this analysis. Also excluded are self-builds (i.e. where the land is purchased separately) and purchases of partially built dwellings. The main RPPI dataset is created from these administrative holdings – Stamp Duty returns, Building Energy Ratings (BER), the Geodirectory and the Pobal Deprivation Index [2].

Stamp Duty Participant Section

The participant section of the Stamp Duty form provides information on individuals purchasing a property. For this analysis, the PPSN is extracted and also the address string provided. The PPSN is converted to a PIK and the address string is utilised to extract a broad geographic location (county or country if not in Ireland).

DSP Child Benefit 

This file is used to identify buyers with a child at the time of the purchase.

Definitions

Type of Transaction

A Sole Transaction is defined as one named purchaser on the Stamp Duty return.

A Joint Transaction is defined as two or more named purchasers on the Stamp Duty return.

Type of Buyer

There are four types of buyer defined:

• Sole buyer with child/children: one named purchaser on the Stamp Duty return who is claiming child benefit at the time of purchase
• Sole buyer without children: one named purchaser on the Stamp Duty return who is not claiming child benefit at the time of purchase
• Joint buyer with child/children: two or more named purchasers on the Stamp Duty return and at least one is claiming child benefit at the time of purchase
• Joint buyer without children: two or more names purchasers on the Stamp Duty return and none of them are claiming child benefit at the time of purchase.

Gross Income

Gross income is defined as the sum of these three items in the year before the filing or execution of the Stamp Duty return:

• Employee income (PAYE income)
• Self-employed profit and Irish rental income (both from ITForm11)
• Social transfers (DSP payments)

Gross income for a sole transaction is the sum of the three items above for the purchaser. For a joint transaction, gross income is the sum of the three items above for all purchasers on the Stamp Duty return.

Unknown 

To calculate full aggregates, a figure for ‘unknown’ is included, which refers to records who could not be assigned to a cohort.

Reference period

Statistics are produced in relation to the year a property transaction was filed with the Revenue Commissioners and the year the property transaction was executed. The ages of the people involved in the transactions are based on the year and month of when the transaction was filed or executed.

Gross income data, (defined above), relates to the calendar year before the Stamp Duty transaction was filed or executed. For example, for a Stamp Duty transaction filed or executed in 2016, the gross income used is for the reference year 2015. Therefore, transactions in year t use income data for year t-1. This was done because there was not enough information for 2019 incomes, (particularly for the self-employed), when this report was compiled.

If no Gross Income information is available in the calender year prior, gross income is set to missing.

Target and covered population 

The target population relates to all (participants) individuals who are involved in the RPPI’s definition of market-based household purchases over the period.

The actual population covered was smaller than the target population because of two main types of missing data – unique identifiers and demographic/income information.

Missing Unique Identifiers

When unique identifiers are absent, it may not be possible to assign a PIK.

Missing Demographic and Income Information 

When a PIK is not assigned, then income variables and demographic characteristics, (such as age or gender) can’t be assigned to records. This means income characteristics can also not be assigned. In some cases, even where a PIK is assigned, it is still possible for income to be missing. Reasons for this can include that the person could:

• have income outside the scope of the Gross income defined above
• be working outside Ireland with income declared in another jurisdiction
• be starting employment in Ireland.

For all these reasons above, the covered population is less than the target population.

In aggregated figures presented in this report, there are several residual individuals who cannot be analysed. Therefore, tables which analyse age, income and identify cohorts with and without children are based on the covered population rather than the target population. As noted, when counts of individuals by different characteristics (age, gender or recorded location) are ‘presented’ a residual category (Unknown) is provided in order to provide a full aggregation for users.

Note that individuals without a PIK are assumed not to have children in the statistics on volume and median price in the ‘Volume and Price’ chapter.

Between 2010 and 2019, 95% of individuals who bought a property were assigned a PIK. This varied from a low of 89.7% in 2010 to a high of 99.8% in 2019. This variation also occurred across regions. Therefore, care should be taken when examining statistics in certain years, but also with certain lower level breakdowns. 

Due to some records having a missing PIK, it was decided to only provide a breakdown of 'Buyer Type' from 2016 onwards. Moreover, in order to have a consistent income definition over a longer time series, income statistics are only produced from 2012 onwards. 

Limitations of analysis

Please note the following limitations of the analysis in this publication.

• Care must be taken in interpreting data over the period 2010-2015 due to missing PIKs. Statistics such as the median age can potentially be distorted because of records which are not randomly missing. For example, older females were more likely to be missing a PIK. Statistical methods to correct this (i.e. imputation) were not successful and therefore no correction was made.
• When analysing trends (changes in prices, income or age) over periods and in different regions there may be increased variablity at lower level breakdowns due to lower number of transactions. 
• When comparing median price values across time, region or cohort, users should be aware that there can be compositional impacts due to the mix of properties. 
• Remember when examining income and price data, that some purchasers may use other means besides income when funding a purchase, (for example, selling another property or using savings).
• This analysis includes a varied mix of purchasers (first-time buyers, previous owners, buying-to-let etc.).
• When examining sex cohorts care must be taken on their classifications. Not all joint transactions are couples. For example, a joint male/female transaction could be a couple but could also be a brother and sister or a parent and child. Similarly, a joint male or joint female transaction could represent a couple, or it could be two brothers, two sisters or a parent and child.
• There is potential that the addresses given by buyers may not be the address they live in at the time of purchase. Therefore, care must be taken when examining the location of purchasers at time of purchase.

Statistical disclosure control

This report includes statistics based on counts from the stamp duty participant database of properties purchased and the number of individuals who purchased a property.

Statistics published on income and prices have values rounded to the nearest hundredth. Statistics published on counts have been rounded to the nearest tenth. 

Also, all count figures are rounded independently and therefore totals are more accurate than the sum of rounded sub-items. (Adding up the sum of rounded sub-items will give a different figure than the total for a category.)

[1] https://www.cso.ie/en/aboutus/lgdp/csodatapolicies/csodataprotocol/

[2] https://www.cso.ie/en/media/csoie/methods/residentialpropertypriceindex/RPPI_Technical_Paper.pdf 

Go to next chapter: Contact Details