The Central Statistics Office has today released a rebased (2015=100) Residential Property Price Index (RPPI) along with the results for February 2020. The rebased index represents price change with reference to the annual average price for the year 2015, rather than the previous base period of January 2005. It is also accompanied by a number of methodological improvements in how the RPPI is compiled.
The reference period for each of the underlying sub-indices and aggregate indices have been updated to 2015=100. For example, if the value of an index was 100 in January 2005, 80 in 2015 and 120 in January 2020 at base January 2005 = 100, the corresponding values at base 2015 = 100 will be:
The relative changes between any two periods will, thus, remain the same after the change, e.g. 120/100 = 150/125 = 1.2 for the example above.
The RPPI continues to publish 13 sub-indices: 11 regional house price indices and 2 indices for apartments inside and outside of Dublin. The aggregate RPPI indices are compiled by weighting the monthly price changes of the sub-indices. The weights are based on the total value of the household market transactions in the previous year in the each of the market segments.
Full details of the data sources and methods used to compile the RPPI are included in a newly updated Technical Paper on the RPPI methods page (linked below*)
To facilitate users to transition to the new base period, the RPPI series to base January 2005 will continue to be compiled by the CSO for a limited period. The relevant tables in CSO’s StatBank (HPM06 and HPA06) will be updated each month.
To reduce volatility and to better identify turning points in the indices, the RPPI is smoothed using the Holt-Winters double-exponential method. From January 2020, smoothing will be applied to the sub-indices with no additional smoothing applied to aggregate indices, i.e. indirect smoothing method. Previously, for the reference years 2005 to 2019, both the sub-indices and the aggregates were smoothed directly. The benefit of using indirect smoothing is that the aggregate indices can be computed as weighted averages of the price changes of the sub-indices. Therefore, the additivity of the RPPI is maintained.
For the RPPI series at base January 2005 = 100, price changes (one, three and twelve-month change) were calculated using index values rounded to one decimal figure. For the rebased series (2015 = 100), price changes are calculated from unrounded index values, which provide more accurate estimates. In some cases, this modification results in the published rates of change being revised by up to 0.2 percentage points.
If you have any questions in relation to the methodological changes, please contact a member of the RPPI team in the CSO.