<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>Official rentals for housing (rent) price inflation statistics are of considerable public interest. Matched-sample estimators, such as that used for nearly two-decades in New Zealand (2000–2019), require an unrealistic assumption of a static universe of rental properties. This article investigates (1) a property fixed-effects estimator that better reflects the dynamic universe of rental properties by implicitly imputing for price change associated with new and disappearing rental properties; (2) length-alignment simulations and property life-cycle metrics to inform the choice of data window length (eight years) and preferred splice methodology (mean-splice); and (3) stock-imputation to convert administrative data from a ‘flow’ (new tenancy price) to ‘stock’ (currently paid rent) concept. The derived window-length sensitivity findings have important implications for inflation measurement. It was found that the longer the data window used to fit the model, the greater the estimated rate of inflation. Using administrative data, a range of estimates from 55% (window length: three-quarters) to 127% (window of 90-quarters) were found for total inflation, over the 25-years to 2017 Q4.</p>
</abstract>

Official rentals for housing (rent) price inflation statistics are of considerable public interest. Matched-sample estimators, such as that used for nearly two-decades in New Zealand (2000–2019), require an unrealistic assumption of a static universe of rental properties. This article investigates (1) a property fixed-effects estimator that better reflects the dynamic universe of rental properties by implicitly imputing for price change associated with new and disappearing rental properties; (2) length-alignment simulations and property life-cycle metrics to inform the choice of data window length (eight years) and preferred splice methodology (mean-splice); and (3) stock-imputation to convert administrative data from a ‘flow’ (new tenancy price) to ‘stock’ (currently paid rent) concept. The derived window-length sensitivity findings have important implications for inflation measurement. It was found that the longer the data window used to fit the model, the greater the estimated rate of inflation. Using administrative data, a range of estimates from 55% (window length: three-quarters) to 127% (window of 90-quarters) were found for total inflation, over the 25-years to 2017 Q4.

Rentals for Housing: A Property Fixed-Effects Estimator of Inflation from Administrative Data

Journal of Official Statistics

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Official rentals for housing (rent) price inflation statistics are of considerable public interest. Matched-sample estimators, such as that used for nearly...

Rentals for Housing: A Property Fixed-Effects Estimator of Inflation from Administrative Data

Pubblicato online: 29 mar 2022

Pagine: 187 - 211

Ricevuto: 01 dic 2019

Accettato: 01 dic 2021

DOI: https://doi.org/10.2478/jos-2022-0009

Parole chiave
Price indexes, multilateral index number methods, quality change, fixed-effects mean-splice, rolling window indexes

© 2022 Alan Bentley, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Rentals for Housing: A Property Fixed-Effects Estimator of Inflation from Administrative Data

Pubblicato online: 29 mar 2022

Pagine: 187 - 211

Ricevuto: 01 dic 2019

Accettato: 01 dic 2021

DOI: https://doi.org/10.2478/jos-2022-0009

Parole chiavePrice indexes, multilateral index number methods, quality change, fixed-effects mean-splice, rolling window indexes

© 2022 Alan Bentley, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Parole chiave
Price indexes, multilateral index number methods, quality change, fixed-effects mean-splice, rolling window indexes