

The aim of this Handbook is to facilitate the settingup of residential property price indices in countries where these are still missing and the improvement of existing price indices where this is deemed necessary. It is designed to give practical guidance on the compilation of house price indices, both in developed and less developed countries, and to increase international comparability of residential property price indices. It explains the different user needs, gives details on data and methods that can be used to compile residential property price indices and provides recommendations. The production of the Handbook was funded and supported by Eurostat.

Residential property is both a source of wealth and, insofar as property owners live in or on their property, an important determining factor in their cost of living. The price of a house is something different from the cost of dwelling services it provides, though the two concepts are obviously interlinked.

There are many areas of society where individuals or organisations use residential property price indices (RPPIs) directly or indirectly either to influence practical decision making or to inform the formulation and conduct of economic policy. Different uses can have a significant impact on the preferred coverage of the index and also on the appropriate methodology applied for its construction.

What makes the construction of a residential property price index (RPPI) so challenging? This question was addressed in Chapter 1 of this Handbook but it will be useful to remind readers about the main problems, which are as follows:
 The compilation of price indices typically relies on matching the prices for identical items over time. However, in the housing context, each property has a unique location and usually a unique set of structural characteristics. Thus, the matched model methodology will be difficult or impossible to apply.
 Transactions are sporadic.
 The desired index number concept may not be clear, or put another way, there are several distinct purposes for which an RPPI is required and, broadly speaking, different purposes require different indices.
 For some purposes, notably the construction of national balance sheets and the estimation of user costs of owner occupied housing, a decomposition of a property price into land and structures components is required but it is unclear how best to accomplish such a decomposition. This issue will be discussed in more detail in Chapter 8 below.

The simplest measures of house price change are based on some measure of central tendency from the distribution of house prices sold in a period, in particular the mean or the median. Since house price distributions are generally positively skewed (predominantly reflecting the heterogeneous nature of housing, the positive skew in income distributions and the zero lower bound on transaction prices), the median is typically used rather than the mean. As no data on housing characteristics are required to calculate the median, a price index that tracks changes in the price of the median house sold from one period to the next can be easily constructed. Another attraction of median indices is that they are easy to understand.

The hedonic regression method recognizes that heterogeneous goods can be described by their attributes or characteristics. That is, a good is essentially a bundle of (performance) characteristics. (1) In the housing context, this bundle may contain attributes of both the structure and the location of the properties. There is no market for characteristics, since they cannot be sold separately, so the prices of the characteristics are not independently observed. The demand and supply for the properties implicitly determine the characteristics’ marginal contributions to the prices of the properties. Regression techniques can be used to estimate those marginal contributions or shadow prices. One purpose of the hedonic method might be to obtain estimates of the willingness to pay for, or marginal cost of producing, the different characteristics. Here we focus on the second main purpose, the construction of qualityadjusted price indices.

The repeat sales method was initially proposed by Bailey, Muth and Nourse (1963). They saw their procedure as a generalization of the chained matched model methodology applied by the pioneers in the construction of real estate price indices like Wyngarden (1927) and Wenzlick (1952). The bestknown repeat sales indices are the Standard and Poor’s/CaseShiller Home Price Indices in the US, which are computed for 20 cities (Standard and Poor’s, 2009). The Federal Housing Finance Agency (FHFA) also computes a repeat sales index for the US, using a slightly different approach. Residex and the UK Land Registry compute repeat sales indices for Australian cities and for the UK, respectively.

As was mentioned in previous chapters, the matched model methodology to construct price indices, where prices of identical items are compared over time, cannot be applied in the housing context. One of the reasons is the low incidence of resales and the resulting change in the composition of the properties sold. The repeat sales method, which was discussed in Chapter 6, attempts to deal with the quality mix problem by looking at properties that were sold more than once over the sample period. However, using only repeatsales data could be very inefficient since all single sales observations are “thrown out” and could also lead to sample selection bias.

In Chapter 3 it was mentioned that for national accounts and CPI purposes, it will be useful or necessary to have a decomposition of the residential property price index (RPPI) into two components: a quality adjusted price index for structures and a price index for the land on which the house is built. The present chapter outlines how hedonic regression can be utilized to derive such a decomposition. Hedonic regression methods were discussed in Chapter 5.

In practice, because of the high cost of undertaking purposedesigned surveys of house prices, the methods adopted by statistical agencies and others to construct residential property price indices have mainly made use of administrative data, the latter usually being a function of the house price data sets generated by a country’s legal and administrative processes associated with buying a house. The indices so constructed can vary according to the point in the house purchasing process at which the price is measured. For example, the final transaction price or the earlier valuation used for securing a loan could be used as the “price” of the property. Furthermore, different administrative data sets will generally collect information on different sets of characteristics associated with the sales of the properties. These differing information sets will generally affect index compilation methods, often acting as a constraint on the techniques available to quality adjust for houses of different sizes, locations, etc. Thus data sets have historically acted as a constraint on index construction.

In practice, the methods used for constructing residential property price indices can be constrained in large part by the nature of the data available. The data required to construct the target index, once defined, are not always available on a regular and timely basis, if at all. Moreover, even where suitable data are available to construct a price index to meet the needs of one set of users, more often than not, the data does not fit the requirements of another set of users. For many countries setting up the required infrastructure and procedures for the collection of the data necessary for producing a property price index can sometimes be prohibitively costly. Also, changes in methodologies and in the underlying data sources can frustrate the construction of historical series, which are often required for econometric modelling and analyses over more than one cycle of housing market developments to inform policy options for the management of the economy. Last but not least, the timeliness and frequency of the data, when available, may not be suitable for producing the kind of house price index that the users want or need.

The purpose of this chapter is to provide additional empirical examples dealing with the construction of house price indices based on the methods that were outlined in Chapters 59. These are broadly defined as follows: measures of central tendency (mean or median), hedonic regression methods, repeat sales methods, and methods based on appraisal data. The following three sections of this chapter illustrate how the first three classes of methods can be implemented on very small data sets. Hopefully, working through these simple examples will enable readers to more readily follow the rather terse algebraic descriptions of the various methods that were provided in Chapters 59.

This handbook provides detailed and comprehensive information on the compilation of residential property price indices (RPPIs). It provides an overview of the conceptual and theoretical issues that arise, explains the different user needs for such indices and gives advice on how to deal with the practical problems that statistical offices are confronted with in the construction of such indices. Earlier chapters cover all relevant topics including: a description of the different practices currently in use; advice on the alternative methodologies available to the compiler; and the advantages and disadvantages of each alternative. The purpose of this chapter is to draw together all this information and make recommendations on best practice for compiling residential property price indices, including how to improve international comparability. The recommendations necessarily take into account the different situations countries are confronted with in terms of data availability and therefore cannot be too prescriptive.


