For the last three decades, Eurostat and OECD have worked together in producing and publishing Purchasing Power Parities (PPPs) for their respective member countries. PPPs are essential tools for the comparison of price and volume levels of GDP and other indicators. The common programme is called the "Eurostat-OECD PPP Programme".

The Eurostat-OECD PPP Programme was established in the early 1980s to compare on a regular and timely basis the GDPs of the Member States of the European Union and the Member Countries of the OECD. This remains the purpose of the Programme, although its coverage has been broadened to include countries that are not members of either the European Union or the OECD. These are either countries that have applied to join the European Union or the OECD or countries with which Eurostat and the OECD have programmes of technical cooperation in statistics. A brief history of the Programme can be found in Annex I.

Eurostat and the OECD make international volume comparisons of GDP from the expenditure side. With the exception of education and a part of housing, the volumes are estimated indirectly with purchasing power parities (PPPs). PPPs are used because they are measures of relative prices. Prices are easier to observe directly than are volumes and measures of relative prices usually have a smaller variation than measures of relative volumes. Eurostat and OECD comparisons start as price collecting exercises, though other data, such as data on GDP expenditures, have also to be collected before they can be concluded. This chapter identifies the price and other data needed by Eurostat and the OECD to make their comparisons. It starts by defining the three principle concepts: consistency, comparability and representativity. It explains why volume comparisons are given priority over price comparisons and why the three concepts are not fully compatible in practice. It finishes by identifying the actual data that participating countries are required to supply.

When the Eurostat-OECD PPP Programme was established in the early 1980s, only OECD Member Countries were covered, benchmark comparisons were made every five years and all data were collected over an 18 month period centred on the reference year. Organisation was straightforward. Eurostat was responsible for those OECD Member Countries that were also Members of the European Union and the OECD was responsible for those OECD Member Countries that were not. This division of responsibilities has changed with time mainly as the result of the enlargement of the European Union1 (and, but to a lesser extent, of the OECD) and the sizeable jump in the number of countries participating in the Programme that occurred during the second half of the 1990s2. The evolution of Eurostat and OECD responsibilities is charted in Annex I which provides a brief history of the Programme.

Eurostat and OECD comparisons of GDP are made from the expenditure side. Each country participating in a comparison is required to provide either Eurostat or the OECD with a detailed breakdown of its expenditure estimate of GDP for the reference year. The detailed expenditures are used as weights in the calculation of PPPs and in the estimation of real expenditures. For this they need to be harmonised across participating countries that are required to report their final expenditures according to a common classification. This chapter explains the definitions and concepts underlying the common classification used for Eurostat and OECD comparisons. It describes the structure and presentation of the classification. The classification itself can be found in Annex III. The chapter also covers the reporting of expenditure data by countries and the subsequent validation of reported data by Eurostat or the OECD. It concludes by identifying those developments that will occur after publication of the manual and which will impact on the classification.

Individual consumption expenditure by households covers the actual and imputed final consumption expenditure incurred by households on the goods and services they require to satisfy their individual needs and wants. It accounts on average for over 60 per cent of GDP and over 85 per cent of actual individual consumption in EU Member States and OECD Member Countries. It is by far the most important of the seven main aggregates that constitute the Eurostat-OECD classification of GDP expenditures and, as such, it is central to the international price and volume comparisons organised by Eurostat and the OECD.

Final expenditure on housing is an important component of GDP. Its share of GDP for countries participating in Eurostat and OECD comparisons is usually between 8 and 12 per cent and tends to increase as GDP per capita rises. Although housing is a part of household consumption expenditure, it is not included in the cycle of price surveys for consumer goods and services described in Chapter 5. Instead it is covered by a special rent survey1 for which participating countries are not expected to collect data specifically but to extract them from existent statistical sources. The survey is organised by Eurostat and is held every year.2 Data are collected on the rents paid by tenants and on the rents imputed to owner-occupiers for a set of broadly-defined dwellings. Quantity and quality data on the housing stock are also collected.

Health expenditure accounts for a large share of GDP in EU Member States and OECD Member Countries. Both governments, as providers of health services, and households, as recipients of the services, are interested in knowing whether the differences in expenditure across countries reflect different amounts of health services being consumed or health services having different price levels. International price and volume comparisons of health expenditures provide countries with a means to assess their national health systems – at least in principle. In practice, inter-country comparisons of health expenditures are difficult to carry out because health services are comparison resistant with the institutional arrangements for their provision and payment varying from country to country.

Education is primarily a non-market service with the majority of pupils and students in participating countries receiving their education free or at prices that are not economically significant from non-market producers. Without economically-significant prices to value output, the expenditure on education provided by non-market producers cannot be derived as it is for market producers by summing their sales. To get round the problem, national accountants have adopted the convention of estimating expenditures on non-market services by summing their costs of production. Previously, to preserve consistency with the prices underlying the expenditure estimates, Eurostat and the OECD calculated PPPs for education and other non-market services with input prices: the input-price approach.

The services that general government provides to households comprise individual services and collective services. Individual services are the services that general government provides to specific identifiable households – that is, services, such as health and education, which are consumed by households individually. Collective services are those that general government provides simultaneously to all members of the community – that is, services, such as defence and public order and safety, which are consumed by households collectively. The principal individual services are covered elsewhere in the manual: health in Chapter 7 and education in Chapter 8. Collective services are covered in this chapter.

Gross fixed capital formation (GFCF) is one of the seven main aggregates in the Eurostat- OECD expenditure classification and accounts for around 20 per cent of GDP in most EU Member States and OECD Member Countries. It comprises three expenditure categories: machinery and equipment, construction and other products1. This chapter concerns the pricing of machinery and equipment and other products. The pricing of construction is dealt with in Chapter 11. Of the other products detailed in footnote 1, only computer software is priced; reference PPPs are used for the rest. Machinery and equipment and computer software are referred to as equipment goods in the manual and the price survey conducted to collect their prices is called the equipment goods price survey.

The chapter, like the previous chapter, concerns gross fixed capital formation (GFCF) and the pricing of capital goods for Eurostat and OECD comparisons. Chapter 10 deals with the equipment goods price survey and the collection of prices for machinery and equipment and computer software. This chapter deals with the construction price survey and the pricing of construction projects. As the chapter explains, pricing a construction project involves collecting unit prices with which to value its components and summing the values obtained to arrive at a total price for the project. PPPs for construction are calculated with the total prices for a set of construction projects. The set covers three types of structures: residential buildings, non-residential buildings and civil engineering works.

Previous chapters describe how price and expenditure data are collected and validated for the international price and volume comparisons of GDP that Eurostat and the OECD make with purchasing power parities (PPPs). This chapter explains how Eurostat and the OECD use the validated data to compute multilateral PPPs for GDP and its component expenditures. The computation has two stages. During the first stage, PPPs are calculated for basic headings. In the second stage, basic heading PPPs are aggregated with GDP expenditures as weights to obtain PPPs for each aggregation level up to and including GDP. Many methods have been developed to calculate and aggregate PPPs1, but the chapter considers only the methods currently employed for Eurostat and OECD comparisons. It should be read in conjunction with Annex V which provides a numerical example to demonstrate how basic heading PPPs are calculated and subsequently aggregated.

The chapter describes how Eurostat and the OECD present and disseminate the results of their comparisons. It also describes how Eurostat and the OECD update the disseminated results of the latest comparison before the results of the next comparison become available. In addition, the chapter explains the revision policy followed by Eurostat and the OECD with respect to results already disseminated for a comparison and subsequent revisions made by participating countries to their estimates of GDP for the year to which the comparison refers. This is an important consideration as such revisions can change the relativities originally established between countries. The chapter concludes by clarifying the access policy of the two organisations with regard to results and underlying basic data that are not in the public domain.

The origins of international price and volume comparisons of GDP can be traced back to the experimental comparisons carried out by the Organisation for European Economic Cooperation (OEEC) in the 1950s. Two approaches were adopted. The first was a comparison made from the expenditure side using mainly price data. Initially, it covered France, Germany, Italy, the United Kingdom and the United States.1 Subsequently, it was expanded to include Belgium, Denmark, the Netherlands and Norway.2 The second was a comparison between the United Kingdom and the United States made from the production side using mainly quantity data.