|
Article Excerpt
Abstract
The objective of this paper is to discuss the possibilities of attaining a time-space harmonization of the consumer price indexes that are elaborated by the 12 European Union (EU) Euro-zone countries and by Eurostat. After focusing on the duality of the time and space domains for price indexes elaboration, the basic elements of the methodology of estimation of consumer price indexes are delineated, both in time and space. Then, the harmonization of formulae and baskets is outlined, with emphasis on the latter. The current system of surveys for price collection is reviewed and its limits are underlined, in order to suggest a methodology for a consumer basket harmonized approach which ensures better harmonized indexes comparability, reduction of list of products, and unification of quality adjustment methods. (JEL E31)
Introduction and Background
In the unified treatment that the authors intend to pursue in this paper, Consumer Price Indexes (CPI) are defined as synthetic price indicators elaborated to measure the relative price changes over time or over space of a basket of consumer goods and services purchased by households. CPI are used for a wide variety of purposes:
1) In time domain:
(a) for inflation measurement;
(b) for the indexation of commercial contracts, wages, social protection benefits, or financial instruments;
(c) as a guide for monetary policy;
(d) as a tool for deflating the national accounts aggregates or calculating changes in national consumption or standards of living.
2) In space domain:
(a) for comparing the price levels and the standards of living in different countries or geographical areas;
(b) for international or inter-area comparisons of Gross Domestic Product (GDP) and/or its components.
In time domain, CPI are elaborated nearly by all countries all over the world. Hereinafter, these will be denoted by Time Consumer Price Indexes (TCPI). Since 1997, besides the specific national TCPI, each Member State of the European Union (EU) has been calculating the Harmonized Indexes of Consumer Prices (HICP), according to rules specified in a series of European Regulations developed by the European Union Statistical Office (Eurostat) in agreement with the EU Member States [Eurostat, 2001a]. (1)
HICP are used to compare inflation rates across the EU, as well as to monitor performance as regards the convergence criterion of price stability in the Maastricht Treaty framework. Since January 1999, the European Central Bank (ECB) has also used them for the measurement of price stability across the Euro-zone [Eurostat, 2004]. Moreover, since the creation of the European Monetary Union (EMU) among 12 EU countries in March 1998, Eurostat calculates the Monetary Union Index of Consumer Prices (MUICP), an aggregate index covering the countries within the Euro-zone, (2) and the European Index of Consumer Prices (EICP) for the Euro-zone, plus the other EU countries. (3) The latter indexes are calculated using statistics provided by the Member States on consumer price changes and the consumption patterns of households within their economic territories. The aggregation across countries uses country weights from household final monetary consumption expenditure.
While HICP provide the best statistical base to make EU comparisons of inflation and represent considerable progress in the harmonization of methodologies, it is still hard to imagine a complete harmonization of TCPI. In this respect, technical agreement on different aspects is still to be proposed. Among them, the treatment of quality adjustments, the source of the weighting structure, the choice of formula (fixed base versus chained), the homogenization in price collection, and the methodological treatment of specific lots. It should be underlined that HICP cover all area of household final consumption monetary expenditure, but the relative importance of consumers' expenditure on each good or service varies from country to country. Hence, there is no uniform basket applying to all Member States. The differences between HICP and the individual national TCPI concern the treatment of subsidies, healthcare and education, (4) and the treatment of owner-occupied housing. (5)
In space domain, Eurostat, in the framework of the European Comparison Programme (ECP), as a specific project of the International Comparison Programme (ICP), elaborates the so-called GDP Parities (GDPP), in order to undertake comparisons of GDP volumes among European countries [Eurostat, 2000]. The GDPP are now elaborated according to the Gini-Elteto-Koves-Szulc (GEKS) [Gini, 1931; Elteto et al., 1964] approach at any aggregation level. For this purpose, Eurostat proposes a unique list of consumer goods and services for which prices must be collected, for all the EU countries, other than the baskets used for the HICP. Within the GDPP, Eurostat elaborates and disseminates the so-called Purchasing Power Parities (PPP), typically space CPI.
Due to the fact that the time and space domains are regarded as disjoint fields and that HICP and the TCPI are calculated individually by each EU member country, whereas the PPP are calculated by Eurostat, the two sets of price indexes are elaborated independently and no attempt has been undertaken to achieve a harmonized treatment. As a consequence, not only the comparability among the TCPI and a full harmonization of the HICP is still missing, but above all, there is no harmonization between HICP and TCPI on the one hand, and PPP on the other hand, whereas there should be because of the dual nature of the question.
It is worth emphasizing that every National Statistical Institute (NSI) of the EU member state draws out two different baskets of goods and services concerning household expenditures for which to survey prices: (1) for the national TCPI and for the HICP, and (2) for the PPP (based on the list provided by Eurostat). Thus, harmonization should be pursued since, as it will be shown later, there is a duality in time and space consumer price indexes domains. This is what the authors intend to do in this paper, as regards to the 12 countries of the Euro-zone.
The harmonization process can be regarded both from the theoretical and methodological point of view and from the practical one of producing a uniform basket through the unification of the two baskets of goods and services into a single basket on which each NSI and Eurostat can conduct the elaborations. Therefore, the paper will deal with the first issue, although briefly, then focus on the practical aspects.
The paper is organized as follows. In the next section, the above duality will be discussed by giving a concise theoretical settlement to an already existing, even though implicit, harmonized way of looking at the problem, including a brief re-reading of the methodology both in time and space frameworks. The discussion will avoid as much as possible the liturgical way of presenting the topic, usually with wide illustration of elementary indexes aggregating formulae, various indexes discussion and comparison of their performance, properties and so on, and will be restrained instead to the very essential questions useful to the harmonization purposes.
In the third section, the authors will highlight the many common points and similarities between time and space consumer price indexes and elaborate a harmonization of time and space formulae and baskets. The conclusion of the paper will focus on the critical appraisal of the existing situation, as compared to the one proposed by the authors, and provide a summary of the suggestions emerging from the overall approach to the matter.
Methodology of Estimation of Consumer Price Indexes
Overview
The duality between time and space consumer price indexes domains as outlined in the introduction can easily be seen by reflecting on the fact that in both domains the aim of the index is to measure changes in levels of prices of a given basket of goods and services, and that the way of pursuing the objective is, mutatis mutandis, exactly the same (this is also the idea of Rao [2001], even though in a more general price indexes context). As anticipated, a discussion of the theoretical foundations will be avoided here, to pass instead, after a brief overview of the two contexts, to the illustration of the points salient to our purpose and traceable in EU NSI and Eurostat estimation activity, following our own re-reading.
While the nature of the questions underlying the two domains is similar, of course, as pointed out by Balk [2001], multilateral international comparisons are not simple translations of multilateral intertemporal comparisons. Indeed, some important differences between these two types of comparisons may hold, but do not in anyway affect the above stressed common nature: (1) time is a continuous variable, whereas the number of the countries involved in the comparison stays fixed; (2) unlike time periods, countries do not exhibit a natural ordering; (3) in a intertemporal comparison, the time periods considered are of the same size, whereas countries are by nature not equally important.
To make duality more evident, we comparatively analyze TCPI and PPP. TCPI can be defined as the number of money units needed to purchase at time t the same basket of consumer goods and services that was purchased at time with one money unit. In other words, they measure the ratio of price levels in time and t, relative to the same basket. Besides representing a tool for inflation measurement, TCPI play a prominent role in monitoring the effects of both economic and monetary government policies and provide an efficient and objective index for automatic adjustment of salaries, wages, rents, and all the other kind of indexation used in welfare state policy. Moreover, they are widely used as National Accounts (NA) time aggregates deflators, first of all, GDP.
PPP are defined as the number of money units needed to purchase in country or region or whatever administrative or conventional space aggregation A, the same basket of consumer goods and services that can be purchased in country or region or whatever administrative or conventional space aggregation B with one unit of money. In other words, they measure the ratio of price levels between space A and B relative to the same basket. PPP represent the tool for spatial inflation measurement and are crucial, too, as spatial deflators in space NA aggregates comparisons, namely for GDP international or inter-area comparisons.
What should...
|
