Harmonized Cross Region and Cross Country CPI Time-Space Integration in the Euro-Zone

On 17th May 2006 the European Commission has expressed favourable opinion to the entering of Slovenia into Euro-zone. Afterwards, this opinion has been ratified by the Ministries of Finance of the 12 countries. Thus, since 1st January 2007, Slovenia has thus substituted the Taler with the Euro, by raising to 13 the number of European countries that adopt the so called ``unique currency’. Slovenia adds itself to the list of the ``twelve’ composed by: Austria, Belgium, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands and Spain.

Notwithstanding the question of elaborating regional PPPs did not receive yet the attention it deserves by researchers and scholars and, above all, by Euro-zone statistical information institutional producers, such as NSOs and Eurostat (and not even by the other world institutional statistical offices such as the United Nations Statistical Division, the OECD, the World Bank, the World Bank), some countries have started working in this direction. As an example, the UK Office for National Statistics (Wingfield, Fenwick, and Smith, 2005) made in 2000 the first exercise of estimating regional price level comparisons, then updated in 2003. In 2004 new estimates have been produced and disseminated. The basic approach to calculating regional price level differences was to measure the cost of purchasing a common basket of goods and services (the Retail Price Index (RPI) basket) in each region and express that cost relative to buying the same basket nationally. Despite this ``star’ approach, and although the estimates are conducted on a time-to-time basis (and not on a regular monthly one as envisaged in this paper), this work represents an important step forward the time-space integration.

The second example is provided by the Australian Bureau of Statistics in a paper by Waschka et alii (2003), with the development of experimental spatial price indexes to measure the price differences between the eight Australian capital cities. Here too, the TCPI basket was used, and the GEKS formula was applied, as it does not require prices of specific commodities to be available in every city, since the price comparisons can be made as long as prices are available in at least two cities.

The third example is given by Aten (2005) who used the CPD method to estimate differences in price levels across 38 geographic areas in the US, based on prices collected for 2003 CPI.

Later on (Aten, 2006), she proposed an experimental index of the price level differences in US selected areas for 2003 and 2004, by refining the analysis previously conducted by Koroski, Cardiff, and Moulton (1994), and Koroski, Moulton, and Zieschang (1999). In the Euro-zone, the Italian NSO, Istat, has recently started a programme aiming at elaborating a system of PPPs at a regional level (Biggeri, De Carli, and Laureti, 2008).

Before the Maastricht Treaty in 1992, these indexes had developed according to the perceived needs of each country, often with the objective of having an index which could be used to uprate the wages of the workers in order to preserve their purchasing power (a “compensation” index). The result was that a single comparison of national CPIs did not necessarily give an accurate comparison of relative rates of consumer price inflation (Astin, 1999).

Eurostat view was that the Member States should be allowed to decide on their own specific procedures, provided that comparability was not threatened (Astin, 1999).

The group of the 15 countries plus the 10 countries (Cyprus, Estonia, Latvia, Lithuania, Malta, Poland, Czech Republic, Slovakia, Slovenia and Hungary) that entered the EU starting from the 1st May 2004.

Since 1st January 2007 two additional countries, Bulgaria and Romania have entered the EU. These countries too, as well as Turkey, a candidate country, are included in the average HTICP. Therefore, the total number of countries involved in average HTICP elaboration accounts for 30.

While the updating is yearly, the survey is conducted every three years. In fact, Eurostat has the task of producing yearly PPPs (with the further obligation set up by the European Commission of publishing final results within 36 months from the reference year), whereas the NSO have given the task to collect and transmit to Eurostat the prices every three years. In order of not overcharging the prices collection work, the goods and services list has been subdivided in six parts and the ONSs conduct six differentiate surveys, two each year (one in the first semester and one in the second semester), over three years. As a consequence, the surveyed prices refer for one third to year t, for one third to year t–1 and for another one third to year t–2. This implies that the NSOs must provide the so-called “time rectification factors” with which to “centre” the prices collected in the three year period to a reference year t.

Actually, in the last decades all countries exhibited consumer price increasing, except Japan, the only big western economy which has experienced a period of consumer prices decreasing.

To this regard, the ICP Tool-Pack software implemented by the WB in TCPI 2003–2006 framework allows to import the information on prices and on products specifications collected through the surveys on national TCPIs that are believed useful also in PPPs implementation.

A discussion of the properties that should be possessed by a harmonized price index is given in Diewert (2002).

“Characteristicity” is a concept first introduced by Drechsler in 1973 and used by United Nations Statistical Office (UNSO) and Eurostat, though the term has acquired quite a different meaning with time, signifying that the item is typical of that country; in other words, that it is consumed in a large quantity.