92000E0276

WRITTEN QUESTION E-0276/00

by Paulo Casaca (PSE) to the Commission

(7 February 2000)

Subject: Lack of credibility in Eurostat's PPP data

For at least the last 25 years, Germany's national statistics office in Wiesbaden has been publishing a series of data comparing prices in Portugal with those in Germany over the last few years, on a monthly basis (Preise, Fachserie 17, Reihe 10, Internationaler Vergleich der Preise für die Lebenshaltung, Statistiches Budesamt, various months).

The data published by the German national statistics office systematically reveal a level of prices for Portugal which is much higher than the level recorded by Eurostat. For example, Eurostat maintains that, in 1995, Portuguese consumer prices were 41,4 % lower than German prices (Purchasing Power Parities and Related Economic Indicators, results for 1995 and 1996, Eurostat), whereas Germany's national statistics office, in the above-mentioned publication for 1995, detected a difference of only 6,3 % for the same indicator.

Since 1998 the data released by Eurostat (which are implicit in GDP statistics, since the PPP results have not been published) continue to present Portuguese prices (GDP prices, since consumer prices have not been divulged) as being much lower than German prices, whilst the German national statistics office has concluded, in the case of consumer prices as well, that the escudo is overvalued.

Does the Commission believe that the use of an indicator which produces totally different (if not to say contradictory) values when published by different statistical bodies can be regarded as completely correct and as reflecting current practice?

Answer given by Mr Solbes Mira on behalf of the Commission

(28 March 2000)

The statistics of the Statistisches Bundesamt publication to which the Honourable Member refers are not produced using the same methodology or with the same purpose as those produced by the Commission to calculate purchasing power parities.

There are three main reasons why the results of the two publications are different. The Eurostat results include all consumption at the level of overall gross domestic product (GDP); in other words, not only the final consumption expenditure of households, but also the final consumption expenditure of government and gross fixed capital formation. The results presented in the German publication cover only the final consumption expenditure, but not even all of that, since rents, cars and insurance are excluded. In the case of rents, for example, which in 1995 represented 17,4 % of private final consumption expenditure in Germany compared with 4,7 % in Portugal, the price index for Portugal was 78 % lower than that of Germany. Similarly, for the final consumption expenditure of government, which in 1995 accounted for 12 % of GDP in Germany and as much as 18,1 % in Portugal, the Portuguese price index was 68 % lower than that of Germany.

In addition, products and services are weighted differently, since Eurostat takes the national structure of household consumption into account for each Member State (the German structure for Germany, the Portuguese for Portugal, and so on), whereas the Statistisches Bundesamt bases its analyses for all countries on Germany's structure of household consumption.

Finally, the two analyses serve different purposes. Eurostat defines the purchasing power parities in order to make a general comparison of all the European countries analysed (multilateral comparisons), whereas the German national statistics office produces results simply with a view to provide bilateral comparisons between Germany and the other countries. Unlike Eurostat, which produces and publishes its statistics on purchasing power parities for an international audience, the Statistisches Bundesamt produces its statistics for domestic purposes.

Because of their differences in methodology and purpose, the two publications do not, therefore, lend themselves to comparison; their results should be seen as complementing, not contradicting, each other.