Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized International Comparison Program [11.01] A New Approach to Recommendation for obtaining interim estimates for purchasing power CIS Statistical Committee 5 th Technical Advisory Group Meeting April 18-19, 2011 Washington DC
Table of Contents Construction of interim PPPs estimates... 3 Calculation of interim estimates of PPPs... 4 Limitation of the interim PPP estimates... 6 2
CIS Statistical Committee Recommendation for obtaining interim estimates for purchasing power parities Construction of interim PPPs estimates International comparison of Gross Domestic Products (GDP) based on purchasing power parities (PPP) is one of the main tools for analyzing and forecasting world economic development. The necessary statistical data are generated through the framework of the International Comparison Program (ICP). In accordance with the decision of the UN Statistical Commission, Global rounds of GDP comparisons based on PPPs are conducted. The World Bank is the coordinator of this work. Global comparison is conducted through the integration of the calculations for six regions using a common methodological and technological basis: the six regions are CIS, OECD-EU, Africa, Asia, Western Asia, and Latin America. The ICP holds year long surveys to collect primary price data using uniform schemes; data are verified and agreed step by step; purchasing power parities are calculated, and GDP volumes of the participating counties are measured. Due to the large scope of work, the full cycle of comparisons usually takes three years. Every such cycle is called a round of comparisons by the year of data collection. The Global round of 2005 involved 146 countries. Official results developed by the World Bank and approved by the UN Statistical Commission were officially published in 2008. The price survey of 2008 served as the basis for comparisons conducted by OECD-EU countries, where some CIS countries took part as well. The results of this comparison were published in the beginning of 2011. At present, in accordance with the decision of the UN Statistical Commission, a new round of Global Comparisons is taking place on the basis of price surveys for 2011. The results are planned for publication in 2014. Only the approved results of the comparisons rounds are considered official data on the purchasing power parities. Before the official data have been published, one may use interim PPP estimates for the purpose of comparison. The main approaches to calculate and use interim PPP estimates are described below. 3
Calculation of interim estimates of PPPs In order to obtain interim estimates of PPPS one can use different approaches. Each of these approaches derives indirect estimates of PPPs. Depending on the initial assumptions and the disaggregation level of calculations, various methods can yield quantitatively different results. Taking into account the indirect and temporary nature of the obtained estimates it is reasonable to use a rather simple method for obtaining interim estimates of PPPs. This saves efforts during calculations and, to some extent, ensures the higher reliability of the calculated results. One of the simplest methods is adjustment of PPPs using GDP deflator ratios. Let us use the following notations: 0 - the benchmark year when data were collected for the round of comparisons and official PPPs were calculated; t - the year for which the indirect estimate of PPP is calculated; PPP (А/В)0 - purchasing power parity of the currencies in countries A and B in the benchmark year 0: Quantity of units of national currency of country A for a unit of currency of country B (for example, if Russian ruble is used in comparisons as the base currency, than the formula given above for the PPP definition means quantity of units of the national currency of country A for one Russian ruble ) def A(t/0) - GDP deflator for country A measured in the national currency of country A in the year t in relation to the year 0; The purchasing power parity for the benchmark year, PPP (A/B)0, is known from the official data published on the basis of data collected through the price survey in the year 0. GDP deflators def A(t/0) and def В(t/0) may be obtained from the official statistical data of countries A and B, respectively. The indirect estimate of the PPP for the countries A and B in the year t may be obtained from the following formula. (1) PPP (А/В)t = PPP (А/В)0 def A(t/0) : def В(t/0) 4
As the GDP deflator is the broadest macroeconomic estimate of price dynamics, the given formula (1) allows for indirect measurement, in general terms, of the change in price ratios in the considered countries during the period from the year when PPP was calculated on the basis of the actual price survey. The indirect estimate of PPP for countries A and B derived from the given formula (1) may be used to calculate comparable GDP estimates for these countries in the year t. For example, the GDP of country A in the year t may be converted into a common currency using the following formula: (2) GDP (А/В)t = GDP (А/А)t : PPP (А/В)t where GDP (А/В)t - GDP of country A in the year t nominated in the national currency of country B using PPP of the year t. GDP (А/А)t - GDP of the country A in the year t, nominated in the national currency of country А in the year t Thus, if, for example, the Russian Ruble is used as the base currency, than expression (2) allows to estimate GDP of country A in Russian rubles at prices of the year t using PPP of the currency A and Russian Ruble in the year t. Converting GDP of two countries into a common currency in the year t allows direct comparison of GDP volumes in these countries. In the table below an example of the calculation of PPP adjustment using GDP deflators ratio is given; the Russian Ruble is used as the base currency. 5
PPP adjustment on the basis of GDP deflator ratio (conditional figures) year 0 year t GDP of the country A in current prices, billion units of national currency (u.n.c) 7 500 10 000 GDP deflator in country A, year t to year 0 1,19 PPP of the national currency of country A and Russian Ruble, u.n.c/rub. 4,50 4,66 (= 4,50 1,19 : 1,15) GDP of country A in current prices, billion Rub at PPP 1 667 2 147 GDP deflator for Russia, year t to year 0 1,15 Year t does not necessarily follow the base year 0; it may be any of the subsequent years till the publication of the next official data from the comparisons round. After official comparisons results of the next round have been published, one should use interpolation to obtain indirect PPP estimates for the period between the benchmark years. Limitation of the interim PPP estimates The presented method allows the calculation of indirect PPP estimates when results of direct statistical calculations are not available. However, this method, as with any other indirect method of PPP calculations, has its limitations which are to be taken into account when put into practice. One of the limitations of the method for PPPs adjustment on the basis of GDP deflator ratios is that it is used for obtaining indirect PPP estimates for each year. It provides the basis for spatial comparisons but cannot be used for compilation of time series in constant prices. A more important limitation of a general character is related to the fact that methods for indirect PPPs estimation do not fully account for the structural changes (both in prices and in GDP) that take place in the economy between the benchmark years even of two subsequent rounds of comparisons. In may result in deviations between the indirect estimates of PPPs and their direct calculations on the basis of large scale price surveys. This can be seen, if we compare Ruble/dollar PPP calculated in the framework of the OECD-Eurostat comparison for 2008, and its indirect estimate obtained by adjusting PPPs using GDP deflator ratios on the basis of 2005 data. 6
Such deviations can be explained from the theoretical point of view, and one should keep in mind that indirect PPP estimates and their application for GDP comparisons may be used only as a temporary solution untill the results based on direct statistical observation are received. Indirect estimates may differ from the subsequent official data and should be replaced as soon as the latter are published. 7