Purchasing power parity (PPP)
- Registration entry for subjects
- Reliability and security
- Access to information
Purchasing power parity (PPP)
0.2 Subject area
Prices and Consumption
0.3 Responsible authority; office, division, person etc.
Price statistics Department
Telephone: 528 1207
0.4 Purpose and history
Purchasing Power Parity (PPP) is used to convert the Gross Domestic Product (GDP) to make it comparable with that of other countries. It is also used to compare relative price level between countries.
PPP is a currency conversion rate used instead of exchange rates. PPP shows how much is needed of the currency of one country to purchase the same amount of goods and services in other countries.
PPPs have been calculated in OECD countries from 1980 while the United Nations have done so from 1970. From 1990 Iceland has taken part in an international price comparison programme where PPPs are calculated and used to compare GDPs.
0.5 Users and application
The main use of PPPs is in the comparison of volume indices between countries and in the comparison of relative price levels. The main users of PPPs are international organisations, such as Eurostat, the International Monetary Fund, the OECD, the United Nations and the World Bank.
However, the largest users are various departments within the European Union, which beyond comparing GDPs, use the PPP in:
Allocating structural funds to reduce disparities between and within EU Member states
Monitoring price developments in the Eurozone
Deciding the cost of living allowance for EU employees
Comparing relative price levels
Prices of goods and services for private consumption are gathered from stores and service providers. Sub-indices of consumer price are used in updating between years. Price of rent is collected from a house rental survey and from the continuous household expenditure survey. The rent supplements system, among other places, provides information used in updating price of rent.
Data on government collective consumption comes from the Ministry of Finance and Statistics Iceland department of wage statistics on the wages of public sector employees. This data is used to find the value of final government consumption.
Prices for the investment heading are found through surveys conducted every second year on the prices of buildings and of equipment goods.
Information on value and the itemization of the GDP gathered from the national accounts department of Statistics Iceland.
0.7 Legal basis for official statistics
There are no specifically Icelandic laws on the calculation or publication of Purchasing Power Parities. EU regulation is underway and it is thought that regulations on PPPs will become valid in 2007.
0.8 Response burden
Every year price surveys are carried out and it is expected that every three years they will cover all factors of the GDP. Anually new results from surveys that are available are included in the calculation, while the results of older surveys are updated using indices. The response burden is therefore not heavy in most cases, only extending to giving price information on a few items every three years. Surveys are to a large extent carried out through visits by interviewers, reducing the response burden to surveyed companies even further. Part of the information is gathered through telephone interviews and/or requests sent via fax or email.
The national accounts department of Statistics Iceland handles the national accounts and their itemization.
The heaviest response burden falls on the Ministry of Finance and the Wage department of Statistics Iceland in surveying the wages of employees in the public sector. Every year data on average annual wages and working hours is collected for the largest groups. The burden lies in processing the data in digital form.
0.9 EEA and EU obligations
Legislation on the calculation of PPPs is underway and it is assumed that it will be ready for acceptance by the administration by the end of 2006. Though there is no legislation specific to PPPs, they are mentioned in other legislation where it is set forth that they should be used. Those are acts no. 1164/94, 1260/99 and 1267/99 (on cohesion and structural funds) and no. 3830/91 on the wages of EU employees.
1.1 Description of content
The PPP is a method of calculation used instead of the exchange rate to compare the GDP in different countries. They are an average of price relatives between participating countries for a basket of similar goods and services, which is supposed to indicate the countries' GDP. The basket is divided into categories (see 1.2) and it should include goods which are reflective of the consumption patterns of each nation. Using price relatives, nominal expenditure values are converted to a comparable price level and thus compare the GDP volume between countries.
1.2 Statistical concepts
The classification frameworks used for comparisons of European Union countries and OECD countries are:
SNA (System of National Accounts, framework used by the United Nations), in accordance with this system national accounts are categorised according to the financing of expenses, e.g. state expenses due to the health service are categorised under final government consumption, only patients' usage under final consumption by private households. Appropriate when evaluating what families have paid for a good or service.
Another used framework is ICP (International Comparison Programme). It is based on consumption, thus state expenses owing to individuals and families are subtracted from final government consumption and added to final consumption by private households. More appropriate for comparison where expenses to health service and education are financed in different ways.
From 1999 the COICOP framework has been in use for consumption by private households. It is also used in the consumer price index, the harmonized consumer price index and in the continuous household expenditures survey carried out by Statistics Iceland.
COFOG (Classification of the Functions of Government) is used for final government consumption, categorised according to purpose and type. COPNI (Classification of the Purpose of Non-Profit Institutions Serving Households) is used to categorise the expenses of non-profit institutions.
The calculation of PPPs takes place in three stages.
Price relatives between all states are calculated for each sub-group of goods and services, e.g. cheese and milk.
Price relatives are calculated for groups of goods and services, e.g. dairy products, weighed for the expenditure classification of GDP for the relevant group.
Price relatives calculated using the EKS (Eletö-Köves-Schultz) method and a geometric mean is calculated for typical goods in the first country, (Laspeyres) geometric mean is calculated for typical goods in the second country, (Paasche) geometric mean calculated for results in both countries (Fischer).
In exceptional instances where a price is not available for all goods in the countries, a special method of calculation is used to estimate the missing values so that price relatives will always be available for comparable goods.
In the calculation of aggregate results the EKS-method is also used by Eurostat as well as OECD. The elementary level PPPs are weighed according to the division of the GDP into expenditure categories and added up and again the EKS method is used to calculate average ratios between all countries. This method ensures that different expenditure patterns do not affect the aggregate results. Most appropriate to comparisons of individual expenses and price relatives between countries.
GK (Geary-Kamis) method (used by the United Nations). Countries are considered as a group and the weight of each country is equal to its part of the total GDP of the group. Thus large states will weigh more than smaller ones. Average prices are calculated with iteration for the whole group. This leads to the patterns and values of expenses affecting the results in individual countries as well as the aggregate results. Changes to price relatives in one country can affect all other countries and aggregate results, e.g. a change to price relatives of clothes in France can affect price relatives for groceries in Britain.
Not appropriate if a country has an expenditure pattern which deviates from the average, in which case the results can be erroneous; this is called the "Gerschenkron effect". The main benefit of the method is that it is possible to get an aggregate result for all the countries, which allows comparison. Most appropriate for comparisons of the interior division of expenses within individual countries.
2.1 Reference periods
The reference period for results of the PPP comparison is one year. Data on each year is collected for a third of the prices used and 2/3 of prices in the household consumption groups are updated with indices.
2.2 Process time
Preliminary statistics, eg. for the year 2004 were published in December 2005. Final and itemized statistics for 2004 were published mid-2005.
Eurostat publishes a release plan according to process time (see 2.2)
2.4 Frequency of releases
General results are published annually. Additionally, results from certain price surveys are published separately when available.
3.1 Accuracy and reliability
Each participating country is responsible for the data that it provide to Eurostat but Eurostat is responsible for calculation of main results and releases of data.
3.2 Sources of errors
In surveys of prices of goods and services for private consumption, which are carried out every three years as well as in building cost and equipment good surveys various errors are possible:Errors in the preparation of lists of goods to be price checked, sampling errors: The population of the survey is goods/services under that heading of household consumption expenditures being surveyed. The countries often have insufficient market information or it can prove difficult to prepare lists of goods which are representative of the consumption patterns of each country. A country with consumption patterns which deviate from those of other states can have difficulties in finding typical goods which are available in other countries as it is necessary that at least one other country can give a price for the same good. It is endeavoured to keep the lists as simple as possible to ease work in translation and price collection. However, the quality of the result will be affected if the coverage is too narrow.Errors in price collection, measurement errors: Inaccuracy in product description and/or translation can cause price collection to become unreliable. If interviewers are not sufficiently trained, or stores / service providers are not chosen according to market share, errors can occur. Similarly, if it proves difficult to find the measured goods errors can occur.Errors during the checking of data, processing errors: If the price collection is insufficient for a good/service it can be difficult to evaluate the reliability of the data. Processing errors can occur when hand written information from interviewers is entered into a computer or when information from telephone interviews is entered.Measurement errors: Can occur when data is collected once per year, such as pricing of rent, salaries of public sector employees and in national account weights. The rent heading is problematic as the housing affairs in different states are arranged in very different ways making any comparison difficult. Errors in information on wages, which Statistics Iceland receives from the Ministry of Finance and Statistics Iceland wage department can occur because categorisation of occupations, according to ISCO-88 (ÍSTARF 95), is not perfect. Differences between countries in working hour definitions can also cause errors in the data. Errors in the weights of national accounts can occur due to different methods of calculation. Much work towards harmonisation has taken place; ESA-95, the system used in most European countries, has been taken into to use. When the countries have handed the data to Eurostat it is processed and aggregates are calculated for the whole group. Then data is looked through and it is particularly checked that states use comparable definitions and methodology in the gathering of data.
3.3 Measures on confidence limits/accuracy
Figures on errors/confidence limits are not used in publications, but in the processing of data prices which are 20% above or below the average of the EU-15 countries are especially verified.
4.1 Comparison between periods
Annual results for the PPP-comparison are not intended for comparisons between periods / over time as they are proportional figures between countries.
4.2 Comparison with other statistics
No comparison has been carried out between the results of the PPP and other statistics.
4.3 Coherence between preliminary and final statistics
Preliminary statistics for the PPP are only published for the GDP, not its sub-groups. If countries revise or correct national accounts, changes can appear between preliminary and final statistics.
5.1 Forms of dissemination
- Website of Statistics Iceland
- Results are published annually in a special issue of the Statistical Series.
- Eurostat publishes results in press releases and on their website, http://epp.eurostat.ec.europa.eu/
5.2 Basic data; storage and usability
Basic data used in the calculation of PPP is stored in digital form and partially in archives (data from price collection) by Statistics Iceland. Access to basic data is not allowed in accordance with rules that no information which could reveal the identity of its provider should be released.
An OECD publication is released every three years on the results of the PPP-comparison for all OECD countries as well as a few additional countries. In that publication there are chapters on the methodology of PPP.
A PPP manual issued by Eurostat and the OECD is available at: http://www.oecd.org/std/prices-ppp/eurostat-oecdmethodologicalmanualonpurchasingpowerparitiesppps.htm
5.4 Other information
New articles on the calculation of PPPs can be found on the OECD website: http://www.oecd.org/std/prices-ppp/
© Hagstofa �slands, �ann 24-3-2014