The Annual National Accounts database (ANA database) gives a complete set of annual national accounts. It is divided into four main subjects:
The database provides data for all OECD member states, key partner countries and groups of countries such as the European Union or the total of OECD member states. Most time series start in 1970, yet some series are also available from 1950 onwards. The data can generally be selected according to following dimensions (depending on the type of table): country, transaction, measure, period, sector and function. Data are expressed in millions (national currencies, US dollars, hours worked), thousands (persons, jobs), index, percentage growth rate per head, or unit (PPPs). The default power code and number of decimals can be changed by the user.
Data are collected through questionnaires covering the different tables. For the European OECD member states, this questionnaire corresponds to the Eurostat questionnaire, whereas the non-European member states receive an OECD-specific questionnaire. The questionnaires are sent to the OECD member countries once a year.
All OECD member states compile their accounts according to the 2008 System of National Accounts (2008 SNA), with the exception of Chile, Japan and Turkey, which are currently using the 1993 SNA and will adopt the new methodology in 2015 and 2016.
Data release dates vary across countries. EU countries usually publish the main aggregrates between February and April. More detailed breakdowns are reported in October. Fixed assets broken down by industry and product are released after 24 months. Non-European OECD countries loosely follow this schedule: Japan releases data in February, Turkey in April, Mexico and New Zealand in May, Chile and Korea in June, Canada and USA in September and Australia at the end of the year.
The data are revised on a regular basis, usually two or three times a year. In general, revisions depend on the national statistitics institutes’ changes or adjustments. Revisions mainly concern the latest three years and reflect readjustments of previous estimates. Besides the regular adjustment of data, exceptional revisions can occur, for instance due to changes in methodologies or sources. When this is the case, the OECD incorporates the new data into a blank version to clearly distinguish the new version from the old one. This may also lead to a reduced coverage of the time period until the country is able to submit historical data in accordance with the new methodology. Main aggregates, however, are linked to former data to provide users with longer time series in the meantime.
The metadata in the database provide country specific information on the sources and methodology and can be accessed by clicking the “i” next to the country name or the transaction (for more information, please see the OECD.Stat's user guide).
The measure covers the valuation of the data (values, volumes, indices etc.). In the annual national accounts, the code and the power code of the measure are indicated in the title (e.g.: C: National currency, current prices, millions).
Current and constant prices
At current prices, the flows of physical quantities are reported at the end of the fical or calendar year and evaluated by the prices reported in this particular year. Across time, GDP value can be affected by quantities and price changes. The drawback of this measure is that it can give the false impression that quantities have changed during the year, while in reality it was due to a change in price (or inflation).
The constant prices (or volume chain-linked) estimates on the other hand will give a better idea of the GDP volume (quantity) on the entire period. The 2008 SNA recommends chain-linked estimates and they are usually used to compare values over time. Chain-linked estimates are more accurate than constant prices because they use the previous year price structure (and thereby being closer to the prices structure of the studied year). The only problem with chain-linked estimates is that the additivity is lost and it can’t be conceptually used on transactions presenting positive and negative data (such as stocks data). Users may also convert data to current or constant PPPs to compare across countries or over time.
In the database, the volume measure depends on the country’s submission to the OECD. All OECD member countries publish chain-linked estimates with the exception of Mexico, which releases constant prices1. Each country has its own national reference year. It varies from one country to the other. By selecting the measure “V”, users display chain-linked data at the national reference year. To ascertain comparability, the measure “VOB” displays all data with the same OECD reference year, which is currently 2010.
List of measures:
|V||Constant prices, national base year|
|VP||Constant prices, previous year prices|
|VOB||Constant prices, OECD base year 2010|
|CXC||US $, current prices, current exchanges rates|
|VXCOB||US $, current prices, constant exchange rates, OECD base year 2010|
|VXVOB||US $, constant prices, constant exchange rates, OECD base year 2010|
|XVP||Previous year prices and previous year exchange rates|
|CPC||US $, current prices, current PPPs|
|VPCOB||US $, current prices, constant PPPs, OECD base year 2010|
|VPVOB||US $, constant prices, constant PPPs, OECD base year 2010|
|PVP||Previous year prices. previous year PPPs|
|HCXC||Per head, US $, current prices, current exchanges rates|
|HVXVOB||Per head, US $, constant prices, constant exchange rates, OECD base year 2010|
|HCPC||Per head, US $, current prices, current PPPs|
|HVPVOB||Per head, US $, constant prices, constant PPPs, OECD base year 2010|
|HCPIXOE||Per head, Index using current PPPs, OECD = 100|
|HVPIXOE||Per head, Index using constant prices, constant PPPs, OECD base year 2010, OECD = 100|
|VIX.OB||Volume index, 2010 = 100|
|DOB||Deflator, 2010 = 100|
|CD||National currency per US dollar|
1. The constant prices use the price structure of a reference year, so that it eliminates the inflation changes and allows the analyst to concentrate on the volume of production alone. This is why usually volume data are used for comparisons across the years.