Tax policy analysis

Revenue Statistics - classification of taxes, Interpretative guide and methodology


Purpose of the publication

This annual Report presents detailed internationally comparable data on tax revenues of OECD countries for all levels of government.  The latest edition provides outturn data on tax revenues in 1965-2014.  In addition, provisional estimates of revenues in 2015 are included for most countries. 

For the purposes of the Report, the term ‘taxes’ is confined to compulsory, unrequited payments to general government.  Taxes are unrequited in the sense that benefits provided by government are not normally in proportion to their payments.  


The OECD classification of taxes an interpretative guide

In the OECD classification, taxes are classified by the base of the tax:

  • Income and profits (heading 1000)
  • Payroll and workforce (heading 2000)
  • Property (heading 4000)
  • Goods and services (heading 5000)
  • Other (heading 6000)
  • Compulsory social security contributions paid to general government are treated as taxes (heading 2000)

Much greater detail on the tax concept, the comparisions between the OECD and other international tax classifications, and the accrual basis of reporting is set out in the OECD Interpretative Guide.

The OECD classification and interpretative guide 2016 (PDF)

Classification des impôts de l’OCDE et Guide d’interprétation 2016 (PDF)



The impact of GDP revisions on reported tax levels

The tax ratios shown in this Report express aggregate tax revenues as a percentage of GDP. It is important to recognise that the value of this ratio depends on its denominator (GDP) as well as its numerator (tax revenue), and that the denominator is subject to historical revision.

The numerator (tax revenue)
  • For the numerator, the OECD Secretariat uses revenue figures that are submitted annually by correspondents from national Ministries of Finance, Tax Administrations or National Statistics Offices. Although provisional figures for most countries become available with a lag of about six months, finalised data become available with a lag of around one and a half years. Final revenue data for 2014 were received during the period May-August 2016.
  • In thirty-one OECD countries, the reporting year coincides with the calendar year. Four countries - Australia, Canada, Japan and New Zealand - have different reporting years. Reporting year 2013 includes Q2/2013-Q1/2014 (Canada, Japan) and Q3/2013-Q2/2014 (Australia, New Zealand) respectively (Q = quarter).
The denominator (GDP)
  • For the denominator, the GDP figures used for this Report are the most recently available on 1 September 2016. By that time, the 2014 and 2015 GDP figures were available for all OECD countries.
  • Using these GDP figures ensures a maximum of consistency and international comparability for the reported tax to GDP ratios.
  • The GDP figures are based on the OECD Annual National Accounts (ANA-SNA) for the thirty-one OECD countries where the reporting year is the actual calendar year.
  • Where the reporting year differs from the calendar year, the annual GDP estimates are obtained by aggregating quarterly GDP estimates provided by the OECD Statistics Directorate for those quarters corresponding to each country's fiscal (tax) year. For example, in the case of Canada Q2/2013-Q1/2014.
Revisions to the numerator and denominator

Both the numerator (tax revenues) and the denominator (the GDP figure) are subject to revisions, as more accurate estimates of the amounts involved become available. Such revisions will directly impact on published tax ratios.

  • If the tax figure rises and the GDP figure remains unchanged, the tax ratio will increase.
  • If the GDP figure is revised downward, the tax ratio will also go up, even though aggregate tax revenues have not increased.
  • Conversely, a higher GDP estimate implies a lower tax ratio, even if the amount collected in taxes has not changed.
  • Revenue data, especially for recent years, can be subject to infrequent and usually minor revisions. GDP figures are revised and updated more frequently, though not necessarily for all countries at the same time, reflecting better data sources and improved estimation procedures. Generally these revisions have a rather limited impact on tax ratios.
  • Occasionally, however, GDP figures may change in a more fundamental way when internationally agreed guidelines to measure the value of GDP are changed. The latest such change relates to the System of National Accounts 2008 (2008 SNA) which has now largely replaced its predecessor, the System of National Accounts 1993 (1993 SNA). An earlier version of the System of National Accounts was set up in 1968.
  • The twenty-two OECD countries that are member states of the EU have to adhere to the European System of Integrated Economic Accounts (ESA) for computing their GDP figures. The ESA is primarily an elaboration of SNA, though differing from it in several minor aspects which are not pertinent to this Report. Following the 2008 revision to the System of National Accounts, the 1995 ESA was replaced by the 2010 ESA.
  • So far, the 2008 SNA applies to the GDP figures presented in this publication for 32 countries. The exceptions are Chile, Japan, and Turkey which will follow over the next few months. The GDP figures for these thirty-two countries are, with the exception of odd figures in particular years, higher following the application of the new SNA because of both methodological changes and the impact of improved data sources.
  • The tax revenue figures reported in Revenue Statistics have hardly been affected by the changes to the definitions and standards in the SNA/ESA but in some countries there have been some adjustments as the opportunity has been taken to improve data sources and methodologies for compiling the national accounts figures (including tax revenues). However these changes have been generally relatively smaller than the changes to the GDP figures and tax ratios have generally fallen for those countries that have implemented the revised statistical framework.
  • One particular problem raised by the 2008 SNA/2010 ESA revisions is that countries have varied in the period for which they have revised their GDP figures. To limit this distortionary impact, the OECD Statistics Directorate/Centre for Tax Policy and Administration have estimated revised GDP estimates for 1965 and later years in those cases where OECD countries have not reported revised GDP figures.

 Back to the Revenue Statistics homepage


Related Documents