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 1965-2012.  In addition, provisional estimates of revenues in 2013 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

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)

OECD classification of taxes.

Classification d’impôts de l’OCDE

Interpretative guide

Much greater detail on the tax concept, the classification of taxes and the accrual basis of reporting is set out in the OECD Interpretative Guide.

The OECD Interpretative Guide.

Le Guide d’interprétation de l’OCDE.



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 2012 were received during the period May-August 2014.
  • In thirty OECD countries, the reporting year coincides with the calendar year. In four countries - Australia, Canada, Japan and New Zealand – have different reporting years. Reporting year 2012 includes Q2/2012-Q1/2013 (Canada, Japan) and Q3/2012-Q2/2013 (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 3 October 2014. By that time, the 2012 and 2013 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 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 is revised upward 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-one 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 29 countries. The exceptions are Chile, Japan, New Zealand, Norway and Turkey, which will follow over the next few months. The GDP figures for these twenty nine 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 levels of these changes are as follows:
    • Australia: increase of between 1.4% and 4.7% from 1965 to 2008.
    • Austria: increase of between 0.8% and 3.6% from 1970 to 2013.
    • Belgium: increase of between 1.6% and 3.3% from 1970 to 2013.
    • Canada: increase of between 1.4% and 2.7% from 1981 to 2012.
    • Czech Republic: increase of between 2.9% and 5.2% from 1990 to 2013.
    • Denmark: increase of between 1.7% and 3.1% from 1966 to 2013.
    • Estonia: variation of between -0.2% and 2.3% from 1993 to 2013.
    • Finland: increase of between 1.3% and 5.1% from 1970 to 2013.
    • France: increase of between 1.5% and 3.3% from 1965 to 2013.
    • Germany: increase of between 2.6% and 3.6% from 1970 to 2013.
    • Greece: variation of between -0.4% and 7.4% from 1965 to 2013.
    • Hungary: increase of between 1.1% and 2.6% from 1991 to 2013.
    • Iceland: increase of between 2.4% and 5.8% from 1970 to 2013.
    • Ireland: increase of between 1.7% and 6.5% from 1970 to 2013.
    • Israel: increase of between 4.0% and 6.9% from 1970 to 2012.
    • Italy: increase of between 3.4% and 3.9% from 1970 to 2013.
    • Korea: increase of between 5.3% and 8.3% from 1970 to 2013.
    • Luxembourg: variation of between -4.1% and 5.1% from 1970 to 2013.
    • Mexico: increase of between 0.9% and 2.2% from 1970 to 2012.
    • Netherlands: increase of between 5.3% and 7.7% from 1969 to 2013.
    • Poland: increase of between 0.1% and 1.9% from 1990 to 2013.
    • Portugal: increase of between 0.8% and 4.1% from 1970 to 2013.
    • Slovak Republic: increase of between 1.3% and 2.3% from 1990 to 2013.
    • Slovenia: increase of between 1.6% and 2.5% from 1990 to 2013.
    • Spain: increase of between 2.3% and 3.4% from 1970 to 2013.
    • Sweden: increase of between 3.7% and 5.9% from 1965 to 2013.
    • Switzerland: increase of between 5.0% and 6.1% from 1970 to 2013.
    • United Kingdom: increase of between 2.6% and 6.2% from 1970 to 2013.
    • United States: increase of between 3.4% and 5.0% from 1970 to 2012.
  • Since the tax figures reported in OECD Revenue Statistics have hardly been affected by these changes, the 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.

Treatment on non-wastable tax credits

Non-wastable tax credits are tax credits that can give rise to a payment to taxpayers when the credit exceeds their liability for that tax. They are sometimes referred to as ‘payable’ or ‘refundable’ tax credits.  The impact of the treatment of these non-wastable tax credits on the level of tax to GDP ratios is shown in Table F

Paragraphs 20 and 21 of the OECD Revenue Statistics Interpretative guide indicate that

  • only that portion of a non-wastable tax credit that is used to reduce or eliminate a taxpayer’s tax liability should be deducted in the reporting of tax revenues. For convenience, this may be referred to as the ‘tax expenditure component’ of the credit.
  • the part of the tax credit that exceeds the taxpayer’s tax liability and is paid to that taxpayer should be treated as an expenditure item and not be deducted in the reporting of tax revenues. This part may be referred to as the ‘transfer component’.
  • In Table F, the “split basis” as shown in columns 5 and 8 represents the treatment consistent with the Interpretative Guide and the OECD tax revenue figures.
Historically there have been significant practical difficulties in implementing these paragraphs of the Interpretative guide, resulting in some lack of uniformity of reporting. In addition, distinguishing between tax and expenditure provisions is conceptually difficult and there are valid arguments for alternative treatments. Consequently there is no ideal solution to the problem of how these tax credits should be treated. Two alternatives to the split basis are presented in Table F:
  • the “net basis” treats non-wastable tax credits entirely as tax provisions, so that the full value of the tax credit reduces reported tax revenues, as shown in columns 4 and 7.
  • the “gross basis” is the exact opposite, treating non-wastable tax credits entirely as expenditure provisions, with neither the transfer component nor the tax expenditure component being deducted from tax revenue, as shown in columns 6 and 9.
Table F reports the values of the non-wastable tax credits and their two components for the years 2000, 2005 and 2012, and shows the results of using them to calculate tax revenue values and their associated tax to GDP ratios on the three possible bases.  In making any comparison of tax to GDP ratios based on these alternative treatments of non-wastable tax credits, the reader should be aware of their potential drawbacks.
  • While the gross basis provides comparability between the treatment of public expenditure on in-work income related benefits and non-wastable tax credits, it does not provide comparability between wastable and non-wastable credits.  Changing a wastable tax credit into a non-wastable tax credit, even if it involves minimal fiscal cost or impact on taxpayers, could produce a large increase in reported revenue.  This is because amounts previously deducted from tax revenues would be treated as an expenditure provision and no longer be deducted.
  • The most serious drawback of the net basis is that it does not ensure comparability between countries with and without non-wastable tax credits. This is because it reduces tax revenues for countries with non-wastable tax credits by amounts that would be treated as expenditure in countries that use comparable expenditure programmes to deliver transfers to those who do not pay taxes.  Even between countries with non-wastable tax credits, reporting on a net basis would produce lower tax revenues (everything else being the same) for countries that are giving greater assistance to non-taxpayers with these credits. Arguably, this may give a misleading impression of the extent of the tax system.

However, with some exceptions, the choice of method for reporting non-wastable tax credits has only a small impact on the ratio of total tax revenue to GDP (Table F). For the countries with available data, the differences between the ratios on a net basis and on a gross basis are

  • One percentage point or more in only Germany, New Zealand, and the United Kingdom.
  • Between half of and one percentage point in Australia, Belgium, Canada, the Czech Republic and the United States.
  • Less than half a percentage point in Austria and Norway.


Downloadable tables/figures

Table E. Customs duties collected on behalf of the European Union

Table F. Effect of alternative treatment of non-wastable tax credits, 2000-12


How to obtain this publication

Readers can access the full version of Revenue Statistics 1965-2013, 2014 Edition as follows:


Back to the Revenue Statistics homepage


Related Documents