1. Tax revenue trends 1965-2022
Revenue Statistics 2023 presents detailed internationally comparable data on tax revenues of OECD countries for all levels of government. The latest edition provides final data on tax revenues for the period from 1965 up to 2021, the second year of the COVID-19 pandemic. In addition, provisional estimates of tax revenues in 2022 – a year marked by Russia’s invasion of Ukraine – are included for almost all OECD countries.1
In Revenue Statistics 2023, taxes are defined as compulsory, unrequited payments to the general government or to a supranational authority. Taxes are unrequited in the sense that benefits provided by government are not normally in proportion to their payments.
In the OECD classification, taxes are classified by the base of the tax:
Compulsory social security contributions paid to general government that are treated as taxes (heading 2000)
Greater detail on the tax concepts, the classification of taxes and the accrual basis of reporting is set out in the OECD Interpretative Guide in Annex A of Revenue Statistics 2023.
Tax ratios for 2022 (provisional data)
According to provisional data provided by OECD countries for this report, tax revenues as a percentage of GDP (i.e. the tax-to-GDP ratio) were 34.0% on average in 2022, a decrease of 0.15 percentage points (p.p.) of GDP relative to 2021. This decline in the OECD’s tax-to-GDP ratio was the first since 2019 and followed two consecutive years of increases against the backdrop of the COVID-19 pandemic in 2020 and 2021.
The tax-to-GDP ratio decreased in 21 of the 36 countries for which full 2022 data are available, increased in 14 and remained the same in one. On average, the increases were larger than the decreases (1.0 p.p. versus 0.9 p.p.). The largest change in 2022 occurred in Denmark, whose tax-to-GDP ratio declined by 5.5 p.p., having been the highest in the OECD in all but two years since 2002. The Netherlands, Poland, Sweden, Switzerland and Türkiye also recorded a decline in their tax-to-GDP ratio larger than 1 p.p.
Notes: Data for 2022 are preliminary. The OECD average in 2022 is calculated by applying the unweighted average percentage change for 2022 in the 36 countries providing data for that year to the overall average tax-to-GDP ratio in 2021.
The 2016 OECD average tax-to-GDP ratio includes the one-off revenues from stability contributions in Iceland. Excluding these revenues, the OECD average tax-to-GDP ratio in 2016 would have been 33.2%.
Source: Table 3.1.
Tax-to-GDP ratios varied considerably across OECD countries in 2022 (Table 1.1). Key observations include:
France had the highest tax-to-GDP ratio in 2022 (46.1%), a position it last held in 2018. Norway had the second-highest tax-to-GDP ratio (44.3%) while Mexico had the lowest tax-to-GDP ratio (16.9%).
Denmark observed the largest fall in the tax-to-GDP ratio between 2021 and 2022. Revenues fell by 5.5 p.p. due to declines in revenues from income taxes (4.5 p.p.) and from taxes on goods and services (0.9 p.p.).
The next-largest decline was observed in Türkiye (2.0 p.p.), where social security contributions fell by 1.6 p.p. and revenues from personal income tax (PIT) fell by 0.7 p.p.
Between 2021 and 2022, the largest increase in the tax-to-GDP ratio was in Korea, at 2.2 p.p. This was largely due to a 1.6 p.p. increase in revenues from corporate income tax (CIT) and a 0.6 p.p. increase in revenues from value added tax (VAT).
The second-largest increase was in Norway (1.9 p.p.), where an increase of 8.8 p.p. in revenues from CIT related to exceptional profits in the energy sector offset declines in revenues from PIT, social security contributions and taxes on goods and services. Chile and Greece were the other countries whose tax-to-GDP ratio increased by more than 1.5 p.p. in 2022 (Figure 1.2).
The OECD average tax-to-GDP ratio was higher in 2022 than in 2010, when it was 31.5% of GDP. The tax-to-GDP ratio increased over this period in 30 countries (including data for 2021 in the case of Australia and Japan) (Figure 1.2). The largest increases were seen in Korea (9.6 p.p.) and Greece (8.7 p.p.); increases of over 5 p.p. were also observed in Japan, the Slovak Republic, Spain and Portugal. In the remaining eight countries, the tax-to-GDP ratio decreased between 2010 and 2022. The largest fall occurred in Ireland, from 27.7% in 2010 to 20.9% in 2022, largely due to an exceptional increase in GDP in 2015. The next largest decrease occurred in Türkiye (3.9 p.p.).
Changes in the tax-to-GDP ratio are driven by the relative changes in nominal tax revenues and nominal GDP (Box 1.2). From one year to the next, if tax revenues rise by more than GDP (or fall by less than GDP) the tax-to-GDP ratio will increase. Conversely, if tax revenues rise by less than GDP, or fall further, the tax-to-GDP ratio will fall. Therefore, a higher tax-to-GDP ratio does not necessarily mean that the amount of tax revenues has increased in nominal, or even real, terms.
In 2022, nominal tax revenues increased from the previous year in 35 out of the 36 OECD countries for which data is available, while nominal GDP increased in all 36 countries. In 20 countries, the tax-to-GDP ratio declined because revenues rose by less than GDP, while in Denmark it declined because tax revenues fell in nominal terms while GDP increased (Figure 1.3). In the 14 countries whose tax-to-GDP ratio increased relative to 2021, nominal tax revenues increased by more than nominal GDP.
Changes between 2020 and 2021 are shown for Australia and Japan in Figure 1.3 because the tax-to-GDP ratio is not available in 2022. In both countries, the tax-to-GDP ratio rose by 1.1 p.p. between 2020 and 2021, with nominal tax revenues increasing by more than GDP in both cases.
The tax-to-GDP ratios shown in Revenue Statistics 2023 express aggregate tax revenues as a percentage of GDP. The value of this ratio depends on its denominator (GDP) as well as its numerator (tax revenue). The denominator – GDP – is subject to historical revision.
The numerator (tax revenues)
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, there is a lag of around one and a half years before finalised data is available. Final revenue data for 2021 were received during the period May-August 2023.
In 35 OECD countries, the reporting year coincides with the calendar year. Three countries – Australia, Japan and New Zealand – have different reporting years. Reporting year 2020 spans Q2/2020-Q1/2021 in Japan and Q3/2020-Q2/2021 in Australia and New Zealand (Q = quarter).
The denominator (GDP)
For the denominator, the GDP figures used for Revenue Statistics 2023 are the most recently available in October 2023. At that point, GDP figures for 2022 were available for all OECD countries.
Using these GDP figures ensures a maximum of consistency and international comparability for the tax-to-GDP ratios reported.
The GDP figures are based on the OECD Annual National Accounts (ANA – SNA) for the 35 OECD countries where the calendar year serves as the reporting year.
Where the reporting year differs from the calendar year, 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.
The average shown in this publication is an unweighted average of all countries for which data is available. The provisional average for 2022 is calculated by applying the unweighted average percentage change for 2022 in the 36 countries providing data for that year to the overall average tax-to-GDP ratio in 2021.
Note: In Türkiye, nominal tax revenues increased by 89% in 2022 while nominal GDP rose by 107%. Data for Australia and Japan show the change between 2020 and 2021, as preliminary data for 2022 was not available for these countries.
Source: Secretariat calculations based on Chapter 4 (tax revenues) and Table 3.19 (GDP).
Tax-to-GDP ratios for 2021 (final data)
The latest year for which tax-to-GDP ratios are based on final data and available for all OECD countries is 2021 (Figure 1.4). These data show that tax ratios varied considerably across countries:
In 2021, Denmark had the highest tax-to-GDP ratio (47.4%), followed by France (45.2%). Six other countries had tax-to-GDP ratios above 40% (Austria, Finland, Sweden, Belgium, Norway and Italy).
Mexico had the lowest ratio at 17.3%, followed by Colombia (19.2%), Ireland (20.7%), Chile (22.2%) and Türkiye (22.8%). Five other countries had ratios below 30% in 2021: Costa Rica, United States, Switzerland, Australia and Korea.
The tax-to-GDP ratio in the OECD area as a whole (unweighted average) was 34.2% in 2021. In 2020, it was 33.6%.
Relative to 2020, the tax-to-GDP ratio rose in 28 countries, fell in nine and stayed the same in one.
The largest increases in the tax-to-GDP ratio were in Norway (3.7 p.p.) and Chile (3.0 p.p.). Korea, Costa Rica and Israel all recorded increases in excess of 2.0 p.p.
The largest declines were in Hungary (2.2 p.p.) and Iceland (1.2 p.p.).
Between 2020 and 2021, the increase in the average tax-to-GDP ratio was driven by increases in revenues from CIT and VAT (of 0.5 p.p. and 0.3 p.p. respectively), which more than offset a decline in social security contributions of 0.2 p.p. The special features in the last two editions of this report analyse the impact of the COVID-19 pandemic on tax revenues in OECD countries (OECD, 2021[1]), (OECD, 2022[2]).
Tax ratio changes between 1965 and 2021
Between 1965 and 2021, the average tax-to-GDP ratio in the OECD area increased from 24.9% to 34.2%, an increase of 9.3 p.p. (Figure 1.1). Before the first oil shock (1973 to 1974), strong, almost uninterrupted income growth enabled tax levels to rise in all OECD countries. In part, tax levels rose automatically through the effect of fiscal drag on PIT schedules. From 1975 to 1985, the tax burden in the OECD area increased by 2.9 p.p. After the mid-1970s, the combination of slower real income growth and higher levels of unemployment apparently limited the revenue raising capacity of governments. But during and after the deep recession that followed the second oil shock (1980), countries in Europe saw tax levels rise further, to finance higher spending on social security and rein in budget deficits.
After the mid-1980s, most OECD countries substantially reduced the statutory rates of their personal and corporate income tax, although the negative revenue impact was often offset by reducing or abolishing tax reliefs. By 1999, the average OECD tax-to-GDP ratio had risen to 33.0%, the highest recorded level at that time. It fell back slightly between 2001 and 2004, but then rose again between 2005 and 2007 before falling back during the Global Financial Crisis in 2008 and 2009. The tax-to-GDP ratio increased in all but two years between 2010 and 2021,2 despite the impact of the COVID-19 pandemic in 2020-21.
For more detailed analysis of the long-term evolution of tax revenues in the OECD, please see Chapter 2, the Special Feature of this publication. This examines the buoyancy of tax revenues in OECD countries between 1980 and 2021 by estimating the relationship between changes in tax revenues (both in total and from individual tax types) and changes in GDP over the short- and long run.
The OECD average tax-to-GDP conceals great variety between countries. In 1965, tax-to-GDP ratios in OECD countries ranged from 10.6% in Türkiye to 33.7% in France. By 2021, the corresponding range was from 17.3% in Mexico to 47.4% in Denmark. The trend towards higher tax levels over this period reflects the need to finance a significant increase in public sector outlays in almost all OECD countries.
Tax structures are measured by the share of major taxes in total tax revenues. In 2021, the tax structures of OECD countries varied. Eighteen countries raised the largest part of their revenues from income taxes (both corporate and personal), ten countries raised the largest part of their revenues from social security contributions, and ten countries raised the largest part of their revenues from consumption taxes (including VAT). Taxes on property and payroll taxes played a smaller role in the revenue systems of OECD countries in 2021, both on average and within most countries (Figure 1.5).
Note: Countries are grouped and ranked by those where income tax revenues (personal and corporate) form the highest share of total tax revenues, followed by those where social security contributions, or taxes on goods and services, form the highest share.
Source: Secretariat calculations based on data in Chapter 4.
While the level of tax revenues has generally been rising on average in the OECD, the tax structure or tax ‘mix’ has been remarkably stable over time. Nevertheless, several trends have emerged up to 2021 – the latest year for which data is available for all 38 OECD countries. These trends are discussed below.
Taxes on income and profits
On average, in 2021, OECD countries collected 34.0% of their tax revenues through taxes on income and profits (personal and corporate income taxes taken together). Taxes on personal and corporate incomes remain the most important source of revenues used to finance public spending in 18 OECD countries; in ten of these – Australia, Canada, Denmark, Iceland, Ireland, Mexico, New Zealand, Norway, Switzerland and the United States – the share of income taxes in the tax mix exceeded 40% in 2021.
Note: The OECD average tax revenue in 2016 from main categories includes the one-off revenues from stability contributions in Iceland. This predominately affects the average revenues from property taxes, as a percentage of total tax revenues, in that year only.
Source: Secretariat calculations based on Tables 3.8 to 3.14.
Within taxes on income and profits, the share of PIT and CIT varies:
Revenues from PIT generated 23.7% of total taxes on average in 2021 compared with around 30% in the 1980s. About two percentage points of this reduction can be attributed to the impact on the average of a number of relatively recent entrants to the OECD from Eastern Europe and Latin America, for which tax revenue data is only available from the 1990s onwards. These countries tend to have relatively low PIT revenues and high revenues from social security contributions or CIT, but this impact is observed in the post-1990 data only.
The variation in the share of PIT between countries is considerable. In 2021, it ranged from 6.2% in Costa Rica to 42.9% in the United States and 52.8% in Denmark (Figure 1.5).
CIT revenues represented between 8% and 9% of total tax revenues, on average, throughout the period 1965 to 2003. They then increased to a high of 11.3% in 2007, before dropping to 9.0% in 2010 after the Global Financial Crisis. They remained between 9.0% and 10.0% of total tax revenues thereafter, except in 2018 and 2021, when they accounted for 10.1% and 10.2% respectively.
The share of CIT in total tax revenues in 2021 varied considerably across countries from less than 5% (Estonia, Greece, Hungary, Italy and Latvia) to over 20% in Mexico (20.2%), Australia (22.5%), Norway (23.6%) and Colombia (23.7%). Apart from the spread in statutory CIT rates, these differences are partly explained by institutional and country-specific factors, including:
the breadth of the CIT base; for example, some narrowing may occur as a consequence of generous depreciation schemes and tax incentives:
the degree of cyclicality of the corporate tax system, for which one of the important elements is loss offset provisions:
the extent of reliance upon tax revenues from the exploitation of oil and/or mineral deposits; or
other instruments to postpone the taxation of earned profits.
Social security contributions
Social security contributions accounted for 25.6% of total tax revenues on average across the OECD in 2021. They were highest in Czechia, Slovenia and the Slovak Republic (47.3%, 43.2% and 43.0%, respectively). In contrast, Australia and New Zealand do not levy social security contributions.
There was wide variation across OECD countries in the relative proportions of social security contributions paid by employees and employers in 2021 (Figure 1.7):
Nine countries (Chile, Greece, Hungary, Israel, Lithuania, Luxembourg, Poland, Slovenia and Switzerland) raised more revenues from employee social security contributions while the remainder raised more from employer social security contributions.
The highest share of employee social security contributions was in Lithuania, at 24.1% of total tax revenues. Employee social security contributions also amounted to over 15% of total revenues in Germany, Greece, Hungary, Japan, Poland and Slovenia. Denmark had the lowest share, at 0.1% of total revenues. Apart from Denmark, only Estonia and Ireland had revenues from employee social security contributions of less than 5% of total revenues.
The highest share of employer social security contributions in total tax revenues was in Estonia, at 31.1%. Employer social security contributions also exceeded 25% of total tax revenues in Czechia (29.1%), the Slovak Republic and Spain (both 25.8%). Denmark and Chile had the lowest shares, at 0.1% and 0.2% of total revenues respectively.
The highest share of self-employed or non-employed social security contributions was in Czechia (8.7%), followed by the Netherlands and Poland, at 7.5% and 7.3% of total revenues respectively.
Note: Australia, Costa Rica, Colombia, Iceland, Mexico and New Zealand are not included in Figure 1.7. Although Colombia, Costa Rica, Iceland and Mexico collect social security contributions, disaggregated data is not available. New Zealand and Australia do not levy social security contributions.
Source: Secretariat calculations based on data in Chapter 4.
Property taxes
Between 1965 and 2021, the share of taxes on property fell from 7.9% to 5.6% of total tax revenues on average across the OECD (Figure 1.6). In Australia, Canada, Israel, Korea, Luxembourg, the United Kingdom and the United States, property tax revenues amounted to more than 10% of total tax revenues in 2021. By contrast, property taxes accounted for less than 1% of total tax revenues in Czechia, Estonia and Lithuania.
Consumption taxes
The share of taxes on consumption (general consumption taxes plus specific consumption taxes) fell from 38.4% to 31.9% between 1965 and 2021 (Figure 1.6).
During this period, the composition of taxes on goods and services changed. A fast-growing revenue source has been general consumption taxes, especially VAT, which is imposed in 37 of 38 OECD countries.3
General consumption taxes accounted for 21.4% of total tax revenues in 2021, compared with only 13.4% in the mid-1970s. In 2021, the vast majority of this was from VAT (20.7% of total tax revenues).
The increased importance of VAT has counteracted the diminishing share of specific consumption taxes, such as excises and customs duties.
Between 1975 and 2021, the share of specific taxes on consumption (mostly on tobacco, alcoholic drinks and fuels, as well as some environmentally-related taxes) more than halved, from 17.7% to 8.5% of total revenues. In 2021, excises were the largest single category of total revenues under this heading, accounting for 6.3% of total revenues (Figure 1.8).
Rates of taxes on imported goods were considerably reduced across all OECD countries, reflecting a global trend to remove trade barriers.
Nevertheless, countries such as Greece, Hungary, Latvia, Mexico, Poland and Portugal (between 11%-12%) and Türkiye (17.6%) still collected a relatively large proportion of their tax revenues through taxes on specific goods and services in 2021.
Note: The unweighted average for each year includes all countries which report revenue in the categories shown in that year. The OECD averages for 2016 include the one-off revenues from stability contributions in Iceland.
Source: Secretariat calculations based on Chapter 4.
This section discusses the share of tax revenues attributed to the various sub-sectors of general government in 2021. The different sub-sectors are:
The guidelines for attributing revenues to different levels of government are based on the final version of the 2008 System of National Accounts. These guidelines are discussed in the special feature S.1 in the 2011 edition of Revenue Statistics.
Revenues of sub-national government in federal and unitary countries
Eight OECD countries have a federal structure. Among these countries, central government received 53.3% of total revenues on average in 2021. The second-highest share of revenues on average was received by social security funds, which are a sub-sector of general government, at 21.2% of total revenues, followed by 17.7% at the state level and 7.5% at the local level (Table 1.3).
Within countries with a federal structure, there was considerable variation around these averages:
In 2021, the share of central government receipts in the eight federal OECD countries varied from 28.3% in Germany to 80.4% in Australia and Mexico.
In 2021, the share of the states ranged from 2.0% in Austria and 4.1% in Mexico to 39.6% in Canada. The share of local government varied from 1.8% in Mexico to 14.5% in the United States and 15.3% in Switzerland.
Between 1975 and 2021, the share of central government revenues declined by over 12 p.p. in Belgium and by more than 5 p.p. in Canada.
The share of central government revenues increased in Austria by over 12 p.p. during this period. There was little change in Australia.
Of the seven federal countries with social security funds, five increased the share of revenue between 1975 and 2021. The exceptions were Canada and Mexico, where the share declined between 1975 (1980 for Mexico due to data availability) and 2021.
Colombia and Spain, which are classified as regional rather than unitary countries because of their highly decentralised political structure, have very different compositions by level of government. In Colombia, the share of central government receipts was 72.7% in 2021, with regional governments receiving 5.1% of total revenues and local governments receiving 12.3%. In Spain, the share of central government receipts in 2021 was 40.3% compared with 15.7% for regional governments and 8.6% for local governments.
The remaining twenty-eight OECD countries have a unitary structure. In these countries, an average of 63.6% of revenues were derived at the central level in 2021, with social security funds accounting for 25.2%. A further 10.8% of revenues were raised by local government.
The share of central government receipts varied from 30.8% in France to 93.8% in New Zealand in 2021.
The local government share ranged from 0.7% in Estonia to 35.3% in Sweden.
Between 1975 and 2021, there were increases in the local government share in excess of 5 p.p. in six countries: France, Iceland, Italy, Korea, Portugal and Sweden. Decreases of 5 p.p. or more in the other direction occurred in three countries: Ireland, Norway and the United Kingdom.4
Between 1975 and 2021, there were increases in the share of social security funds of 10 p.p. or more in three countries (France, Japan and Korea) and corresponding decreases in two countries (Italy and Norway).
Composition of central and sub-central government revenues
Figure 1.9 shows revenues from each major category of tax revenue for central and sub-central governments. For federal and regional countries, the sub-central level includes revenues received by both state and local governments. Figure 1.9 demonstrates that:
Central government revenues in almost all OECD countries are predominantly derived from taxes on income and on goods and services, with a negligible share from property taxes.
Property taxes provide a much larger share of revenues at the subnational level, and account for over 90% of revenues in four countries (Israel, Ireland, Greece and the United Kingdom).
By contrast, the share of income taxes and taxes on goods and services is lower at the sub-central level, the exceptions being Finland, Luxembourg and Sweden, where over 90% of sub-central revenues were derived from income taxes in 2021.
Note: The left-hand panel (a) refers to only those taxes that are classified as central government taxes. Social security contributions paid to social security funds are excluded. The right-hand panel (b) refers only to those taxes that are classified as sub-central taxes (local and [where relevant] state taxes). Social security contributions paid to social security funds are excluded.
Source: Secretariat calculations based on Tables 3.16 to 3.18.
Revenues paid to a supranational authority
The 22 member states of the European Union (EU) that are also members of the OECD collect taxes on behalf of the EU, as did the United Kingdom prior to 2020. These taxes primarily consist of customs duties and contributions to the Single Resolution Fund (SRF).5 Both taxes are collected on behalf of the EU by national tax administrations and are included in the total tax figures under headings 5123 and 5126 at the SUPRA level of government. In addition, they are shown as a memorandum item separately from the main figures since they represent a tax imposed by the EU and collected by national administrations.6
Table 1.5 shows the level of taxes collected on behalf of supranational governments in EU countries that are also OECD members, divided into countries in the Euro area and other EU member countries.
In 2021, the combined total of payments collected for the EU was highest in the Netherlands (0.5% of GDP), Belgium and Luxembourg (both 0.4% of GDP). All other EU countries that are also members of the OECD collected revenues on behalf of the EU equivalent to 0.2% of GDP or higher except Denmark, the Slovak Republic, Sweden and Hungary. In all countries except Finland, France and Luxembourg, customs duties were the primary source of these revenues.
OECD countries apply two kinds of tax credits to income taxes (both personal and corporate):
Non-payable or wastable tax credits are those that can only ever be used to reduce or eliminate a tax liability. They cannot be paid out to either taxpayers or non-taxpayers as a benefit. They are, therefore, the same as a tax allowance or relief.
Payable or non-wastable tax credits can be divided into two parts. One part is used to reduce or eliminate a tax liability in the same way as a wastable tax credit. The other part can be paid directly to recipients as a benefit payment when the value of the benefit exceeds the tax liability.
The OECD methodology for classifying non-wastable tax credits is set out in paragraphs 25 and 26 of the Interpretative Guide. These state that only the part of a non-wastable tax credit that is used to reduce or eliminate a taxpayer’s tax liability should be subtracted in the reporting of tax revenues. This is referred to as the ‘tax expenditure component’ of the credit. In contrast, the part of the tax credit that exceeds the taxpayer’s tax liability and is paid to that taxpayer is treated as an expenditure item and not subtracted in the reporting of tax revenues. This part is referred to as the ‘transfer component’.
Table 1.6 provides information on non-wastable tax credits in 2021 for those countries reporting them in Revenue Statistics 2023 (it may be that some countries with non-wastable tax credits do not report them and thus do not appear in the table). It shows the amount of the non-wastable tax credits and their two components together with the results of using the figures to calculate tax revenue values and the associated tax-to-GDP ratios. Table 1.6 also shows two alternative treatments for non-wastable tax credits:
The ‘net basis’, which 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 opposite, treating non-wastable tax credits entirely as expenditure provisions, with neither the transfer component nor the tax expenditure component deducted from tax revenues, as shown in columns 6 and 9. This is the approach followed by the GFSM and the SNA.
Table 1.6 shows that, with a few exceptions, the choice of method for reporting non-wastable tax credits has only a small impact on the ratio of total tax revenues to GDP. For countries with available data, the difference between the ratios on a net basis and on a gross basis only exceeds one percentage point for Germany and the United States, and is between half a percentage point and one percentage point for Canada, Czechia, France, Italy, New Zealand and the United Kingdom.
A memorandum item7 in Revenue Statistics 2023 describes the financing of social security-type benefits in OECD countries. Unlike social assistance benefits, which are funded from general government revenues, social security-type benefits are funded via contributions to social security funds or to private insurance schemes, or by other earmarked sources of funding. These sources of financing include:
Figure 1.10 shows the relative contribution of each of these sources to financing for social security-type benefits in OECD countries, based on data provided by countries for inclusion in the memorandum item in Revenue Statistics 2023.
Note: Two countries (Australia and New Zealand) provide social benefits via social assistance rather than via social security, so are not included in the table. In addition, Costa Rica and the Netherlands are also not included in the figure as complete data on contributions were not available in Revenue Statistics. The figures for Denmark should be interpreted with care as the level of social security-type benefits is very small compared to the level of social assistance benefits. Further, there may be borderline issues in some countries when distinguishing between quasi-compulsory and voluntary schemes.
Source: Secretariat calculations based on chapter 4
Taxes represent the largest source of earmarked financing for social security-type benefits, predominantly via social security contributions. Together, social security contributions and other earmarked taxes account for over 90% of the financing of social security-type benefits in 26 of the 34 OECD countries that provide this level of data (including 10 countries where they account for 100%). In the remaining eight OECD countries that provide this data, six countries report that compulsory contributions to the private sector play a significant role in financing social-security type benefits, including Chile (where they account for 82.8%), Colombia (67.3%) and Switzerland (53.2%). Voluntary contributions accounted for a significant share of funding in only a few countries, notably the United Kingdom (20.7%) and Denmark (77.7%).
Figure 1.11 shows tax-to-GDP ratios (as in Table 1.1 and Figure 1.4) both exclusive of earmarked funding for social security-type benefits (i.e. tax-to-GDP ratios less social security contributions and other earmarked taxes) and inclusive of all non-tax earmarked financing for social security-type benefits (i.e. tax-to-GDP ratios – including social security contributions and other earmarked taxes – plus compulsory contributions to the private sector and voluntary contributions to government).
The countries with the largest share of social security-type schemes financed by non-tax earmarked contributions are Switzerland (8.4% of GDP), Iceland and Chile (6.4% and 5.7% respectively), which materially affects their rankings:
Switzerland has a relatively low tax-to-GDP ratio among OECD countries, at 28.5%, but its combined ratio is above halfway in the OECD distribution.
Iceland has a tax-to-GDP ratio of 35.1%, in the top-third of OECD countries, and a combined ratio of 41.5%, which is the ninth-highest in the OECD.
Chile has the fourth-lowest tax-to-GDP ratio at 22.2% and the sixth-lowest combined ratio at 27.9%.
Excluding earmarked financing for social security benefits from the tax-to-GDP ratio does not affect Australia, Denmark and New Zealand, where benefits are funded out of general taxation. Figure 1.11 highlights that the largest share of earmarked funding for social security-type benefits is seen in France, at 25.0% of GDP, as indicated by the difference between the highest and lowest points on the figure. Austria, Belgium, Czechia, the Slovak Republic and Slovenia have the next highest shares, at between 15% and 18% of GDP.
References
[2] OECD (2022), Revenue Statistics 2022: The Impact of COVID-19 on OECD Tax Revenues, OECD Publishing, Paris, https://doi.org/10.1787/8a691b03-en.
[1] OECD (2021), Revenue Statistics 2021: The Initial Impact of COVID-19 on OECD Tax Revenues, OECD Publishing, Paris, https://doi.org/10.1787/6e87f932-en.
Notes
← 1. At the time Revenue Statistics 2023 was published, provisional data on tax revenues in 2022 for Australia was not available nor were provisional figures on social security contributions in Japan.
← 2. In 2016, Iceland received revenues from one-off stability contributions from entities that previously operated as commercial or savings banks and were concluding operations. The revenue from these contributions led to unusually high tax revenues for a single year and consequently, Iceland’s tax-to-GDP ratio rose from 35.1% in 2015 to 50.3% in 2016 before dropping to 37.1% in 2017. This led to an artificial high in the OECD average tax-to-GDP ratio in 2016 of 33.6%. Without these one-off revenues in Iceland, the OECD average tax-to-GDP ratio would have been 33.2%, an increase of 0.3 p.p. relative to 2015.
← 3. The terms “value-added tax” and “VAT” are used to refer to any national tax that embodies the basic features of a value-added tax by whatever name or acronym it is known e.g. “Goods and Services Tax” (“GST”).
← 4. For 1975, please see Table 1.4 of Revenue Statistics 2023.
← 5. The Single Resolution Fund (SRF) has been in place since 2015. Countries in the Eurozone are required to make SRF contributions under the Single Resolution Mechanism (Regulation (EU) No 806/2014). Contributions are paid on an ex-ante basis and contributions are transferred from the national authorities to the SRF. So far, contributions have been collected for the years 2015 to 2022.
← 6. In addition, EU civil servants pay income taxes and social security contributions directly to the EU. These revenues are not included in the data for total tax revenues in this publication as they are not paid to or collected by a national government. However, for the four countries with the highest number of EU civil servants (Belgium, Luxembourg, Italy and Germany), a memorandum account at the end of the respective country table in Chapter 5 provides information on the scale of these payments.
← 7. The financing of social security-type benefits is shown in Table 4.77 on a comparable basis (percentage of GDP) and in Table 5.39 on a national currency basis.