The 2026 edition of Corporate Tax Statistics contains a further year of anonymised and aggregated country-by-country reporting (CbCR) statistics covering fiscal year (FY) 2023.
Sixty jurisdictions out of a potential one hundred and nine submitted CbCR statistics to the OECD detailing the financial and business activities of over 9400 multinational enterprises (MNEs), with a further five jurisdictions reporting that they received zero CbCRs.
The data suggest modest increases in high-level indicators of base erosion and profit shifting in recent years, though indicators are still below earlier levels. While these indicators could reflect continuing BEPS behaviour, these data may be affected by turbulence in the global economy during high inflation periods during 2023. All these indicators remain far higher in investment hubs relative to other jurisdictions, pointing to the continued existence of BEPS activity.
From FY 2023, the data includes a disaggregation by MNE group size, as measured by unrelated party revenues, and by tax jurisdiction.
The share of revenues raised from large MNEs has increased in recent years. Large MNEs contributed an average of 44.5% of total corporate tax revenues in 2023 compared to 42.8% in 2017.
In FY 2022 and 2023, total profits returned to levels comparable to those recorded in 2019, prior to the COVID-19 pandemic. This supports the view that the substantial rise in total profits reported by the covered MNEs in FY 2021 was largely driven by post-pandemic recovery or by increases in inflation across many IF member jurisdictions.
The composition of business activity differs across jurisdiction groups. The most predominant activity in investment hubs is “holding shares” which also includes other equity instruments.
Corporate Tax Statistics 2026
Country-by-country reporting was implemented as part of Action 13 of the OECD/G20 BEPS Project to support jurisdictions in combating base erosion and profit shifting (BEPS). Under BEPS Action 13, all large multinational enterprises (MNEs) are required to share a country-by-country (CbC) report with tax administrations, including aggregate data on the global allocation of income, profit, taxes paid and economic activity for all tax jurisdictions in which it operates.
While the main purpose of CbCRs is to support tax administrations in the high-level detection and assessment of transfer pricing and other BEPS-related risks, data collected from CbCRs can also play a role in supporting the economic and statistical analysis of BEPS activity and of multinational enterprises in general. Under Action 11 of the BEPS Project (OECD, 2015[1]), acknowledging the need for additional sources of data on MNEs, jurisdictions agreed to regularly publish anonymised and aggregated CbCR statistics to support the ongoing economic and statistical analysis of MNE activities and BEPS. This section outlines progress on the implementation of Action 13, as well as the country-by-country reporting statistics published by the OECD under Action 11.
Jurisdictions have provided the OECD with anonymised and aggregated tabulations of the country-by country reporting information. Aggregation is performed according to certain sub-group or group characteristics (Box 7.1) and reported according to these different criteria in several tables. Table 7.1 provides an overview of the tables submitted to the OECD as part of CbCR statistics, a brief description of their content and the number of individual jurisdictions that submitted each table for FY 2023.
An MNE group is a collection of enterprises related through ownership or control such that the group is either required to prepare consolidated financial statements for financial reporting purposes under applicable accounting principles or would be so required if equity interests in any of the enterprises were traded on a public securities exchange.
An entity is any separate business unit of an MNE group that is included in the consolidated financial statements of the MNE group for financial reporting purposes.
The UPE directly or indirectly owns a sufficient interest in one or more other entities of the MNE group such that it is required to prepare consolidated Financial Statements.
A sub-group is formed by the combined entities of an MNE group operating in one tax jurisdiction.
The aggregated CbCR data are subject to a number of limitations that need to be borne in mind when carrying out any economic or statistical analysis (see Corporate Tax Statistics Explanatory Annex).
Comparisons of CbCR data over time should also be interpreted with caution, as a number of structural features of the dataset affect its consistency across years. In particular, the set of jurisdictions included in the dataset evolved over time as additional countries begin to submit CbCR statistics, while the population of MNE groups covered has also changed from year to year. Furthermore, differences in the level of disaggregation applied in compiling the data affects the comparability of reported aggregates across jurisdictions and periods. As a result, observed year-on-year changes may reflect, in part, variations in coverage and reporting practices rather than underlying economic developments.
Comparability is especially limited between the 2016 sample and subsequent years (2017–2022). This reflects the transition, in a number of jurisdictions, from voluntary filing of CbCR data in the initial year to mandatory filing in later years, which significantly altered both the scope and composition of the reporting population. In addition, differences in fiscal year coverage across reporting entities introduce further inconsistencies, as data for a given reference year may not correspond to uniform accounting periods. Taken together, these factors imply that caution is warranted when interpreting trends over time, particularly when drawing comparisons involving the 2016 data.
Nonetheless, the data provide important information on MNEs and their activities relative to other data sources:
The CbCR data provide global information on MNEs’ activities, with more granular information than is available in other data sources such as consolidated financial accounts.1
The CbCR data include information on number of CbCRs, number of sub-groups, number of entities, total unrelated and related party revenues (and their sum, total revenues), profit or loss before income tax, income tax paid (on a cash basis), current year income tax accrued, stated capital, accumulated earnings, number of employees, tangible assets other than cash and cash equivalents, and the main business activity (or activities) of each constituent entity.
The data ensure inclusion of all global activities of included MNEs.
At a minimum, the data allows for the domestic and foreign activities of MNEs to be separately identified.2 Depending on the reporting jurisdiction, it allows for an analysis of MNEs’ activities by tax jurisdiction due to a detailed geographical disaggregation. The CbCR data also provide cross-country information on MNEs’ business activities (e.g., manufacturing, intellectual property (IP) holding, sales) in different jurisdictions, allowing researchers to relate financial outcomes to these functions for the first time.
Note: The collection of Table 2, where the data is aggregated according to the MNEs size, has been introduced from FY 2022. The collection of Table 3, where the data is aggregated according to the MNEs sector has been postponed. The Inclusive Framework will consider whether to expand the dataset to include Table 3 in future years. The ETR of the MNE group and sub-group in Tables 4 and 5 should not be directly compared to the effective tax rates mentioned in the chapter on corporate effective tax rates.
While there are 146 members of the Inclusive Framework, 109 have implemented mandatory reporting for the FY 2023. Sixty jurisdictions submitted CbCR statistics to the OECD with a further five jurisdictions reporting that they received zero CbCRs in 2023. The 2026 edition of Corporate Tax Statistics includes CbCR statistics on CbCRs filed in 60 headquarter jurisdictions, covering over 9 400 MNE groups (see Table 7.2).
The number of jurisdictions providing aggregated and anonymised CbCR statistics has increased yearly since their introduction in 2016. Figure 7.1 shows that the total number of jurisdictions that could potentially provide CbCR statistics to the OECD increased from 58 in 2016 to 109 in 2023. This total is calculated as the number of jurisdictions that have implemented mandatory CbCR filing along with those that accepted voluntary filing in the specific year. For example, in 2016, 49 jurisdictions implemented mandatory filing while a further 9 accepted voluntary filing. The number of jurisdictions that provided CbCR statistics increased from 26 to 60 over the same period. Despite the large increase in the number of jurisdictions that could potentially submit CbCR statistics, the number of jurisdictions that did not provide CbCR statistics to the OECD has only increased from 32 to 44 with an additional five jurisdictions reporting that they have received zero CbCRs in 2023. Many jurisdictions receive too few CbCRs to be able to provide the statistics under their confidentiality standards.
Source: Anonymised and aggregated CbCR statistics and OECD Country-by-Country Reporting Requirements.
Figure 7.2 shows that the number of MNEs covered in the CbCR statistics has increased over time, from 3 628 in 2016 to 8 985 in 2023. Anonymised and aggregated CbCR data provide an overview of where large MNE groups are headquartered. Table 7.2 shows that, the number of reported MNEs varies considerably among jurisdictions, ranging from a minimum of two in Morocco to 2 098 in the United States. The median number of reported MNEs per jurisdiction is 68. 309 MNEs filed CbCRs as surrogate parent entities (where the jurisdiction of tax residence is different from the UPE’s jurisdiction of tax residence in cases where CbCR reporting rules may not be in place in the UPE’s jurisdiction of tax residence). Jurisdictions provided detailed statistics for 306 out of the 342 surrogate CbCRs that were filed.3
Source: Anonymised and aggregated CbCR statistics.
Note: Totals reflect the number of CbCRs for which data are reported by the UPE jurisdiction, excluding filings by Surrogate Parent Entities (SPEs) and reports containing only foreign data.
Note: Currency values (all values except the number of CbCRs and number of employees) are reported in millions of USD. Level of data disaggregation provided depends on data confidentiality standards applicable in each reporting jurisdiction. Average values have not been calculated for Czechia as the number of CbCRs has not been supplied for confidentiality reasons.
Source: 2023 Anonymised and Aggregated CbCR statistics.
Foreign and domestic MNEs account for significant shares of CIT revenues in several jurisdictions. For a selection of countries, Figure 7.3 reports total tax accrued based on CbCR statistics, as a fraction of the total national CIT revenues, taken from the OECD’s Global Revenue Statistics Database. The figure allows an examination of the relative importance of foreign and domestic MNE contributions as covered in the 2023 data.4
Note: The percentages above are calculated by dividing the amount of total tax accrued reported in CbCR statistics by total CIT revenues as reported in the OECD’s Global Revenue Statistics Database. The figure shows total revenues of both domestic and foreign MNEs as a percentage of total CIT revenues, with jurisdictions ranked according to the total contribution of MNEs to CIT revenues. As there might be some timing differences in recording tax payments between tax accrued reported in CbCR data and CIT revenues reported in Global Revenue Statistics, percentages should be considered as indicative. Revenues from foreign MNEs are calculated as the sum of tax accrued reported in the jurisdiction by MNEs headquartered in other jurisdictions. Foreign MNEs’ tax revenues should be considered as a lower bound as they can be reported exclusively where the geographical disaggregation is available at the jurisdiction level. Data for missing jurisdictions are not included because these jurisdictions are not covered in the 2023 OECD Global Revenue Statistics data. The US ratio of MNE tax revenues to total tax revenues is not presented in this chart due to a one-time transition tax imposed as part of the 2017 Tax Cuts and Jobs Act, which created a mismatch between the numerator and denominator of this ratio. MNEs generally report this transition tax as part of income taxes accrued and income taxes paid on the CbCR. However, the US Bureau of Economic Analysis does not classify this transition tax as CIT revenue (https://www.bea.gov/help/faq/1293). Therefore, the ratio of income tax accrued in CbCR data to US CIT revenues would be significantly upward biased and not indicative of the amount of CIT revenue contributed by MNEs in 2023. This mismatch is likely to persist for a number of years as taxpayers can elect to pay the tax over several years.
Source: 2023 Anonymised and Aggregated CbCR statistics and the OECD Global Revenue Statistics Database.
There is evidence of misalignment between the location where profits are reported and the location where economic activities occur. The data show continuing differences in the distribution across jurisdiction groups of employees, tangible assets, and profits.5 Figure 7.4 presents the distribution of MNEs’ foreign activities across jurisdiction groups.6 For example, high- and middle-income jurisdictions account for a higher share of total employees (respectively 36% and 44%) and total tangible assets (respectively 41% and 30%) than of profits (respectively 30% and 28%). On the other hand, in investment hubs, on average, MNEs report a relatively high share of profits (25%) compared to their share of employees (6%) and tangible assets (15%). High-income jurisdictions, middle-income jurisdictions, and investment hubs account for 44%, 29%, and 14% of tax accrued, respectively.7
Note: The profit variable could include intracompany dividends in several instances and therefore be upward biased. The bars represent jurisdiction groups’ shares of different variables (e.g., profit in group x/total profits booked in foreign jurisdictions) across all jurisdictions included in the CbCR sample. The percentages are calculated using Table 1A Panel A (all subgroups). “Other” reflects aggregate geographic groupings and Stateless entities.
Source: 2023 Anonymised and Aggregated CbCR statistics.
The composition of business activities also varies across jurisdiction groups, while the overall distribution provides a useful benchmark. Figure 7.5 presents the average share of main business activities across all jurisdictions, showing that sales, manufacturing, services, and provision of services account for the largest proportions of reported activities. Holding shares also represents a notable share, alongside manufacturing or production.
Source: 2023 Anonymised and Aggregated CbCR statistics. These data are based on the business activities data in Table 1A of the CbCR data.
Figure 7.6 illustrates how the distribution differs by jurisdiction group. In high-, middle- and low-income jurisdictions, sales, manufacturing and services remain among the most prevalent activities, although their relative importance varies. High- and middle- display very similar overall profiles while services play a more important role in low-income jurisdictions.
In contrast, investment hubs display a markedly different profile, with a substantially higher concentration of entities engaged in holding shares or other equity instruments. This activity represents by far the largest share within this group and is accompanied by relatively lower shares of activities such as manufacturing or sales.
Note: The ratios are calculated by dividing the number of the activities performed in a jurisdiction group by the total number of all activities performed in this jurisdiction group where data is available. For example, 19% of all activities performed in high-income jurisdictions are in the “sales” category. Entities could be attributed to one or more of the following activities: research and development; holding or managing IP; purchasing or procurement; manufacturing or production (manufacturing); sales, marketing or distribution (sales); administrative, management or support services; provision of services to unrelated parties (services); internal group finance; regulated financial services; insurance; holding shares or other equity instruments (holding shares); dormant; other activities. For the United States, other activities also include holding or managing IP; insurance; internal group finance; and research and development.
Source: 2023 Anonymised and Aggregated CbCR statistics.
This release of anonymised and aggregated CbCR data (FY 2023) provides some insights on BEPS. Revenues and profits per employee tend to be higher in investment hubs. Figure 7.7 and Figure 7.8 shows that the ratio of total revenues and profits to the number of employees is higher in investment hubs. In investment hubs, median revenues per employee are USD 1 811 000 while in high-, middle- and low-income jurisdictions median revenues per employee are USD 477 000, USD 211 000 and USD 153 000 respectively. While this may reflect differences in capital intensity or in worker productivity, it is likely also at least partially an indicator of BEPS.
Note: “Other” reflects aggregate geographic groupings and Stateless entities.
Source: Anonymised and Aggregated CbCR statistics.
Note: “Other” reflects aggregate geographic groupings and Stateless entities.
Source: Anonymised and Aggregated CbCR statistics.
On average, the share of related party revenues in total revenues is higher for MNEs in certain jurisdictions. Figure 7.9 plots the distribution of related party revenues as a share of total revenues, by income group. On average, the share of related party revenues in total revenues is higher in investment hubs than in high-, middle- and low-income jurisdictions. In investment hubs, related party revenues account for over 30% of total revenues, whereas the median share of related party revenues in high-, and middle-income jurisdictions is 20% and 14% respectively. The median share of related party revenues in low-income jurisdictions is much lower at just 7%. While high levels of related party revenues may be commercially motivated, they are also a high-level risk assessment factor and could be evidence of tax planning.
Note: “Other” reflects aggregate geographic groupings and Stateless entities.
Source: Anonymised and Aggregated CbCR statistics.
References
[1] OECD (2015), Measuring and Monitoring BEPS, Action 11 - 2015 Final Report, OECD/G20 Base Erosion and Profit Shifting Project, OECD Publishing, Paris, https://doi.org/10.1787/9789264241343-en.
← 1. In the case of the United States, CbCR data are less granular than Inland Revenue Service (IRS) Form 5471, 8865, and 8858 data.
← 2. With the exception of stateless income, which could relate to either domestic or foreign activities.
← 3. The total number of MNEs covered in the 2023 CbCR statistics is 9400. This includes all headquarter MNEs, MNEs that provide foreign information only and MNEs that have chosen surrogate filing.
← 4. Foreign MNEs’ contributions might be understated for two main reasons: first, some jurisdictions provided limited geographical disaggregation; second, the contributions of MNEs with parents headquartered in jurisdictions that did not provide data are missing.
← 5. Profits may be overestimated due to the inclusion of intra-company dividends as described in the CbCR disclaimer available in the Corporate Tax Statistics Explanatory Annex. To evaluate the potential magnitude of included dividends country specific analyses are available at: Netherlands: https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/corporate-taxation/netherlands-cbcr-country-specific-analysis.pdf; Ireland: https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/corporate-taxation/ireland-cbcr-country-specific-analysis.pdf; Italy: https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/corporate-taxation/italy-cbcr-country-specific-analysis.pdf; Sweden: https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/corporate-taxation/sweden-cbcr-country-specific-analysis.pdf; United Kingdom: https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/corporate-taxation/united-kingdom-cbcr-country-specific-analysis.pdf.
← 6. Jurisdiction groups (high-, middle- and low-income) are based on the World Bank classification resulting in 63 high-income jurisdictions, 101 middle-income jurisdictions, and 24 low-income jurisdictions. The 29 investment hubs are defined as those jurisdictions with a total inward Foreign Direct Investment (FDI) position above 150% of gross domestic product (GDP).
← 7. Tax accrued depends on both effective tax rates and taxable profits in a jurisdiction.