OECD releases 26 countries’ CbCR data: First look for non-US countries
The OECD just released its second Corporate Tax Statistics publication, which includes County-by-Country Reports (CbCR) data tabulations for the first time. Similar to the first publication, the release can be viewed as half-full or half-empty, especially with the scope and limitations of the CbCR tabulations. [Couldn’t resist reusing image from 1/24/19 blog on the OECD’s first Corporate Tax Statistics publication.]
In 2015, I fought hard (and successfully) multiple times to retain eight words in the Base Erosion and Profit Shifting (BEPS) Action 13 report allowing the CbCR to be used “where appropriate for economic and statistical analysis.” Thus, it is gratifying to finally see the initial CbCR reports from 26 countries. The U.S. released its 2016 CbCR data in December 2018, but the OECD Corporate Tax Statistics data is the first time CbCR has been released publicly for at least 24 other countries. Those countries should be applauded for finally releasing publicly their data, although some of their data is in truncated form. Other countries, such as Germany and the United Kingdom, should start reporting their CbC data and help lead in this effort at improved tax transparency.
This new information will help international tax policy analysts understand a broader group of MNEs’ cross-border activities, and also to assist policymakers in the 2020 BEPS Inclusive Framework (IF) review of the current CbCR template. The release will also focus the debate between public disclosure of individual MNEs’ data vs. public aggregated and anonymized tabulations by tax administrations.
The OECD’s CbCR data tabulations are helpful, but also disappointing for several reasons:
· Although intra-group dividends were explicitly excluded from the revenue measure, their treatment was left unclear in the pre-tax profits/income measure defined in the 2015 BEPS Action 13 report. Countries and companies used different approaches resulting in varying degrees of double-counting, which understate tax-to-income and income-to-revenue ratios, as well as distort the geographic allocation of income. Clarification of this issue was made by the OECD in 2019, but it won’t affect the data until 2020 (which will then be confounded by the Covid-19 pandemic). In the meantime, effective tax rate metrics need to be viewed with caution, although many would argue that it may be the Best Available Data (BAD).
· Data for “Stateless entities” were requested to capture companies such as Apple Sales International that had tens of billions of income not reported to any tax authority in 2011-2013. However, stateless entities also include many “tax transparent” entities, such as partnerships which aren’t taxed at the entity level, but rather at the owners’ level. Thus, “stateless entity” data includes both BEPS avoidance along with income on which tax is paid but not at the entity level. The BEPS Inclusive Framework’s 2020 CbCR review is considering alternative meaningful reporting of stateless entities’ data, and that decision should be made quickly so it can apply to 2020 data.
· Details on specific BEPS planning channels is needed. As part of the political process of achieving mandatory CbC reporting, several key financial items were scrapped from the initial draft CbCR template in 2014. The 2020 IF review should include a public debate on the benefits and costs of including related-party royalty payments and related-party interest payments, key features of significant BEPS planning. MNEs are incurring non-significant costs in filing CbCRs, so the data should include important BEPS risk-relevant items in a standardized format for tax administration and tax policy analysis. A few additional key data items would have a high benefit-to-cost relationship.
· Confidentiality is cited by many countries not reporting much, if any, detail on the geographic location of their MNEs’ activities. Finland, Ireland, Korea and Netherlands lumped together all foreign jurisdictions, even though they had 1,216, 1,068, 2,127, and 3,530 foreign subgroups, respectively. Aggregating subgroup information when there are at least 10 or more subgroups in more disaggregated country displays defeats the purpose of country-by-country reporting.
The debate on public individual MNE reporting of CbC data will continue, and the now-apparent limitations of the aggregated and anonymized tabulations will give additional rationale for individual MNE reporting. Nonetheless, the OECD’s Corporate Tax Statistics publication is a significant, albeit incremental, step forward in improved tax transparency about large MNEs’ global tax activities. Issues, such as intra-group dividends, stateless income, and details on BEPS channels, would also be important for analysis of public individual MNE CbC reporting.
The OECD’s compilation of countries’ 2016 CbCR data has already improved tax policy analysis of potential anti-BEPS proposals, such as the Inclusive Framework’s digital tax proposals, by supplementing other existing cross-country data. I look forward to next year’s Corporate Tax Statistics with a second year of CbCR data, with expanded scope and refinements. In the meantime, hopefully academics will find new insights from the large amount of new CbCR data.
Tom Neubig