As the international tax community seeks revenue estimates from the OECD’s two-pillar tax reform project, some finance ministries are playing their cards close to the vest. In the United States, Republican lawmakers have repeatedly sought pillar 1 revenue information from the Treasury Department, only to be rebuffed each time. Why is Treasury reluctant to share this data? It says it would be premature to release estimates.
“To our knowledge, no country has published interim data of its estimates of Pillar One reallocation, or provided such estimates to its parliament before Pillar One negotiations are complete, presumably because doing so could undermine that country’s national interests and its negotiating position,” Treasury said in a March 2022 letter. “We . . . remain committed to provide comprehensive estimates around these issues when doing so would not undermine leverage.”
In Singapore, the government is working through the details of the OECD’s latest base erosion and profit-shifting project, and has noted that it will have to relinquish some taxing rights over profits under pillar 1 while receiving very little in return, because the country has a small domestic market.
In a BEPS Explainer, the Singapore government said that while it’s difficult to assess the full impact of both pillars without finalized rules and specific data points, it anticipates that Singapore may be able to collect higher tax revenue under pillar 2 if it can retain the profits of all of its domestic economic activities.
The United Kingdom is a notable exception to the tight-lipped trend. In the government’s 2022 autumn statement, it offered a five-year estimate of pillar 2 revenues, revealing that it expects to collect £335 million in the 2023-2024 tax year. By the 2027-2028 tax year, it expects to collect £2.25 billion. But the numbers are subject to change; the government noted uncertainties in the behavioral responses of groups, other jurisdictions, and future distribution of low-taxed profits and revenue.
The international tax community probably shouldn’t expect many official country-level revenue assessments in the coming months. On January 18, the OECD released new estimates on the economic impact of both pillars. The OECD’s release wasn’t a full, comprehensive economic impact assessment; that will be released in a few months. Nor was the assessment data broken down by specific jurisdictions. On that point, the OECD secretariat has deferred to inclusive framework members, who for various reasons are uncomfortable with the release of jurisdiction-specific estimates. That means the OECD secretariat has been fielding calls for jurisdiction-specific estimates, while not being authorized to release them.
In the interest of transparency, the OECD released some new data, because the design of both pillars has changed significantly since the OECD released a preliminary economic impact assessment in 2020. According to the OECD, the new figures were calculated on updated data, new pillar 1 design features highlighted in the amount A progress report, and the OECD’s global anti-base-erosion model rules.
The updated assessment is based on recent country-by-country data from 2017 and 2018 and projections for 2019-2021. But the data is limited as it mostly predates the COVID-19 crisis, recent global inflation, the ongoing implementation of BEPS measures, the U.S. Tax Cuts and Jobs Act, and behavioral reactions to those two regimes. The bottom line is that the OECD’s update is a work in progress, a rough sketch presented in a 46-page PowerPoint presentation. By comparison, the OECD’s 2020 economic impact assessment clocked in at nearly 300 pages.
In the meantime, the OECD is fully aware of the limitations mentioned above, and others are beginning to delve into those issues and more. The Tax Foundation recently released an analysis pointing out some key issues the OECD’s update missed. However, the OECD’s release may offer the greatest value for stakeholders if they use it as a springboard and a benchmark tool for their own subsequent analysis and estimates.
New OECD Analysis
Per the OECD’s newest estimates, pillar 1 reallocated profits and global revenue from pillar 2 could increase by roughly 60 percent and 50 percent, respectively, compared with its previous analysis. These are substantial increases subject to large caveats.
When the OECD created its 2020 economic assessment, it relied on anonymized and aggregated CbC data from 2016 to create jurisdiction-by-jurisdiction matrixes. The updated estimate relies on a broader batch of anonymized and aggregated CbC data from 2017 and 2018, projections, and an expanded use of Orbis data.
The updated estimate also relies on a new approach in which the OECD built individual matrixes for multinational enterprises expected to be subject to amount A. Those matrixes addressed their sales, profit, depreciation, and payroll on a CbC basis. These matrixes combine available unconsolidated and consolidated MNE data along with extrapolations based on industry data and jurisdiction-level matrixes, according to the OECD.
One of the OECD’s key challenges in calculating estimates for pillar 1 is access to comprehensive data. CbC data was available through 2018, which enabled the OECD to estimate MNE jurisdiction-level matrixes for 2017 and 2018. However, corporate data for 2019 through 2021 was consolidated, so the OECD in response took the 2018 geographic distribution of economic activity (sales, profit, payroll, and depreciation), projected it forward to 2021 and scaled it to match yearly consolidated account values for each MNE group.
Use of the expanded CbC data allows the updated assessment to cover 82 percent of profit as compared with 63 percent in the 2020 assessment.
From there, the secretariat took several steps to verify and validate the data it used. Among the data used were corporate annual reports to fill MNE-level matrixes, and CbC data confidentially provided by some MNEs to the OECD secretariat. The OECD also shared jurisdiction-group-level matrixes with some jurisdictions in which in-scope MNEs had already conducted their own analyses. Lastly, it asked some jurisdictions to validate shares of relief provided under the pillar 1 elimination of double taxation rules, as well as validate the impact of a de minimis threshold. None of the jurisdictions that the OECD worked with, however, shared MNE-level data, according to the secretariat.
Pillar 1
For pillar 1, the OECD now estimates that $200 billion in profits could be reallocated to market jurisdictions annually under amount A, based on 2021 figures. This is expected to lead to annual global tax revenue gains of between $13 billion and $36 billion, based on 2021 data. As a quick refresher, pillar 1’s amount A would reallocate 25 percent of an in-scope MNE’s residual profits — profits exceeding 10 percent of revenue — to jurisdictions in which the MNE has nexus. Initially, amount A would apply only to MNEs with at least €20 billion in global turnover and profitability above 10 percent. The OECD previously estimated in 2021 that $125 billion of MNE profits would be reallocated to market jurisdictions annually, so this is a 60 percent increase from its previous calculation. That said, 2021 was a highly profitable year for in-scope MNEs, and the OECD noted that the average reallocation between 2017 and 2021 would be $132 billion.
According to the updated assessment, between 82 and 108 MNEs would have fallen in scope of amount A between 2017 and 2021. That is consistent with the OECD’s previous calculations that amount A would capture about 100 companies.
One of the biggest issues in the pillar 1 negotiations was whether the OECD should reduce the amount A threshold so that it can apply to more companies, particularly midsized companies operating in developing and smaller countries that do not meet the €20 billion revenue threshold. In a nod to those countries, the OECD might lower the turnover threshold to €10 billion after the first seven years of amount A if an OECD review finds that amount A is working and being implemented as intended. Some had suggested the OECD could go even lower. The African Tax Administration Forum previously suggested a €250 million threshold. Previous 2021 analysis from Michael Devereux and Martin Simmler of the Oxford University Centre for Business Taxation suggested that if the OECD were to lower the turnover threshold to €750 million, the amount of reallocated profit could reach $240 billion.
The latest revenue estimates, if taken at their 2021 value, suggest that the OECD may not have to significantly lower the amount A revenue threshold to capture a larger amount of reallocated profit, although there are a few caveats. The first is that in-scope MNEs experienced high profitability in 2021 and the OECD notes that there is significant uncertainty about whether they will maintain those high levels. Second, the number of companies estimated to fall under amount A is still fairly small — by design — leaving the midsized company issue still open.
Nonetheless, middle- and low-income countries are expected to experience more revenue gains than high-income countries as a share of their existing corporate income tax revenues according to the OECD. This is a significant update, because the OECD’s previous impact assessment estimated that low- and high-income countries would experience revenue gains in roughly similar proportions. The IMF, in its April 2022 Fiscal Monitor, also found that low- and high-income countries could expect revenue gains in similar proportion when it analyzed the OECD’s previous data. In low-income countries, corporate tax revenue could rise by 0.7 percent and in advanced economies by 0.9 percent, according to the IMF.
The updated analysis doesn’t include a percentage breakdown of just how much revenue may rise across the three jurisdiction blocs, but hopefully this is something the expanded assessment will cover.
On the other hand, investment hubs (which the OECD defines as jurisdictions with a total inward foreign direct investment position above 150 percent of their GDPs) could lose even more revenue than the OECD initially forecast. According to the updated estimate, losses in investment hubs modestly increased, although the OECD does not mention the percentage of revenue these jurisdictions might see reallocated. Hopefully the organization will shed more light on this in its forthcoming impact assessment.
By comparison, the IMF’s 2022 analysis found that pillar 1 would reallocate 2 percent of global profits, largely shouldered by low-tax investment hubs, which would see about 2 percent of their corporate tax revenue reallocated to other countries.
The OECD also listed the key sectors it expects will be affected by pillar 1. They are:
electronics manufacturing;
chemicals and pharmaceutical manufacturing;
programming and information;
broadcasting and software;
telecom; and
food, beverage, and tobacco manufacturing companies.
This is an update from the 2020 economic impact assessment, which did not itemize the most affected sectors.
Pillar 2
The OECD estimates that pillar 2 would generate about $220 billion in global revenue gains based on 2018 data, as compared with its previous $150 billion estimate. This is a nearly 50 percent increase from the original figure and will account for 9 percent of global corporate income tax revenues, according to the OECD. We’ll have to wait for the OECD’s impact assessment for more details on what drove the increase, but on a high level, the OECD states that the figure is partly based on better, more recent data on global low-taxed profits.
Compare this with the IMF’s 2022 analysis, which agreed with the OECD’s $150 billion figure, and found that pillar 2 would initially increase global annual corporate income tax revenues by about 5.7 percent. That calculation included the pillar 2 formulaic substance carveout, which will initially exclude 8 percent of tangible assets and 10 percent of payroll from MNEs’ pillar 2 calculations.
At the time, the IMF noted that the benefits of reduced tax competition from pillar 2 could push revenues an additional 8.1 percent higher because of reduced competition.
Now that the OECD is saying pillar 2 will account for 9 percent of global corporate income tax revenues, a follow-up question is just how much higher these revenues may climb because of reduced competition.
Updated Low-Income Country Benefits
The OECD’s prediction that middle- and low-income countries will fare better is based on several design changes to both pillars.
First, under pillar 1 developing countries with less than €40 billion in annual GDP will be eligible for a new special purpose nexus rule that triggers amount A when in-scope multinationals have at least a €250,000 income nexus. This compares with a €1 million nexus for high-income jurisdictions.
Second, the OECD added a new tail-end revenue provision to pillar 1 that is designed to allocate extra amount A revenue to low-income jurisdictions. That provision, which was introduced in the OECD’s July 2022 Progress Report on Amount A, addresses the treatment of revenues from finished goods sold to a final customer through an independent distributor. The OECD has decided that unsourced revenues will be treated on a pro rata basis as arising in low-income jurisdictions, using a low-income jurisdiction allocation key up to an aggregate 5 percent limit.
Then, there are the de minimis rules for eliminating double taxation, which the OECD says will ensure that smaller jurisdictions are less likely to surrender their taxing rights.
As for pillar 2, the OECD said that its revised allocation key for the UTPR (formerly known as the undertaxed payments rule), which now includes employees, will also generate modest gains for low-income jurisdictions.
The OECD also mentioned that qualified domestic minimum top-up taxes (QDMTTs), which give source countries a first taxing right on income that is taxed below pillar 2’s 15 percent floor, will be a beneficial design feature for low-income countries. The OECD is still modeling the potential impact of QDMTTs, however, and did not include them in its most recent analysis. Any information the OECD can provide on QDMTTs in its upcoming impact assessment will be important.
Although the QDMTT is important for source countries, its impact on high-tax source countries (which includes many low-income countries) may be limited, given the 15 percent rate. As such, high-tax source countries may not find it helpful, although this will also depend on the nature of tax incentives and holidays they offer to in-scope MNEs. The OECD is exploring the issue and says it found increasing evidence that a high share of low-tax profit is located in high-tax jurisdictions. It will conduct additional modeling to better determine where low-tax profits are located.
Conclusion
Now that the OECD has sketched and updated economic impacts, it is up to the international tax community to take this information, along with data that the OECD will release in the coming months, and use it to conduct their own analyses. As the OECD’s update shows, there are many open questions and information gaps in the assessment process that can only benefit from a broad array of input and perspectives. That feedback is unlikely to be turned away; the OECD says it welcomes comments and feedback on its latest findings.