Executive summary

On October 5, 2015, the Organization for Economic Co-operation and Development (OECD) released its final report on measuring and monitoring BEPS (Action 11) under its Action Plan on Base Erosion and Profit Shifting (BEPS). This report was released in a package that included final reports on all 15 BEPS Actions.

The document, Measuring and Monitoring BEPS (the Final Report), does not include any new proposals for changing international tax rules; rather, it focuses on measuring the size and extent of BEPS activities. Action 11 is intended to estimate the size of BEPS, to identify indicators of BEPS, and to provide recommendations for improving the measurement of BEPS. The report estimates that BEPS reduces global corporate income tax revenue by 4%-10% (i.e., US$100 billion to US$240 billion annually).

The six indicators of BEPS identified in the Final Report are: (1) the concentration of foreign direct investment (FDI) relative to gross domestic product (GDP); (2) the profit rates of multinational enterprise (MNE) affiliates in low-tax countries compared to those in high-tax countries; (3) the profit rates of MNE affiliates in low-tax countries compared with the profit rate of their own MNE groups; (4) the effective tax rates of MNEs compared to those of domestic-only enterprises; (5) the separation of intangible property from the location of its production; and (6) the concentration of debt in MNE affiliates located in higher-tax rate countries. The Final Report recommends greater cooperation between the OECD and taxing authorities in the collection and sharing of data. It also identifies several additional measures of BEPS that will become possible using the data collected under Actions 5 (harmful tax practices), 12 (disclosure of aggressive tax planning), and 13 (transfer pricing documentation and country-by-country reporting).

The Final Report is very similar to the discussion draft issued in April 2015.¹ The major differences from the draft are the addition of section providing a quantitative estimate of the scale of BEPS and a greater recognition of the difficulty of establishing a "counterfactual" on how the identified BEPS indicators would look in a world without BEPS activities.

Detailed discussion

Action 11 is different from other BEPS action items because it focuses on measuring BEPS activity rather than correcting it. The Final Report covers an assessment of existing data sources for BEPS analysis, indicators of BEPS, measuring the scale and impact of BEPS, and recommendations for improving data and monitoring of BEPS.

Assessment of data sources for BEPS analysis

The Final Report acknowledges that analysis of BEPS is especially challenging in terms of data because the strategies used to achieve what may be considered to be BEPS are often complex and may involve financial entities and transactions that are not publicly reported. By its nature, BEPS is not directly observable and must be estimated from available data. This is made more challenging by the need to separate BEPS from what the OECD calls "real economic effects." The Final Report indicates that some of the government investigations into BEPS activities by MNEs have highlighted the limitations of the public data by demonstrating the significance of data that is not public. Even the tax data directly controlled by governments is often not made public in a form useful for analysis. Limited government capacity for analyzing the data already collected by tax authorities was demonstrated by the fact that only 8 out of 37 countries surveyed by the OECD could report the total amount of tax revenue collected from MNEs in their jurisdictions.

BEPS indicators

While acknowledging the serious shortcomings of the available data, the Final Report proposes six indicators of BEPS activities. Collectively, these six indicators are intended to form a "dashboard" that can assist policymakers in monitoring the direction of change in BEPS activities over time.

The six indicators of BEPS identified in the report are: (1) the concentration of FDI relative to GDP; (2) the profit rates of MNE affiliates in low-tax countries compared to those in high-tax countries; (3) the profit rates of MNE affiliates in low-tax countries compared with the profit rate of their own MNE groups; (4) the effective tax rates of MNEs compared to those of domestic-only enterprises; (5) the separation of intangible property from the location of its production; and (6) the concentration of debt in MNE affiliates located in higher-tax rate countries.

Concentration of FDI relative to GDP

This indicator compares the average ratio of FDI to GDP for countries with the highest ratios to the FDI to GDP ratios for other countries. The rationale provided in the Final Report is that GDP is a measure of real economic activity taking place in a country and FDI is a measure of foreign investment within related groups of companies. High concentrations of FDI relative to GDP therefore could indicate foreign investment by MNEs that is not proportional to real economic activity taking place in that country. The OECD uses a group of 14 high FDI to GDP countries to form the numerator of the ratio while 200 lower FDI to GDP countries form the denominator. The Final Report says that this indicator more than doubled in value from 2005 to 2012, suggesting an increase in BEPS activity. Among the limitations of this indicator, however, is that it includes no measure of tax policy differences and does not control for non-BEPS reasons for increased FDI concentration.

Profit rates of MNE affiliates in low-tax countries compared to those in high-tax countries

This indicator compares the distribution of profits within each MNE group's affiliates between low-profit-rate affiliates and high-profit-rate affiliates, and between those with low effective tax rates (ETRs) and those with high ETRs, forming a two-by-two matrix. The rationale provided in the Final Report is that a concentration of an MNE's profits in its affiliates with the lowest ETRs could be a result of BEPS activity. The Final Report says that the lower ETR and higher profit quadrant accounts for 45% of total income and that this indicator has increased in value by 32% from 2007 to 2011. The most serious limitation of this indicator, however, is that random year-to-year variation in profit rates would also result in a larger proportion of income accruing to affiliates in the lower ETR and higher profit quadrant. This does not mean that this indicator could not capture BEPS behavior; without a good counterfactual estimate of what this indicator would show in the absence of BEPS, however, it is difficult to interpret.

Profit rates of MNE affiliates in low-tax countries compared with the profit rate of their global groups

This indicator compares the profit rate of MNE affiliates in low-tax jurisdictions with the MNE's worldwide profit rate. The rationale in the Final Report for this indicator is that profit shifting to low-tax jurisdictions should result in higher profit rates (measured as profits/assets) than the MNE's worldwide profit rate. The Final Report shows that the ratio of the profit rate of affiliates in lower-tax countries to MNEs' worldwide profit rate is 1.9 in 2007 and rises to 2.0 in 2011. Among the limitations of this indicator, however, is that it could only be calculated for 171 of the largest MNEs, making it useful as a measure for this group but of uncertain value as an indicator for the broader set of all MNEs.

Effective tax rates of MNEs compared to those of domestic-only enterprises

This indicator compares the ETRs of large MNE affiliates with domestic-only enterprises using a regression equation. The rationale in the Final Report for this indicator is that, if two entities, one an MNE affiliate and the other not, are otherwise similar and yet have different ETRs, this indicates that the MNE affiliate may be using hybrid-mismatch arrangements or other BEPs measures to reduce its tax expense. The Final Report says that the average large MNE affiliate had an ETR 3.3 percentage points lower than comparable domestic entities. This indicator has declined from a high of 4.5 percentage points in 2001 to a difference of 3.3 percentage points in the most recently available year, 2010. In addition to the limitations imposed by the limited availability of relevant data, however, there are many non-BEPS related differences between MNE and non-MNE affiliated entities that may affect their ETRs, such as their capital intensity, productivity, and use of available local preferential tax treatments.

Separation of intangible assets from the location of their production

This indicator compares the concentration of royalty receipts relative to research and development (R&D) spending. It is calculated by comparing the ratio of royalty payments to R&D spending in a group of high-ratio countries to the average ratio for the other countries in the sample. The rationale in the Final Report for this indicator is that moving intangible property from its place of origin to a lower-tax jurisdiction facilitates BEPS and that therefore mismatches between the location of R&D spending and royalties may indicate BEPS. The Final Report says that this indicator fell from 2.8 in 2005 to 2.7 in 2009 before increasing to 5.8 in 2011. Among the limitations of this indicator however, is that the sudden doubling of the indicator over two years is not fully explained. There were a total of 59 countries used in the analysis, of which only four were the high-ratio countries that make up the numerator of this indicator, making the indicator highly sensitive to any changes in this small set of countries. The high-ratio countries collectively accounted for only 3% of royalties for the 59 countries, making it unclear what the magnitude of this BEPS channel might be. In addition, royalties are also paid on intangible property such as trademarks and brands that are not generally the products of R&D spending.

Concentration of debt in MNE affiliates located in higher tax rate countries

This indicator compares the interest-to-income ratio of affiliates in higher statutory tax rate and lower statutory tax rate countries. The rationale in the Final Report for this indicator is that MNEs may strategically allocate debt to their affiliates in countries with tax rates higher than the average for their worldwide group in order to reduce their worldwide tax liability. The Final Report says that affiliates with above average interest-to-income ratios, located in countries with statutory tax rates above the average for the MNE's worldwide group, had interest-to-income ratios of 29%. This was 19 percentage points above average for all affiliates. The OECD therefore calculates this indicator to be 19 percentage points. This indicator, however, has one of the same limitations as the second indicator, namely that affiliates with above average interest-to-income ratios will, by definition, have higher income-to-interest ratios, so a finding that this is the case is not necessarily evidence of BEPS. The other limitation of this indicator is that all enterprises, both MNE and non-MNE, have an incentive to increase their use of debt financing as statutory tax rates increase, so a finding that affiliates in higher tax rate countries have higher interest-to-income ratios is also not necessarily evidence of BEPS. Taken together, these limitations make it difficult to know what the value of this indicator would be in the absence of BEPS.

This section of the Final Report concludes with a discussion of two additional indicators that may become possible with improved data availability. The first is profits compared to tax rates for the MNE's operations in its headquarters country and in its foreign affiliates. The second is differential rates of return for FDI made by special purpose entities (SPEs).

Toward measuring the scale and economic impact of BEPS and countermeasures

The measurement of BEPS will never be straightforward. Measuring BEPS will require sophisticated econometric techniques to separate BEPS activities from non-BEPS activities undertaken by MNEs. Shifting real economic activity from one country to another in response to differences in tax rates is not BEPS, but artificial arrangements that shift taxable income without shifting real economic activity are considered BEPS by the Final Report. The challenge for economists seeking to estimate the magnitude of BEPS is distinguishing between these two types of behavior while relying on financial and tax data. A further challenge is determining which economic activities generate profits and where those economic activities are taking place. Every analysis of BEPS has to make choices on these issues. The OECD's analysis in the Final Report uses company assets as described in financial accounting to determine the location of activities and to calculate profit rates.

The Final Report surveys the academic literature on BEPS and finds that it provides strong evidence for the existence of BEPS behaviors. Using a wide variety of data sources from multiple countries, researchers have found evidence on the magnitude of BEPS supporting a range of estimates from 5% to 30% of MNE profits. Another notable finding is that BEPS may have a much greater effect on the corporate income tax revenue of developing countries as on developed countries. The International Monetary Fund (IMF) estimated that BEPS reduces corporate income tax revenue globally by 5% and in developing countries by 13%. The United Nations Conference on Trade and Development similarly estimated that BEPS reduced corporate income tax revenue globally by 8% and in developing countries by 7.5-14%.

The OECD's own estimate of the global corporate income tax revenue loss from BEPS uses the same data sources as the OECD's six BEPS indicators introduced in the Final Report. The estimate is the sum of two separate econometric models that attempt to control for a variety of factors that may influence an MNE affiliate's profits so as to isolate the effect of BEPS behaviors. The first model is intended to measure the corporate income tax (CIT) revenues lost from profit shifting, and is similar to the third BEPS Indicator. The second model is intended to measure CIT revenue lost from MNEs' use of mismatches between tax systems and their relative use of domestic tax preferences. It is similar to the fourth BEPS Indicator. Both are based on financial accounting data and define profit as a return on assets. These estimates are combined to reach the OECD's estimate of the global CIT revenue lost to BEPS at 4%-10% (i.e., US$100 billion to US$240 billion annually).

The Final Report does not provide individual estimates of the CIT revenue that could be raised by each of the BEPS Action Items, but it reviews the existing literature on each one and, where the OECD has done its own analysis, provides its estimates as well.

Action 2 — Neutralizing the effects of hybrid mismatch arrangements: The OECD estimates that large MNEs have average effective tax rates 2.5 to 5 percentage points lower than similar domestic companies. This estimate comes from one of the two models that comprise the OECD's estimate of global CIT revenue lost to BEPS. By this estimate, Action 2 would address roughly half of all CIT revenues lost to BEPS.

Action 3 — Strengthening CFC rules: The Final Report indicates that research has found that controlled foreign company (CFC) rules can reduce passive investments in low-tax countries and reduce the activity of MNE affiliates in "tax haven" countries.

Action 4 — Limit base erosion via interest deductions: The Final Report reviews an extensive literature showing evidence of greater use of leverage by MNE affiliates in high-tax countries. Analyses of German interest limitation rules are described as reducing the use of debt financing. The OECD estimates that MNE affiliates raise their debt/equity ratios by 1.3% for every 1% increase in their statutory corporate tax rate above the average affiliate tax rate for that MNE group.

Action 6 — Prevent treaty abuse: The Final Report describes one recent piece of research that found treaty shopping reduces the withholding effective tax rate from nearly 8% to 3%.

Actions 8-10 — Assure that transfer pricing outcomes algin with value creation: The Final Report describes research showing that a 1% difference in tax rates produces a 2% increase in intra-firm import prices relative to non-intra-firm goods. Evidence on the migration of intellectual property to low-tax countries is described. The OECD's analysis in the Final Report finds that the tax sensitivity of profit shifting is almost twice as large for MNEs with patents as it is for those without patents.

This section of the Final Report concludes with a discussion of the tax incidence and real economic effects of BEPS countermeasures. The discussion is theoretical and concludes that, to the extent BEPS countermeasures succeed in raising the ETRs of MNEs, capital owners will largely pay the cost and real economic activity will decline somewhat in countries that recover the most tax revenue. The less competitive the markets in which MNEs compete, the more the tax incidence will fall solely on capital owners and the less real economic activity will be affected.

Toward better data and tools for monitoring BEPS in the future

The final section of the Report offers six recommendations to improve the data and analytical tools available for the understanding of BEPS and BEPS countermeasures.

1. The OECD should work with countries to publish new corporate tax statistics relevant to BEPS analysis on an internationally consistent basis. Among the data included should be aggregated and anonymized data collected under the country-by-country reporting provided in Action 13.

2. The OECD should work with governments to produce periodic reports on the estimated revenue impacts of proposed and enacted BEPS countermeasures.

3. The OECD should continue to produce BEPS Indicators to monitor BEPS and BEPS countermeasures.

4. Governments should improve the public reporting of business tax revenue statistics.

5. Governments should improve the non-tax data relevant for BEPS analysis on such topics as FDI, SPEs, and trade-in services and intangible property.

6. Governments should support BEPS research by academic and government economists.

Implications

The element of the Final Report that attracts the most attention is likely to be its quantitative estimate of the CIT revenue lost to BEPS. The range of 4%-10% of global CIT is fairly broad and encompasses earlier estimates by the International Monetary Fund and the United Nations. It is expected that this estimate will be quoted in discussions regarding BEPS.

If the OECD's methodologies for measuring BEPS become accepted, a more subtle implication of the analysis in the Final Report could be an implicit acceptance of assets as the key factor of production in assessing where profits are earned. Similar methodologies could be developed based on labor, sales, or some other factors. The OECD's choice to not discuss these other possibilities could be re-visited by governments or other stakeholders.

Another possible result of the Final Report is the use of the OECD's BEPS Indicators as tools to examine individual companies. Most of the indicators are calculated at the individual company level and then aggregated to form the global BEPS Indicators. Governments could attempt to use the same publicly available data used by the OECD, or the data provided by country-by-country reporting, to produce BEPS Indicators for each MNE with an affiliate in their country. Depending on what those indicators showed, particular MNEs could be subject to additional scrutiny.

Webcasts

EY is hosting a series of eight tax webcasts that will provide a comprehensive review of the final BEPS reports and the outlook for country action (with replays available after the live webcast):

— OECD BEPS Project Outcomes: Highlights and Next Steps — October 15, 10 a.m. EDT

— New Reporting under BEPS Action 13 — October 20, 10 a.m. EDT

— Digital Economy Developments and BEPS Action 1 — October 27, 12 noon EDT

— Permanent Establishment Developments and BEPS Action 7 — November 5, 10 a.m. EST

— Transfer Pricing and BEPS Actions 8-10 — November 12, 10 a.m. EST

— Anti-abuse Measures under BEPS Actions 3, 5, 6 and 12 — November 19, 10 a.m. EST

— Financial Payments and BEPS Actions 2 and 4 — December 3, 10 a.m. EST

— Dispute Resolution and BEPS Action 14 — December 10, 10 a.m. EST

For more information and to register for the webcast series, click here.

———————————————

———————————————
ENDNOTES

¹ See EY Global Tax Alert, OECD issues discussion draft on economic analysis of base erosion and profit shifting under BEPS Action 11, dated 22 April 2015.