Macroeconomic impacts of limiting the tax deductibility of interest expenses of inbound companies

Transcription

1 Macroeconomic impacts of limiting the tax deductibility of interest expenses of inbound companies Prepared on behalf of the Organization for International Investment June 2015

2 (Page intentionally left blank)

3 Executive summary Inbound companies in the United States have long been an important contributor to the US economy. Inbound companies bring capital and create employment opportunities for US workers, support other US companies through the purchase of intermediate inputs, and engage in R&D activities in the United States. There is broad agreement that there is a need for US corporate income tax (CIT) reform, which can be expected to have significant impacts on decisions by global companies on where to invest. Section 163(j) of the Internal Revenue Code restricts the deduction for interest paid by a domestic corporation on related-party debt or third-party debt guaranteed by a related party. Recent tax reform proposals include provisions that would broaden Section 163(j) and further limit the deductibility of interest expenses. These provisions could have significant impacts on companies investment decisions. This report analyzes the impacts of potential changes to further limit the deductibility of interest expenses by foreign multinationals interested in investing in the United States ( inbound companies ). The report finds that, even when paired with a revenue-neutral reduction in the US CIT rate, such a change would raise the cost of capital, deter US investment, and reduce US GDP. Thus, even when combined with a lower CIT rate, this report finds that a further limit on the deductibility of interest expenses by inbound companies would make the United States a less attractive place to invest for inbound companies. The overall economic impacts of limiting the deductibility of interest expenses by inbound companies to a percentage of EBITDA (i.e., Earnings Before Interest, Taxes, Depreciation and Amortization) are estimated using a macroeconomic model of the US economy. The model is similar to one of the models used by the US Congress s non-partisan Joint Committee on Taxation (JCT) to analyze the macroeconomic effects of the Tax Reform Act of An important issue for a revenue-neutral tax reform is how a lower CIT rate or lower tax rates generally will be paid for. Understanding the potential impacts and tradeoffs associated with using the revenue from specific revenue-raising provisions is an important consideration in designing a pro-growth tax reform plan. This analysis pairs further limits on the deductibility of interest expenses of inbound companies with offsetting uses of the additional revenue a revenue-neutral reduction in the CIT rate or a revenue-neutral increase in government spending. This report finds that further limiting the deductibility of interest expenses by inbound companies would reduce US GDP in the long-run, after taking into account the effect of a revenue-neutral reduction in the CIT rate (Table ES-1). A 10% of EBITDA limitation on net interest expenses is estimated to reduce annual GDP by $11.1 billion, in the long-run, even though paired with a revenue-neutral reduction in the CIT rate (scaled to the size of the US economy in 2013). i

4 A 30% of EBITDA limitation on net interest expenses paired with a revenue-neutral reduction in the CIT rate is estimated to reduce annual GDP by $4.4 billion (scaled to the size of the US economy in 2013), in the long-run. Annual GDP is also estimated to decline if the increased revenue were instead used to finance higher government spending. Table ES-1: Estimated annual impacts of limitations on interest expense deduction, in the long-run Billions of US dollars change in GDP; percent change in GDP Effect on GDP Effect on GDP Policy Disallowed interest Reduce corporate income tax rate Increase government spending 10% limitation 41% -$ % -$ % 30% limitation 18% -$ % -$ % Note: Long-run estimates are scaled to the 2013 US economy. The 10% and 30% limitations are estimated to fund a 0.22 and 0.08 percentage-point reduction in the CIT rate, respectively. Source: EY analysis. The report also estimates the impacts of further limits on the deductibility of interest expenses by inbound companies on foreign direct investment (FDI) and employment without netting the impacts with a revenue-neutral change in the CIT rate. These estimates reflect the direct impacts of the limitation plus the indirect impacts on their suppliers and at the businesses where workers spend their income. Potential changes to further limit the deductibility of interest expenses by inbound companies are estimated to reduce FDI into the United States by between 2.4% ($66 billion) (under a 30% of EBITDA net interest limitation) and 7.1% ($196 billion) (under a 10% of EBITDA net interest limitation) and reduce the total employment contribution of inbound companies by between 515,000 (under a 30% of EBITDA net interest limitation) and 1.5 million (under a 10% of EBITDA net interest limitation). Table ES-2: Estimated long-run national FDI and employment impacts Percent change in FDI stock; number of full- and part-time employees (thousands) Policy FDI Direct employment Indirect employment Induced employment Total employment 10% limitation -7.1% ,505 30% limitation -2.4% Note: Employment impacts are scaled to the 2012 US economy. Figures may not sum due to rounding. Source: 2012 IMPLAN model of the US economy; EY analysis. Understanding the potential impacts and tradeoffs associated with using the revenue from specific provisions, such as further limitations of the deductibility of interest expenses by inbound companies, is an important consideration in designing a pro-growth tax reform plan. ii

5 Contents I. Introduction... 4 II. Investment by inbound companies in the US economy... 6 III. Limitations on the deductibility of interest expense of inbound companies... 7 Recent discussions on changes to interest deduction limitations... 8 Interest expense deduction limitations analyzed by this report and corresponding estimates of disallowed interest expense... 8 IV. Macroeconomic analysis of an interest expense limitation on inbound companies...10 Estimated macroeconomic impacts of an interest expense limitation on inbound companies for the US economy...11 Estimated change in FDI and employment contribution of inbound companies without a revenue-neutral CIT rate reduction...14 V. Limitations and caveats...16 VI. Summary...17 Appendix A. EY General Equilibrium Model of the US Economy, additional estimated impacts and sensitivity...18 Description of EY General Equilibrium Model of the US Economy...18 Estimated macroeconomic impacts of an interest expense limitation on inbound companies paired with a revenue-neutral increase in government spending...20 Sensitivity of macroeconomic impacts to key parameters...21 Appendix B. Impact of an interest limitation for inbound companies on investment incentives..24 METR and the cost of capital framework...24 Taxation and investment incentives in an international context...25 Endnotes...28 iii

6 Macroeconomic impacts of limiting the tax deductibility of interest expense of inbound companies I. Introduction Inbound companies in the United States have long been an important contributor to the US economy. Inbound companies bring capital and create employment opportunities for US workers, support other US companies through the purchase of intermediate inputs, and engage in R&D activities in the United States. There is broad agreement that there is a need for US corporate income tax (CIT) reform, which can be expected to have significant impacts on decisions by global companies on where to invest. The Section 163(j) of the Internal Revenue Code restricts the deduction for interest paid by a domestic corporation on related-party debt or third-party debt guaranteed by a related party. Recent tax reform proposals include provisions that would broaden Section 163(j) and further limit the deductibility of interest expenses. These provisions could have significant impacts on companies investment decisions. Proposals to limit the deductibility of interest expenses may also be at cross-purposes with some of the objectives of tax reform. Limiting the deductibility of interest expenses can raise the cost of capital even when paired with a revenue-neutral reduction in the CIT rate. 1 The higher cost of capital can be expected to deter investment. In the international context, this would translate into less foreign direct investment (FDI) into the United States, as well as changes in the extent to which investments are funded through debt or equity, with commensurate adverse impacts on output and employment. Rather than making the United States a more attractive place to invest, such a policy shift with respect to the deductibility of interest expenses could make the United States a less attractive place to invest. This report estimates the overall economic impacts of limiting the deductibility of interest expenses by foreign multinationals interested in investing in the United States ( inbound companies ) to a percentage of EBITDA (i.e., Earnings Before Interest, Taxes, Depreciation and Amortization) using a macroeconomic model of the US economy. The macroeconomic model used by this analysis is similar to one of the models used by the US Congress s non-partisan Joint Committee on Taxation (JCT) to analyze the macroeconomic effects of the Tax Reform Act of 2014 proposed by former Chairman of the House Ways and Means Committee Dave Camp. A key component of this analysis is estimating how a limitation that tightens the rules with respect to interest deductibility may influence the investment decisions of inbound companies interested in investing in the United States. An important issue in the tax reform debate is how a lower CIT rate or lower tax rates generally would be paid for. Understanding the potential impacts and tradeoffs associated with using the revenue from specific revenue-raising provisions is an important consideration in designing a pro-growth tax reform plan. This analysis pairs further limits on the deductibility of interest expenses of inbound companies with offsetting uses of the additional revenue a revenueneutral reduction in the CIT rate or a revenue-neutral increase in government spending. 4

7 A 10% of EBITDA limitation on net interest expenses is estimated to reduce annual GDP by $11.1 billion in the long-run, and a 30% of EBITDA limitation on net interest expenses is estimated to reduce annual GDP by $4.4 billion in the long-run, even though the revenue raised from these policies is used to fund a revenue-neutral reduction in the CIT tax rate. Annual GDP is also estimated to decline in the long-run if the increased revenue were instead used to finance higher government spending. The report also estimates the impact of further limits on the deductibility of interest expenses by inbound companies on foreign direct investment (FDI) and employment without netting the impact with a revenue-neutral change in the CIT rate. These estimates reflect the direct impact of the limitation, plus the indirect impact on their suppliers and at the businesses where workers spend their income. Interest limitation policies are estimated to reduce FDI into the United States by between 2.4% ($66 billion) (30% of EBITDA net interest limitation) and 7.1% ($196 billion) (10% of EBITDA net interest limitation), and reduce the total employment contribution of inbound companies by between 515,000 (30% of EBITDA net interest limitation) and 1.5 million (10% of EBITDA net interest limitation). 5

8 II. Investment by inbound companies in the US economy Inbound companies contributed 40% of manufacturing capital investment in US workers employed at inbound companies earn wages 33% higher than average. Inbound companies in the United States have long been an important contributor to the US economy. Inbound companies bring capital and create employment opportunities for US workers, support other US companies through the purchase of intermediate inputs, and engage in R&D activities in the United States. Inbound companies contributed $221 billion of capital investment in 2012; this amounts to approximately 11% of all private nonresidential fixed investment in the United States. 2 Inbound companies are especially important to the US manufacturing industry, which produces products for US and global consumers (Figure 1). Nearly 40% of total private nonresidential fixed investment in the manufacturing sector was attributable to inbound companies in The investments of inbound companies in the United States not only results in new jobs, but relatively high-paying jobs. Inbound companies directly employ 5.8 million people in the United States and pay wages that are 33% higher than the national average. 4 Most inbound companies provide on-the-job training to their employees. Although the type of training may vary depending on the industry, the types of operations, and local conditions, such training helps upgrade the skills of US workers. Inbound companies also support the growth of local small and medium-sized businesses. When inbound companies set up their operations in the United States, they rely on US suppliers for inputs and services. The establishment of local supply chain networks between inbound companies and their local suppliers helps many small and medium-sized US companies develop their businesses. Figure 1. Share of US capital investment by inbound companies, 2012 Manufacturing Wholesale trade Mining, quarrying, and oil and gas extraction Finance and insurance Utilities Real estate and rental and leasing Transportation and warehousing Information Retail trade Services Construction Agriculture, forestry, fishing, and hunting 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% Note: The shares sum to 100%. Source: US Bureau of Economic Analysis and EY analysis. 6

9 III. Limitations on the deductibility of interest expense of inbound companies A fixed ratio limitation on interest expenses deduction was included in the President s FY 2016 Budget and OECD BEPS discussion draft. Limits on the deductibility of interest expenses could disallow more than 40% of the interest expenses of multinational companies. The ability of business to deduct interest expenses is a common feature of the CIT system in most countries. Income tax principles dictate that interest expenses and other legitimate expenses incurred by businesses in the course of earning income should generally be deductible. Concerns may arise in the context of multinationals investing across boarders in countries with different tax rates. Host country governments may be concerned about the reduction in tax revenue resulting from inappropriate use of large interest deductions by foreign multinationals and their domestic subsidiaries that can have the effect of shifting taxable income into a different tax jurisdiction. Countries have adopted or considered interest expense deduction limitation rules applicable to locally incorporated subsidiaries of foreign multinationals to address these concerns, including: 1. A fixed ratio limitation on interest expenses or debt, such as an interest-to-earnings ratio, interest-to-asset ratio, or debt-to-equity ratio. 2. Limitation on interest expenses by comparing the subsidiary s debt level with the multinational group s overall debt position. 3. Limitation with reference to the arm s length principle, which compares the level of interest expenses or debt with that which would have occurred if borrowing was from an independent entity. 4. Restrictions on deductions for interest expenses arising in specific transactions. The United States currently has rules under Internal Revenue Code Section 163(j) that restrict the deduction for interest paid by a domestic corporation on related-party debt or third-party debt guaranteed by a related party. 5 The disallowed amount of interest expenses is capped by the excess of net interest expenses over 50% of the taxpayer s adjusted taxable income (ATI). There is also a debt-to-equity safe harbor of 1.5, below which the Section 163(j) restriction is not applicable. 7

10 Recent discussions on changes to interest deduction limitations The OECD released its discussion draft on Base Erosion and Profit Shifting (BEPS) Action 4: Interest Deductions and Other Financial Payments, on December 18, 2014 (the Discussion Draft ). The Discussion Draft discussed various approaches to limit interest deductions of multinationals. A key focus in the discussion draft is whether interest deductions should be limited with reference to a group-wide approach, a fixed ratio, or some combination thereof. 6 Compared with a group-wide approach, a fixed ratio rule may be easier for a member of a multinational group to apply as there is no need to collect additional information about the rest of the group. A fixed ratio rule can be applied with reference to assets or earnings. A fixed interestto-earnings ratio rule may be volatile to the extent earnings fluctuate. In the FY 2016 Administration s Budget (and FY 2015), rules were proposed to modify the Section 163(j) rules and broaden interest deduction limitations on all debt of inbound companies. In general, a US member belonging to a foreign multinational group would have two options to determine its interest expenses disallowed for deduction under the rules proposed in the FY 2016 Budget: Option 1: A group allocation approach that uses earnings as the allocation factor. The US member s interest expense deduction would be capped by a fraction of the third-party interest expenses of the multinational group to which it belongs: (US member earnings/multinational group earnings) * multinational group third-party interest Option 2: A fixed interest-to-earnings ratio approach that caps the US member s interest expense deduction by a specific proportion of its ATI calculated under the Section 163(j) rules. The rule would be applied on net interest such that the cap on the deduction would be 10% of ATI plus the US member s interest income. Interest expense deduction limitations analyzed by this report and corresponding estimates of disallowed interest expense This report analyzes 10% and 30% interest-to-ebitda ratio limitations based on net interest expenses (i.e., the amount of interest expenses allowed for deduction is capped by an inbound company s interest income plus 10% or 30% of its EBITDA). This analysis estimates that the percent of total interest expenses disallowed for deduction by inbound companies operating in the United States would rise from 4.5% under current law to 18.3% under a 30% interest limitation and to 41.4% under a 10% interest limitation (Table 1). The 10% and 30% interest limitations would be applied to the US subsidiaries of foreign headquartered multinationals, but there is little publicly available data on the EBITDA or interest expenses of individual subsidiaries. Tax return data available from the IRS allow for the calculation of net interest and EBITDA for inbound companies, but those data are aggregated at the industry level and, therefore, do not provide a clear view of interest expenses that would be disallowed under the 10% or 30% limitation. The aggregated data shows only whether the net 8

11 interest expense of an industry as a whole would fall above or below the limitation, not how the range of companies within each industry may be affected by the limitation. Table 1. Estimated percentage of interest expenses disallowed for deduction for inbound companies under proposed limitations % of total interest expense deduction disallowed Current law 4.5% Policies analyzed: 10% limitation 41.4% 30% limitation 18.3% Note: The current law estimate for the percentage of interest expenses disallowed for deduction for inbound companies is based on IRS data in 2010 from Form 8926, the last year for which such data are available. The estimates for the percentage of interest expenses disallowed for deduction for inbound companies under each of the policy scenarios are based on a statistical match between the IRS SOI data from 2002 to 2011 on foreign controlled domestic corporations and financial statement data for publicly traded global companies headquartered in foreign countries with US subsidiaries from 2002 through Source: IRS, S&P Capital IQ, and EY analysis. To address the short-comings of the IRS data, this analysis combines the publicly available IRS tax return data for inbound companies with financial statement data from the S&P Capital IQ database for 3,731 individual foreign headquartered global companies with US subsidiaries. These two data sets are combined using a statistical matching routine and then estimates of the disallowed interest for inbound companies are made on the basis of the combined data set. The financial statement data from the S&P Capital IQ database is at the company level, but for the global operations of the company, not the inbound company operating in the United States. The statistical routine used to combine the two data sources, in effect, takes the range of interest-to-ebitda ratios found in the company level financial data for each industry, and applies it to the industry-level IRS tax return data for inbound companies. 7 9

12 IV. Macroeconomic analysis of an interest expense limitation on inbound companies A 10% net interest to EBITDA ratio limitation for inbound companies paired with a revenue-neutral reduction in the CIT rate would reduce annual GDP by $11.1 billion (scaled to the US economy in 2013), in the long-run. A 30% limitation would also result, in the long-run, in less GDP a $4.4 billion reduction in annual GDP when combined with a revenue-neutral reduction in the CIT rate (scaled to the US economy in 2013). US GDP would also decline if the increased tax revenue from either interest expense limitation were used to increase government spending. Interest deduction limitation policies would, in the long-run, reduce the stock of FDI in the United States by between 2.4% ($66 billion) (30% limitation) and 7.1% ($196 billion) (10% limitation) and the total employment contribution of inbound companies by between 515,000 (30% limitation) and 1.5 million (10% limitation) (scaled to the US economy in 2012). The proposed interest expense limitations analyzed by this report would affect businesses through changes in their cost of capital and the corresponding impacts on investment. Further, investment shifts between industries and sectors of the US economy, as well as between the United States and the rest of the world. The share of disallowed interest is used to estimate the increase in the cost of capital under each of the interest limitations. The macroeconomic impacts of implementing a new limitation on the deductibility of interest expenses on inbound companies are estimated using the EY General Equilibrium Model of the US Economy (the EY GE Model ). This model is designed to capture the major features of the US economy and the key economic decisions of businesses and households affected by tax policy. It is an overlapping generations (OLG) model similar to those used by the Congressional Budget Office (CBO), JCT, and US Department of the Treasury. 8 Businesses and households incorporate the after-tax return from work and savings into their decisions of how much to produce, save, and work. The model is initially calibrated to reflect a stylized 2013 US economy. 9 A technical description of the EY GE Model is provided in Appendix A. An important aspect of this type of model is that policy changes are assumed to be financed by an offsetting change in fiscal policy, either through a change in tax policy or government spending. In the case of an analysis of a policy that reduces taxes, this element of the model means that tax cuts are required to be paid for in a manner that leaves the federal government on a fiscally sustainable path. In the case of tax increases, such as the enactment of a new limitation on the deductibility of interest expenses, the additional revenue is assumed to be used to finance either an offsetting reduction in the CIT rate or an increase in government spending. 10 The macroeconomic impacts of both a 10% and 30% interest-to-ebitda ratio limitation on the deductibility of net interest expenses of inbound companies in the United States are estimated in this analysis. The simulations pair these limitations with corresponding revenue-neutral reductions in the CIT rate, as well as an alternative policy scenario of increased government 10

13 spending discussed in Appendix A. Additionally, the sensitivity of the estimated macroeconomic impacts to key model parameters is examined. Estimated macroeconomic impacts of an interest expense limitation on inbound companies for the US economy Both interest expense limitations on inbound companies are estimated to reduce annual US GDP in the long-run, even when combined with a revenue-neutral reduction in the CIT rate. Long-run annual GDP is also estimated to decline if the increased revenue were instead used to finance higher government spending. 11 If the increased revenue from the implementation of the 10% limitation were used to reduce the CIT rate in the framework of a base-broadening and rate-reducing tax reform the GDP decline would be $11.1 billion each year, in the long-run. If the interest-to-ebitda ratio is set at 30%, GDP would decline would by $4.4 billion each year in the long-run (scaled the US economy in 2013). Table 2 summarizes these impacts for the interest expense limitations on annual GDP, in the long-run. Table 2. Estimated annual impacts of limitations on interest expense deduction, in the long-run Billions of US dollars change in GDP; percent change in GDP Effect on GDP Effect on GDP Policy Disallowed interest Reduce corporate income tax rate Increase government spending 10% limitation 41% -$ % -$ % 30% limitation 18% -$ % -$ % Note: Long-run estimates are scaled to the 2013 US economy. The 10% and 30% limitations are estimated to fund a 0.22 and 0.08 percentage-point reduction in the CIT rate, respectively. Source: EY analysis. As shown in the table, using the increased revenue from the interest expense limitations to increase government spending would have more negative impacts because, unlike reducing the CIT rate, the increase in government spending is assumed to not impact the cost of capital. 12 In general, if increased revenue from an interest expense limitation was used to finance government spending instead of reducing CIT rate, the decline in annual GDP would be $0.4 billion greater under a 10% limitation and $0.1 billion greater under a 30% limitation. Additional economic impacts for both the interest limitations paired with the revenue-neutral CIT rate reduction and increased government spending are detailed below. 10% limitation on interest expense of inbound companies The impacts of the limitation on interest expenses vary by industry depending on whether an industry has a sizable share of inbound companies, prevalent use of debt to finance investments, and capital-intensive production processes. These results are described in Figure 2. 11

14 Figure 2. Estimated annual impacts of 10% limitation under revenue-neutral reduction in the CIT rate scenario, in the long-run Billions of US dollars change in GDP; percent change in GDP 10% Limitation Agriculture Mining Utilities Construction Manufacturing Wholesale trade Retail trade Transportation and warehousing Information Finance and insurance Real estate and rental and leasing Services Owner-occupied housing % -0.03% -0.15% -0.03% -0.11% -0.09% -0.05% -0.07% -0.04% -0.02% -0.11% -0.02% -0.03% Note: Long-run dollar figures are scaled to the 2013 US economy. The business sector is the combination of the corporate sector (C corporations) and the pass-through sector (S corporations, partnerships, limited liability companies, sole proprietorships); it excludes owner-occupied housing. Source: EY analysis. Each of the most impacted industries are capital intensive, with manufacturing and real estate seeing reductions of economic activity greater than $2.5 billion per year. 13 The economic impacts also reflect the shifting of economic activity in the US economy due to changes in relative prices. The manufacturing and mining industries, for example, are predominantly organized in corporate form and include substantial corporate activity both inside and outside of inbound companies. The concentrated impacts on capital-intensive industries are a result of the policy causing a net increase in the cost of capital that raises the price of capital relative to labor. Accordingly, this change in relative prices leads to both the substitution of labor for capital, as well as a shift in economic activity from capital-intensive industries towards laborintensive industries. Moreover, the increased labor intensity in a smaller economy is estimated to reduce the after-tax wage and labor income. 14 The economic impacts on major macroeconomic indicators for this simulation are presented in Table 3. The reduction in GDP in the US economy is driven primarily by a reduction in investment, especially foreign investment. Foreign investment in the US economy is estimated to fall by $4.8 billion under the 10% limitation. 12

15 Table 3. Annual effect of 10% limitation on macroeconomic indicators under revenueneutral reduction in the CIT rate scenario, in the long-run Billions of US dollars; percent change 10% Limitation GDP % Consumption % Investment % Foreign investment % Capital stock % Labor income % After-tax wage n/a -0.13% Note: Long-run dollar figures are scaled to the 2013 US economy. Source: EY analysis. 30% limitation on interest expense of inbound companies The second set of policy simulations is for the enactment of a 30% interest-to-ebitda ratio limitation on deductibility of interest expenses for inbound companies. Similar to the 10% limitation, the impacts of the limitation vary significantly depending on the amount of economic activity occurring in inbound companies for an industry, the reliance on debt financing to fund investments in affected industries, and the overall capital intensity of an industry. This can be seen in Figure 3. Figure 3. Annual effect of 30% limitation on long-run GDP under revenue-neutral reduction in the CIT rate scenario Billions of US dollars change in GDP; percent change in GDP 30% Limitation Agriculture Mining Utilities Construction Manufacturing Wholesale trade Retail trade Transportation and warehousing Information Finance and insurance Real estate and rental and leasing Services Owner-occupied housing % -0.08% -0.06% -0.01% -0.04% -0.03% -0.02% -0.03% -0.01% -0.01% -0.04% -0.01% -0.01% Note: Long-run dollar figures are scaled to the 2013 US economy. The business sector is the combination of the corporate sector (C corporations) and the pass-through sector (S corporations, partnerships, limited liability companies, sole proprietorships); it excludes owner-occupied housing. Source: EY analysis. 13

16 The economic impacts on major macroeconomic indicators for the 30% limitation are presented in Table 4. The 30% limitation paired with a revenue-neutral reduction in the CIT rate is estimated to reduce long-run annual GDP by $4.4 billion. Foreign investment in the US economy is estimated to fall by $1.9 billion. Table 4. Annual effect of 30% limitation on macroeconomic indicators under revenueneutral reduction in the CIT rate scenario, in the long-run Billions of US dollars; percent change 30% Limitation GDP % Consumption % Investment % Foreign investment % Capital stock % Labor income % After-tax wage n/a -0.05% Note: Long-run dollar figures are scaled to the 2013 US economy. Source: EY analysis. Estimated change in FDI and employment contribution of inbound companies without a revenue-neutral CIT rate reduction The report also estimates the impacts of further limits on the deductibility of interest expenses by inbound companies on FDI and employment without netting the impacts with a revenueneutral change in the CIT rate. The United States had the world s largest inward FDI stock of nearly $2.8 trillion at the end of Inbound companies directly employ 5.8 million people in the United States and pay wages that are 33% higher than the national average. The total employment impacts of decreased investment from inbound companies are the sum of three distinct types of impacts, described below: Direct impacts of inbound companies. The direct employment impacts reflect the change in total number (headcount) of full- and part-time employees that are directly employed by inbound companies. The limitations on the deductibility of interest expenses would increase the cost of capital faced by foreign companies in the United States and can be expected to decrease the amount of FDI that flows into the US economy. As a result, the reduction in investment from inbound companies into the US economy would lead to fewer jobs in these companies. Indirect impacts related to suppliers. The indirect employment impacts of the limitations on the deductibility of interest expenses are the supplier-related economic activities that result from purchases of goods and services by inbound companies included in the direct employment impacts. Purchases by inbound companies from other US businesses providing goods and services support employment in these businesses. In turn, these suppliers 14

17 purchase operating inputs, which supports additional rounds of indirect employment impacts. Induced impacts related to consumer spending. The induced employment impacts reflect the impacts from employee spending. Employees of companies directly and indirectly affected by the proposed limitations on the deductibility of interest expenses use a portion of their incomes to purchase goods and services from US businesses. These transactions support employment at businesses such as retailers, restaurants, and service companies. Raising the cost of capital for inbound companies investing in the United States would cause them to reduce their investments and, over time, reduce the stock of FDI in the US economy and employment at inbound companies, their suppliers, and at the businesses where those workers spend their income. The interest limitations are estimated to reduce FDI into the United States by between 2.4% ($66 billion) (30% interest limitation) and 7.1% ($196 billion) (10% interest limitation) and reduce the total employment contribution of inbound companies by between 515,000 (30% interest limitation) and 1.5 million (10% interest limitation). Table 5 summarizes the estimated FDI and employment impacts under the potential interest limitations. A summary of the estimated long-run national FDI and employment impacts of the two interest limitations across all 50 states and the District of Columbia is provided in Appendix C. Table 5. Estimated long-run national FDI and employment impacts without a revenueneutral CIT rate reduction Percent change in FDI stock; number of full- and part-time employees (thousands) Policy FDI Direct employment Indirect employment Induced employment Total employment 10% limitation -7.1% ,505 30% limitation -2.4% Note: Employment impacts are scaled to the 2012 US economy. Figures may not sum due to rounding. Source: 2012 IMPLAN model of the US economy; EY analysis. 15

18 V. Limitations and caveats Any modeling effort is only a rough approximation of potential impacts, and the modeling used for this analysis is no exception. Although various limitations and caveats might be added to the analysis, several are particularly noteworthy: Estimates based on stylized depiction of the US economy. The general equilibrium model used for this analysis is, by its very nature, a highly stylized depiction of the US economy intended to capture key details important to analyzing the impacts of a potential tax policy change. United States on a fiscally sustainable path. The model assumes the United States is on a fiscally sustainable path under current law and remains on a fiscally sustainable path after the policy change, when neither may necessarily be the case. Estimates limited by calibration. This model is calibrated to the recent US economy (in 2013) and, because any particular year contains unique events, no particular baseline year is completely generalizable. Multinational corporations responsive to normal and supernormal investment margins. Multinational corporations in this model are assumed to earn economic profits and be responsive to the marginal effective tax rate (METR) for the normal return to investment and statutory CIT rate for the supernormal return to investment. This contrasts with other sectors of the US economy that are assumed to only earn normal returns on investment and are thereby only responsive to the METR. Estimates are limited by available public information. The analysis relies on information reported by federal government agencies (primarily the Bureau of Economic Analysis, the Bureau of Labor Statistics, and US Census Bureau), financial statement data of publicly traded companies, and aggregate tax return information (from the IRS). The analysis did not attempt to verify or validate this information using sources other than those described in the report. Estimates are limited by available data on subsidiaries. The proposed limitations are calculated at the level of the individual subsidiary, but there is little publicly available data on the EBITDA or interest expenses of individual subsidiaries. Therefore, this analysis uses public financial statement data from non-us multinational companies to estimate the proportion of interest expenses that would be disallowed for deduction. 16

19 VI. Summary Recent tax reform plans have included provisions to further limit the deductibility of interest expenses for inbound companies to help finance lower tax rates or otherwise reform the international tax system. An important aspect of tax reform, however, is carefully balancing competing objectives: While lower tax rates themselves can help encourage greater economic growth, how such lower tax rates are financed can also impact economic growth. This analysis examines the macroeconomic impacts of further limitations on the deductibility of interest expenses of inbound companies paired with alternative uses of the revenue a lower CIT rate and higher government spending. The analysis finds that limiting the deductibility of interest expenses to 10% or 30% of EBITDA increases the cost of capital in the economy, even when combined with lower CIT rates. The higher cost of capital is found to discourage business investment, which adversely affects the overall economy. The combined effect of these changes is to adversely affect the economy across different uses of the associated revenue and under a range of modeling assumptions. Understanding the potential impacts and tradeoffs associated with using the revenue from specific provisions, such as further limitations of the deductibility of interest expenses for inbound companies, is an important consideration in designing a pro-growth tax reform plan. 17

20 Appendix A. EY General Equilibrium Model of the US Economy, additional estimated impacts and sensitivity This appendix describes the EY General Equilibrium Model of the US economy, presents estimated macroeconomic impacts pairing the interest limitations with a revenue-neutral increase in government spending, and reports the sensitivity of estimated impacts to changes in key parameters values. Description of EY General Equilibrium Model of the US Economy The EY GE Model was used to estimate the long-run macroeconomic impacts associated with the limitations on the deductibility of interest expenses of inbound companies. In this model tax policy affects the incentives to work, save and invest, and allocate capital and labor among competing uses. Representative individuals and firms incorporate the after-tax return from work and savings into their decisions of how much to produce, save, and work. The general equilibrium methodology accounts for changes in equilibrium prices in factor (i.e., capital and labor) and goods markets and simultaneously accounts for the behavioral responses of individuals and businesses to changes in tax treatment. Behavioral changes are estimated in the OLG framework, whereby representative individuals incorporate changes in current and future prices when deciding how much to consume and save in each period of their life. The EY GE Model is similar to those that have been used by the CBO, JCT, and US Treasury Department. 15 An overview of the model follows: Production Firm production is modeled with a constant elasticity of substitution (CES) functional form in which firms choose the optimal level of capital and labor subject to the gross-of-tax cost of capital and gross-of-tax wage. The model includes industry-specific detail for more than 30 industries through use of differing elasticities of substitution between capital and labor, factorintensities, and scale parameters. Such a specification accounts for differential use of capital and labor between industries as well as distortions in factor prices introduced by the tax system. Further, the production of each industry is modeled for the domestic corporate, multinational corporate, and pass-through sectors; each industry is responsive to changes in the relative gross-of-tax cost of capital by organizational form and allocates production between these sectors. Multinational corporations are responsive to the effective average tax rate (EATR) a weighted average of the marginal effective tax rate (METR) and statutory CIT rate and other sectors of the US economy are responsive to the METR. This reflects the assumption that multinational corporations earn economic profits. The industry detail included in this model corresponds approximately with 2- to 3-digit North American Industry Classification System (NAICS) codes and is calibrated to a stylized version of the 2013 US economy. Additional industry-specific modeling is included in computing the gross- 18

21 of-tax cost of capital by industry as the US tax code discriminates by asset type, organizational form, and source of finance. Specifically, each industry differs in its mix of capital types, concentration in organizational form, and debt-equity ratio. Because industry outputs are typically a combination of value added (i.e., the capital and labor of an industry) and the finished production of other industries (i.e., intermediate inputs), each industry s output is modeled as a fixed proportion of an industry s value added and intermediate inputs to capture inter-industry linkages. These industry outputs are then bundled together into consumption goods that are purchased by consumers. Consumption Consumer behavior is modeled through use of an OLG framework that includes 55 generational cohorts (representing adults age 21 to 75). Thus, in any one year, the model includes a representative individual optimizing lifetime consumption and savings decisions for each age 21 through 75 (i.e., 55 representative individuals). For each generational cohort, the endowment of human capital changes with age growing early in life and declining later in life following the estimate of Altig et al. (2001). The endowment of human capital is assumed to grow between generational cohorts at an assumed rate of technical progress (1.9%). Additionally, the population of the United States is assumed to grow at the rate of 1.5%, implying that each generational cohort is 1.5% larger than that born in the previous year. The utility of representative individuals is modeled as a CES function allocating a composite commodity consisting of consumption goods and leisure over their lifetimes. Representative individuals optimize their lifetime utility through their decisions of how much to consume, save, and work in each period subject to their preference parameters and the after-tax returns from work and savings in each period. In determining their labor supply, representative individuals respond to the after-tax return to labor, as well as their overall income levels, in determining whether to work and thereby earn income that is used to purchase consumption goods or to consume leisure by not working. Other features The model includes a simple characterization of the government. Government spending is assumed to be used for either (1) transfer payments to representative individuals or (2) the provision of public goods. Public goods are assumed to be provided by the government through the purchase of industry outputs as specified in a Cobb-Douglas function. This spending is financed in the model by collecting corporate income, individual income, and payroll taxes. Tax policy changes are assumed to be offset by a contemporaneous and offsetting change in government spending or taxes. Additionally, international capital flows are modeled through a constant portfolio elasticity approach similar to Gravelle and Smetters (2006). The approach of this model assumes that international capital flows are responsive to the difference in EATRs in the United States and the rest of the world through a constant portfolio elasticity expression. This approach represents 19

22 a compromise between the closed economy approach and the alternative of a small open economy in which capital is perfectly mobile and the international after-tax return to capital is fixed. The model also captures the impacts of the CIT on income shifting between the United States and the rest of the world. Estimated macroeconomic impacts of an interest expense limitation on inbound companies paired with a revenue-neutral increase in government spending Table A1 summarizes the estimated impacts for the interest expense limitations paired with a revenue-neutral increase in government spending on long-run annual GDP. If the increased revenue from the implementation of the 10% limitation were used to increase government spending it is estimated that long-run annual GDP would decline by $11.5 billion. Under a 30% limitation, long-run annual GDP would decline by $4.5 billion. Table A1. Long-run annual effect of 10% and 30% limitation on macroeconomic indicators under revenue-neutral increase in government spending scenario Billions of US dollars; percent change 10% Limitation 30% Limitation GDP % % Consumption % % Investment % % Foreign investment % % Capital stock % % Labor income % % After-tax wage n/a -0.10% n/a -0.10% Note: Long-run dollar figures are scaled to the 2013 US economy. Source: EY analysis. The reduction in GDP in the US economy is driven primarily by a reduction in investment, especially foreign investment. Foreign investment in the US economy is estimated to fall by $5.4 billion under the 10% limitation and $2.1 billion under the 30% limitation. While the supply of labor is approximately unchanged, the increased labor intensity in a smaller economy is estimated to reduce the after-tax wage and labor income. The impacts of the limitation on interest expenses vary by industry depending on whether an industry has a sizable share of inbound companies, prevalent use of debt to finance investments, and capital-intensive production processes. These results are described in Figure A1. Each of the most impacted industries is capital intensive, with manufacturing seeing reductions of long-run economic activity of $3.2 billion per year. 20

23 Figure A1. Long-run annual effect of 10% and 30% limitation on GDP each year sector under revenue-neutral increase in government spending scenario Billions of US dollars change in GDP; percent change in GDP 10% Limitation Agriculture Mining Utilities Construction Manufacturing -3.2 Wholesale trade Retail trade Transportation and warehousing Information Finance and insurance Real estate and rental and leasing Services Owner-occupied housing % -0.04% -0.16% 0.00% -0.14% -0.10% -0.05% -0.08% -0.04% -0.04% -0.08% -0.02% -0.04% 30% Limitation Agriculture Mining Utilities Construction Manufacturing Wholesale trade Retail trade Transportation and warehousing Information Finance and insurance Real estate and rental and leasing Services Owner-occupied housing % -0.10% -0.06% 0.00% -0.05% -0.04% -0.02% -0.03% -0.01% -0.01% -0.03% -0.01% -0.02% Note: Long-run dollar figures are scaled to the 2013 US economy. Source: EY analysis. Sensitivity of macroeconomic impacts to key parameters Ultimately, the estimated impacts depend on a combination of the structure of the model and the assumption on how responsive businesses and households are to changes in after-tax rewards, such as the wage rate and the after-tax return to capital. In the baseline simulations this analysis uses parameter values reflecting key business and household behaviors that approximate central tendency estimates from prior research and recent analyses that use models of similar structure. However, uncertainty underlies the exact magnitude of these parameters. This analysis considers the sensitivity of the estimated impacts by assuming sets of 21

24 low and high values for these parameters. This approach provides a general sense for how the estimated results could vary depending on alternative views on how responsive businesses and households might be to changes in tax policy. The key model parameters chosen for the baseline, high, and low scenarios are each in the range of parameters reported in a recent Congressional Research Services (CRS) review of economic models of similar structure to the EY GE Model. 16 Specifically, for recent models of this type, the parameter range is 0.25 to 0.50 for the intertemporal elasticity of substitution, 0.50 to 1.00 for the intratemporal elasticity of substitution, and 0.30 and 0.60 for the leisure share of time endowment. 17 The baseline portfolio elasticity for capital a key determinant of the responsiveness of foreign investment to taxation comes from Gravelle and Smetters (2006). An examination of the sensitivity of the estimated macroeconomic impacts to key model parameters is presented in Table A2. For comparison purposes, the previously discussed baseline impacts are reported in addition to the low and high scenarios. Note that the impacts below are for the 10% limitation paired with a revenue-neutral reduction in the CIT rate. The long-run annual GDP is estimated to decline by between 0.04% (low responsiveness) and 0.08% (high responsiveness), bounding the baseline estimated impact of a 0.06% decrease in the long-run annual GDP. Similar results are estimated for the other macroeconomic indicators. Investment, for example, is estimated to decrease by 0.11% in the low responsiveness scenario and by 0.21% in the high responsiveness scenario relative to the baseline result of a 0.16% decrease. The investment responsiveness is largely due to the sensitivity of the change in foreign investment to parameter values; the impacts on foreign investment range from 1.58% (low responsiveness) to 4.99% (high responsiveness). 22