Offshore Profit Shifting and Domestic Productivity Measurement

Transcription

1 Offshore Profit Shifting and Domestic Productivity Measurement F. Guvenen, R.J. Mataloni Jr., D.G. Rassier, and K.J. Ruhl STL FRB Macro-Trade Workshop April 2017

2 MNE profit shifting Multinational enterprises (MNE) 1. Intangible assets can be located anywhere 2. Access to heterogeneous tax locations MNE may assign assets in ways that do not reflect production 1

3 MNE profit shifting Multinational enterprises (MNE) 1. Intangible assets can be located anywhere 2. Access to heterogeneous tax locations MNE may assign assets in ways that do not reflect production Breaks the link between production and location in NIPAs 1

4 Profit shifting example: iphone Developed in California, built by contract manufacturer in China Hypothetical numbers Parts and assembly labor $250 Sale price $750 No further costs, all phone sold outside the U.S. $500 gross profit is a return to design, software, etc. made in U.S. $250 cogs is not U.S. GDP. If Apple U.S. retains ownership of intangible property $500 is an export in U.S. GDP If Apple U.S. legally transfers ownership to affiliate outside of U.S. Affiliate may underpay for use of intangible property Return on intangible asset accrues to affiliate; not in U.S. GDP 2

5 Assets in U.S.-owned foreign affiliates, 2012 Ratio of U.S.-owned foreign affiliate total assets to PPE Compensation Employment (mil. USD) World Canada Ireland Luxembourg 1, , Netherlands Switzerland Barbados , Bermuda , U.K.I., Caribbean , Hong Kong Singapore

6 Profit shifting and mismeasurement Foreign affiliate uses intangible capital (IC) created by parent For use of IC, affiliate may: 4

7 Profit shifting and mismeasurement Foreign affiliate uses intangible capital (IC) created by parent For use of IC, affiliate may: Make payment to parent +GDP 4

8 Profit shifting and mismeasurement Foreign affiliate uses intangible capital (IC) created by parent For use of IC, affiliate may: Make payment to parent +GDP Keep payment as profit in affiliate + income on USDIA Make payment to a low-tax affiliate + income on USDIA 4

9 Profit shifting and mismeasurement Foreign affiliate uses intangible capital (IC) created by parent For use of IC, affiliate may: Make payment to parent +GDP Keep payment as profit in affiliate + income on USDIA Make payment to a low-tax affiliate + income on USDIA GNP = GDP + income on USDIA income on FDIUS + Income on USDIA explodes in the 2000s 4

10 4.5 Income on U.S. direct investment abroad share of business sector value added (percent)

11 Adjusting for profit shifting Reallocate income on USDIA across units of the MNE Reallocations to parent increase GDP GNP = GDP + income on USDIA income on FDIUS + What part of income on USDIA is owed to the parent? 6

12 Adjusting for profit shifting Reallocate income on USDIA across units of the MNE Reallocations to parent increase GDP GNP = GDP + income on USDIA income on FDIUS + What part of income on USDIA is owed to the parent? Our approach: formulary adjustment Popular multi-jurisdictional tax adjustment Allocate income proportional to apportionment factors Apportionment factors: compensation and sales For data reasons, we focus on U.S.-owned MNEs operating abroad Affiliates of foreign-owned MNEs operating in U.S. in progress 6

13 Aggregate formulary adjustment USDIA Income share Weighted Adjustment

14 Overview of results In the aggregate, our adjustment adds : nothing : $3.6 trillion to GDP : 1.5 pps to cumulative productivity growth 8

15 Overview of results In the aggregate, our adjustment adds : nothing : $3.6 trillion to GDP : 1.5 pps to cumulative productivity growth Adjustment matters most for R&D-intensive industries These industries have the strongest slowdown 8

16 Overview of results In the aggregate, our adjustment adds : nothing : $3.6 trillion to GDP : 1.5 pps to cumulative productivity growth Adjustment matters most for R&D-intensive industries These industries have the strongest slowdown 2008: adjustment is 8 percent of GDP : 4.0 pps to cumulative productivity growth : 0.6 pps to annual productivity growth rate 8

17 Overview of results In the aggregate, our adjustment adds : nothing : $3.6 trillion to GDP : 1.5 pps to cumulative productivity growth Adjustment matters most for R&D-intensive industries These industries have the strongest slowdown 2008: adjustment is 8 percent of GDP : 4.0 pps to cumulative productivity growth : 0.6 pps to annual productivity growth rate Mismeasurement likely to continue (currently 2.5 percent GDP) 8

18 Roadmap 1. Introduction 2. Accounting framework 3. Aggregate measurements 4. Industry measurements 5. Profit shifting by foreign MNEs operating in the U.S. 9

19 Conceptual framework One U.S. parent m and one foreign affiliate a Take stock of intangible capital as given and created in the U.S. Physical capital k, skilled and unskilled labor (l s, l u ), intangible capital h, and productivity z Parent produces final goods: y m = f (z m, k m, l sm, l um, h) Affiliate produces final goods: y a = f (z a, k a, l sa, l ua, h) Intangible capital can be shared 10

20 Conceptual framework One U.S. parent m and one foreign affiliate a Take stock of intangible capital as given and created in the U.S. Physical capital k, skilled and unskilled labor (l s, l u ), intangible capital h, and productivity z Parent produces final goods: y m = f (z m, k m, l sm, l um, h) Affiliate produces final goods: y a = f (z a, k a, l sa, l ua, h) Intangible capital can be shared Assumptions Constant returns to scale Physical capital and labor: located in their production locations Intangible capital: located anywhere Perfect factor and good markets 10

21 Accounting Who owns intangible capital? MNE assigns share λ of intangible capital to parent Does not effect production, but determines location to book intangible capital payments 11

22 Accounting Who owns intangible capital? MNE assigns share λ of intangible capital to parent Does not effect production, but determines location to book intangible capital payments Affiliate earnings π a = py a w s l sa w u l ua r k k a r h (1 λ)h r h λh+r h (1 λ)h = r h (1 λ)h Parent earnings π m = py m w s l sm w u l um r k k m r h λh r h (1 λ)h + r h λh = r h λh 11

23 Accounting with MNEs Expenditure-based U.S. GDP Income-based U.S. GDP Y E = py m + r h λh r h (1 λ)h Y I = w s l sm + w u l um + r k k m + r h λh + r h λh = py m r h h + 2r h λh Payment to parent is receipts by U.S. reporters [parents] on the use of intellectual property by foreign persons, part of income on USDIA Y USDIA = π a = r h (1 λ)h 12

24 Productivity Labor productivity (to be consistent with the literature) A = YE l m = py m + 2r h λh rh l m When λ = 1 entire return to h counted in U.S. GDP When λ < 1 U.S. GDP and productivity fall 13

25 Formulary adjustment Apportionment factors: variables that correlate with production Compensation and sales Apportionment weights ω n = 1 2 w s l sn + w u l un w s l sm + w u l um + w s l sa + w u l ua py n py m + py a n = a, m 14

26 Formulary adjustment Apportionment factors: variables that correlate with production Compensation and sales Apportionment weights ω n = 1 2 w s l sn + w u l un w s l sm + w u l um + w s l sa + w u l ua Allocate income across members of MNE π ω n = ω n (π a + π m ) n = a, m And the formulary adjustment to each member is ɛ n = π ω n π n Parent s formulary adjustment + GDP n = a, m py n py m + py a n = a, m Ỹ GDP = py m + λr h h + ɛ m 14

27 Aggregate adjustments to productivity 15

28 Constructing the adjustment in the data Focus on business-sector value added For each entity in each U.S.-owned MNE 1. Compute ω n 2. Compute π ω n = ω n (π a1 + π a2 + + π m ) 3. Compute ɛ n = π ω n π n Adjusted U.S. business-sector value added is Ỹ VA = Y VA + m M ɛ m 16

29 Aggregate formulary adjustment USDIA Income share Weighted Adjustment Compensation Sales

30 Constructing the adjustment in the data Focus on business-sector value added For each entity in each U.S.-owned MNE 1. Compute ω n 2. Compute π ω n = ω n (π a1 + π a2 + + π m ) 3. Compute ɛ n = π ω n π n Adjusted U.S. business-sector value added is Ỹ VA = Y VA + m M ɛ m Adjusted productivity is à = ỸVA l m 18

31 Aggregate cumulative labor productivity growth Adjusted Unadjusted 50 percent

32 Increase in aggregate cumulative labor productivity growth percent Cumulative Annual Unadjusted Adjusted Unadjusted Adjusted

33 Adjustments in other countries, 2012 Positive adjustments: Japan, France, Italy, Russia, Argentina, Greece, Turkey, Libya, Germany, and Kenya Adjustments are too small to pass confidentiality checks Japan, France, Italy, Greece, and Germany have tax rates that exceed the OECD average 21

34 Adjustments in other countries, 2012 Positive adjustments: Japan, France, Italy, Russia, Argentina, Greece, Turkey, Libya, Germany, and Kenya Adjustments are too small to pass confidentiality checks Japan, France, Italy, Greece, and Germany have tax rates that exceed the OECD average Negative adjustments: Netherlands, Bermuda, Ireland, Luxembourg, U.K.I. Caribbean, Singapore, U.K., Switzerland, Canada, Qatar Tax havens: Netherlands, Bermuda, Ireland, Luxembourg, U.K.I. Caribbean, Singapore, Switzerland Important locations of U.S. MNE production: Canada, U.K. 21

35 Reattribution of U.S. MNE earnings, 2012 (bil. USD) Canada: $13.7 Ireland: $29.5 United Kingdom: $14.7 United States: +$280.1 Bermuda: $32.4 Luxembourg: $23.6 Netherlands: $73.0 Switzerland: $12.7 Qatar $10.0 Singapore: $19.0 U.K.I. Caribbean: $

36 Industry-level adjustments 23

37 Industry-level adjustments Assign industries to groups based on IT using, IT producing (Bloom et al 2012; Fernald 2014) R&D intensity (75th percentile MNEs by R&D/Sales) As before, compute a formulary adjustment for each MNE Add the MNE s adjustment to the industry value added Ỹ RD = i I RD Y i + m M RD ɛ m 24

38 8 Adjustment by R&D intensity share of group's unadjusted value added (percent) R&D intensive non-r&d intensive

39 Adjustment by IT usage and production share of group's unadjusted value added (percent) non-it using IT using share of group's unadjusted value added (percent) IT producing 0.5 non-it producing

40 Increase in cumulative labor productivity growth: R&D intensity R&D intensive percent non-r&d intensive R&D intensive non-r&d intensive Unadjusted Adjusted Unadjusted Adjusted

41 Increase in cumulative labor productivity growth: IT producing 4.0 IT producing percent non-it producing IT producing non-it producing Unadjusted Adjusted Unadjusted Adjusted

42 Increase in cumulative labor productivity growth: IT using non-it using 1.5 percent IT using IT using non-it using Unadjusted Adjusted Unadjusted Adjusted

43 Foreign MNEs operating in the United States 30

44 Adjusting for foreign-owned affiliates Survey: U.S.-owned affiliates operating abroad very good Survey: foreign parents operating affiliates in U.S. not very good U.S. tax rates create incentives to understate income earned in U.S. by affiliates of foreign multinationals (FDIUS) GNP = GDP + income } on {{ USDIA } income }{{ on FDIUS } + overstated understated 31

45 Adjusting for foreign-owned affiliates Survey: U.S.-owned affiliates operating abroad very good Survey: foreign parents operating affiliates in U.S. not very good U.S. tax rates create incentives to understate income earned in U.S. by affiliates of foreign multinationals (FDIUS) GNP = GDP + income } on {{ USDIA } income }{{ on FDIUS } + overstated understated Match BEA data on affiliates in the U.S. with data from Orbis on the foreign parent. No common identifier. Match about 100 technology intensive foreign-owned affiliates operating in U.S. 31

46 Profits and production of foreign MNEs degree line degree line U.S. share of worldwide profit OLS fitted value U.S. share of worldwide profit OLS fitted value 0.1 Lowess smoothed value 0.1 Lowess smoothed value U.S. share of worldwide production U.S. share of worldwide production MNEs with significant U.S. operations earn most of their profit outside of the United States 32

47 Summary Intangible capital and multinationals increasingly important Increasing potential for misattribution of value added 33

48 Summary Intangible capital and multinationals increasingly important Increasing potential for misattribution of value added Formulary apportionment shows Measurement error has increased since mid-1990s Larger for R&D-intensive industries 33

49 Summary Intangible capital and multinationals increasingly important Increasing potential for misattribution of value added Formulary apportionment shows Measurement error has increased since mid-1990s Larger for R&D-intensive industries Other places this mismeasurement could matter Capital s share of income Rate of return on U.S.-owned foreign assets vs. foreign-owned U.S. assets Importance of C- and S-corps in business income 33