Using DINA to Evaluate U.S. Capital Income Tax Proposals

Patrick Driessen

First WID.world Conference

December 14, 2017 Author solely responsible for content

DINA from 1913 to … 2037?

• Issues on distributing income effects of U.S. capital tax and other policy proposals on recent display

• Distributional framework choices • Background on U.S. historical versus policy distributions • 3 Models: Cash-plus, DINA, and Adjusted DINA • Exercise Mechanics • The capital income tax prototypes • Results • Lessons

50 Days of U.S. agency distributions

• Unlike revenue estimates and macro analysis, no formal legislative requirement for distributions though there is political pressure

• Under one view that emphasizes meeting logistical demands, U.S. agencies

(CBO, JCT, TPC, etc.) have performed well under constraints – Healthcare mandate repeal in Senate bill (nod to DINA) – JCT separated “business” from “individual” effects for first time

• But no deficits, no comprehensive outlays, no formal accounting for foreign investors, no growth. Issues on territoriality, corporate tax incidence

• Presenting major capital income tax policy results with model that

understates baseline capital income (and overstates capital income tax rates)

• Agencies in this position because of affinity for a cash-based approach – During last two decades focused on “dynamic” scoring, corporate tax incidence,

Distributional framework choices

• Income (accrual, DINA, or realization/cash based aka “Cash-plus”) • Government outlay inclusivity (transfers, in-kind, state/local), plus

deficits/surpluses • Breadth of taxes (geography, type)

• Units of observation (households, tax units, adults)

• Welfare indicators (SWFs, pre- and post-tax, outlays, mobility) • Outstanding technical issues (lifetime tax incidence, integrating

macroeconomic effects when looking forward with policy proposals)

U.S. historical vs. policy distributions

• Historical modeling from the tax side requires managing endogeneity of tax and other law, as well as regulatory, changes

• Legislative policy distributions look forward, anywhere from 1 to 10 years (perhaps to 20 and beyond soon, depending upon budget period rule)

• PSZ ,CBO regularly offer U.S. historical distributions; JCT, OTA, and TPC not – CBO technically is involved with both historical and policy analysis

• All U.S. agencies use similar cash/realization approaches - “Cash-Plus” – Early PS research also a version of Cash-plus

• Historical/policy distributional conformity desirable

• Accrual approach, another method used for historical evaluation, likely collapses to national income or DINA-based method going forward

Cash-plus, DINA, and Adjusted DINA Models

• Capital income tax proposals good for testing models • Cash-plus concepts relatively familiar to legislators and lay people,

closer to administrative data – But see Medicare/Medicaid, fringe benefits, corporate tax incidence – For history, concerns for tax policy endogeneity, both pre- and post-tax

• DINA jumps the shark by using national income to calibrate – For pre-tax offers income classifier flexibility – Deficit allocation – All government levels, all government spending

• “Adjusted DINA” modifies DINA – BEPS link to labor bearing some corporate tax – Discounting deferred foreign earnings to lay in for territoriality – Incorporation of deemed growth for policy proposals

• For all proposals, foreign investor benefits are tracked

8 Capital Income Tax Prototypes Tested

• Corporate income tax repeal (-$400 billion) • 20% corporate tax rate (-$140 B) • Lax territoriality (-$100 B) • Corporate integration shifting liability to individuals* • Harsher worldwide taxation with corporate rate reduction* • Harsher worldwide taxation with deficit reduction* • Simple version of pending (Dec. 2017) U.S. tax legislation (-$115 B) • Pending (Dec. 2017) U.S. legislation with repeal of healthcare

mandate (-$115 B)

* Revenue neutral

Mechanics

• Baseline • Revenue Estimates + growth • Foreign investors, deficit allocation, growth allocation • Focus on post-tax (with post-tax broadly defined) • 1/99, 10/90, 60/40 (Cash-plus), 50/50 (DINA) • If growth deemed, “growth risk” analysis offered in

Adjusted DINA model

Pre-Tax Income U.S. Distributions by Model and Year*

Model

2012 2013 2014 2015

$Trillions

As a %

of PSZ

$Trillions

As a %

of PSZ

$Trillions

As a %

of PSZ

$Trillions

As a %

of PSZ

CBO $12.3 85.3%

JCT $11.7 81.1% $12.7 84.1% $13.3 84.8%

LBAA $12.5 88.8% $15.4 106.9%

OTA $13.9 88.8%

TPC $13.9 91.7%

PSZ $14.1 $14.5 $15.2 $15.7

* Author responsible for calculations. See Table 1 in text of accompanying paper. CBO is Congressional Budget Office; JCT is

Joint Committee on Taxation; LBAA is Jeff Larrimore, Richard V. Burkhauser, Gerald Auten, and Philip Armour; OTA is Office of

Tax Analysis, U.S. Treasury Department; TPC is Tax Policy Center; and PSZ is Thomas Piketty, Emmanuel Saez, and Gabriel

Zucman. LBAA is an accrual model; PSZ is DINA; and others are Cash-plus.

Cash-Plus, DINA, and Adjusted DINA U.S. Baselines, at 2013 Levels* ($ in Billions, otherwise %)

Cash-plus DINA Adjusted DINA Total Income:

Pre-tax $12,330 $14,450 $14,310

Post-tax $9,860 $14,450 $14,310

Pre-tax and post-tax

changes to DINA to

create Adjusted DINA:

Deferral discount -$40

Profit shifting -$100

Top 1% shares: Of pre-tax income 15.00% 19.60% 19.58%

Of post-tax income 12.40% 15.34% 15.28%

Top 10% shares: Of pre-tax income 37.98% 46.32% 46.31%

Of post-tax income 33.97% 38.75% 38.67%

Top 60% shares: Of pre-tax income 85.51%

Of post-tax income 83.09%

Top 50% shares: Of pre-tax income 87.23% 87.22%

Of post-tax income 80.68% 80.60%

* Author responsible for calculations, see Table 2 in accompanying paper.

Charts 1A, 1B, 1C: Annual Post-tax Income Effects of Repealing Corporate Tax, by Pre-tax Income Splits and Distributions (2013 levels, $ billions)

157 131 73 104 120 183 189

-143 -34

57 60 80 70 80 87

-500

0

500

Chart 1A. By Pre-tax 1/99 Split and Alternate Distributions

Top 1% Low 99% Foreign Investors

229 230 98

165 207 111 90

-168 -95 -29

60 80 70 80 87

-500

0

500

Chart 1B. By Pre-tax 10/90 Split and Alternate Distributions Top 10% Low 90% Foreign Investors

323 315

54

178 265

17 5

-124 -108 -87

60 80 70 80 87

-200

-100

0

100

200

300

400

Cash-plus DINA (except before deficit) DINA Adjusted DINA (pre-growth) Adjusted DINA (growth +.8%)

Distributional Approaches 60/40 Cash-plus ,50/50 DINA …

Chart 1C. By Pre-tax 60/40 and 50/50 Splits and Alternate Distributions Top Income Group Low Income Group Foreign Investors

Chart 1R: Corporate Tax Repeal - Annual Post-tax Income Growth Range Risk by Pre-tax Income Splits for Adjusted DINA (2013 levels, $ billions)

165

104

178

120

207

265

-150

-100

-50

0

50

100

150

200

250

300

Chart 1Rb. Post-Tax Growth Risk Range by High Income Group

(low number = no growth, higher number = +.8%)

Top 1% Top 10% Top 50%

57

-29

-87

-34

-95 -108

-150

-100

-50

0

50

100

150

200

250

300

Chart 1Ra. Post-Tax Growth Risk Range by Low Income Group

(low number = no growth, higher number = +.8%)

Low 99% Low 90% Low 50%

Charts 3A, 3C: Annual Post-tax Income Effects of Adopting Lax Territoriality, by Pre-tax Income Splits and Distributions (2013 levels, $ billions)

39

18

31

46

-36 -34

15 18

28

-40

-20

0

20

40

60

80

100

Cash-plus DINA Adjusted DINA (nogrowth expected)

Distributional Approaches

Chart 3A. By Pre-tax 1/99 Split and Alternate Distributions

Top 1% Low 99% Foreign Investors

81

13

32

4

-31 -34

15 18

28

-40

-20

0

20

40

60

80

100

Cash-plus DINA Adjusted DINA (nogrowth expected)

Distributional Approaches (60/40 Cash-plus and 50/50 DINA income splits)

Chart 3C. By Pre-tax 60/40 and 50/50 Splits and Alternate Distributions

Top Income Group Bottom Income Group Foreign Investors

Chart 4A: Annual Post-tax Income Effects of $200 Bils. Shift of Corporate Tax, by 1/99 Income Split and Distributions (2013 levels, $ billions)

447

-16 -16

323

-24 -24

30 40 40

-100

0

100

200

300

400

500

Cash-plus DINA Adjusted DINA (no growthexpected)

Distributional Approaches

Top 1% Low 99% Foreign Investors

Charts 5A and 5C: Annual Post-tax Income Effects of Harsh Worldwide with Corporate Tax Cut, by Pre-tax Income Splits and Distributions (2013 levels, $ billions)

14 20

86

114

2

0

20

40

60

80

100

120

Cash-plus DINA Adjusted DINA(pre-growth)

Adjusted DINA(growth +.25%)

Distributional Approaches

Chart 5A. By Pre-tax 1/99 Split and Alternate Distributions

Top 1% Low 99% Foreign Investors

85

112

15

22

2

0

20

40

60

80

100

120

Cash-plus DINA Adjusted DINA(pre-growth)

Adjusted DINA(growth +.25%)

Distributional Approaches (Cash-plus 60/40. and DINA 50/50 income splits)

Chart 5C. By Pre-tax 60/40 and 50/50 Split and Alternate Distributions

Top Income Group Low Income Group Foreign Investors

Charts 6A and 6C: Annual Post-tax Income Effects of Harsh Worldwide to Pay Down Debt, By Pre-tax Income Splits and Distributions (2013 levels, $ billions)

-39

-18

-1

-46

36

136

-15 -18 -16

-100

-80

-60

-40

-20

0

20

40

60

80

100

120

140

Cash-plus DINA Adjusted DINA(growth +.125%)

Distributional Approaches

Chart 6A. By Pre-tax 1/99 Split and Alternate Distributions

Top 1% Low 99% Foreign Investors

-81

-13

85

-4

31

49

-15 -18 -16

-100

-80

-60

-40

-20

0

20

40

60

80

100

120

140

Cash-plus DINA Adjusted DINA(growth +.125%)

Distributional Approaches (Cash-plus 60/50 , DINA 50/50 )

Chart 6C. By Pre-tax 60/40 and 50/50 Splits and Alternate Distributions

Top Income Group Bottom Income Group Foreign Investors

Charts 7C and 8C: Annual Post-tax Income Effects of Pending (Dec.2017) U.S. Legislation (with and without ACA mandate repeal) by Pre-tax Income Splits and Distributions (2013 level,$bils.)

89

24

130

5

-30

-12

6 5

22

-100

-50

0

50

100

150

200

Cash-plus DINA Adjusted DINA (growth +.8%)

Distributional Approaches (60/40 Cash-plus, 50/50 others)

Chart 7C. Without ACA Mandate Repeal, By Pre-tax 60/40 and 50/50 Split and Alternate Distributions

Top Income Group Low Income Group Foreign Investors

119

54

160

-25

-60

-42

6 5

22

-100

-50

0

50

100

150

200

Cash-plus DINA Adjusted DINA (growth +.8%)

Distributional Approaches (60/40 Cash-plus, 50/50 others)

Chart 8C. With ACA Mandate Repeal, By Pre-tax 60/40 and 50/50 Split and Alternate Distributions

Top Income Group Low Income Group Foreign Investors

Chart 8R: Pending (Dec.2017) U.S. Legislation (with ACA mandate repeal) - Annual Post-tax Income Growth Range Risk by Pre-tax Income Splits for Adjusted DINA (2013 levels, $ bils.)

-60 -46

-63

45

6

-42

-80

-40

0

40

80

120

160

Chart 8Ra. Post-Tax Growth Risk Range by Low Income Group

(low number = no growth, higher number = +.8%)

Low 99% Low 90% Low 50%

70

56

73 73

111

160

-80

-40

0

40

80

120

160

Chart 8Rb. Post-Tax Growth Risk Range by High Income Group

(low number = no growth, higher number = +.8%)

Top 1% Top 10% Top 50%

Results Summary

• Adjusted DINA offers way of modeling BEPS’s impact on labor as substitute for current U.S. agency corporate income tax distributional approach

• Adjusted DINA begins to show why investors/companies want territoriality

• Cash-plus can be misleading for integration proposals, both as to level of post-tax benefits and progressivity engendered with corporate tax change

• Tracking growth risk formalizes distributional intuition

• Generally, Cash-plus approaches show more favorable overall results, and particularly better relative and absolute results for low-income groups, for capital income tax reduction policies than DINA and Adjusted DINA models

Lessons

• A policy model that distributes deficits and outlays also should distribute growth (or non-growth) to get everything in the story

• May need more sensitivity analysis (i.e., income grouping or even going to more specific percentiles) with policy distributions than needed for historical evaluation

• Much to be done on aggregating source /home country taxation of foreign investors

• Pending U.S. tax legislation has potentially long phase-ins and phase-outs, big wealth effects, all affecting policy and historical distributions

• A consequence of U.S. tax legislation is likely greater diversion of Cash-plus from accrual and DINA-based models in historical evaluation

– And the latter two models also may diverge more from each other as well)