Taxing Sales under the FairTax What Rate Works?

Similar documents
Taxing Sales Under the FairTax: What Rate Works?

University of Toronto November 9, 2006 ECO 209Y MACROECONOMIC THEORY. Term Test #1 L0101 L0201 L0401 L5101 MW MW 1-2 MW 2-3 W 6-8

University of Toronto November 9, 2006 ECO 209Y MACROECONOMIC THEORY. Term Test #1 L0101 L0201 L0401 L5101 MW MW 1-2 MW 2-3 W 6-8

Finance 402: Problem Set 1 Solutions

Money, Banking, and Financial Markets (Econ 353) Midterm Examination I June 27, Name Univ. Id #

Lecture Note 2 Time Value of Money

2) In the medium-run/long-run, a decrease in the budget deficit will produce:

Education Maintenance Allowance (EMA) 2017/18 Notes to help you complete the Financial Details Form

FORD MOTOR CREDIT COMPANY SUGGESTED ANSWERS. Richard M. Levich. New York University Stern School of Business. Revised, February 1999

Education Maintenance Allowance (EMA) 2018/19

Understanding Annuities. Some Algebraic Terminology.

Clearing Notice SIX x-clear Ltd

Domestic Savings and International Capital Flows

Macroeconomic Theory and Policy

Quiz on Deterministic part of course October 22, 2002

Members not eligible for this option

Survey of Math: Chapter 22: Consumer Finance Borrowing Page 1

Welsh Government Learning Grant Further Education 2018/19

Quiz 2 Answers PART I

Fall 2017 Social Sciences 7418 University of Wisconsin-Madison Problem Set 3 Answers

15-451/651: Design & Analysis of Algorithms January 22, 2019 Lecture #3: Amortized Analysis last changed: January 18, 2019

Chapter 10 Making Choices: The Method, MARR, and Multiple Attributes

Members not eligible for this option

Macroeconomic Theory and Policy

Ch Rival Pure private goods (most retail goods) Non-Rival Impure public goods (internet service)

UNIVERSITY OF NOTTINGHAM

Chapter 4 Calculation of the weight (W0)

Harmonised Labour Cost Index. Methodology

Fall 2016 Social Sciences 7418 University of Wisconsin-Madison. Transactions and Portfolio Crowding Out

Advisory. Category: Capital

Macroeconomic equilibrium in the short run: the Money market

Analysis of the Influence of Expenditure Policies of Government on Macroeconomic behavior of an Agent- Based Artificial Economic System

Tests for Two Correlations

ECO 209Y MACROECONOMIC THEORY AND POLICY LECTURE 8: THE OPEN ECONOMY WITH FIXED EXCHANGE RATES

Chapter 5 Bonds, Bond Prices and the Determination of Interest Rates

Survey of Math Test #3 Practice Questions Page 1 of 5

IND E 250 Final Exam Solutions June 8, Section A. Multiple choice and simple computation. [5 points each] (Version A)

Financial mathematics

Evaluating Performance

YORK UNIVERSITY Faculty of Science Department of Mathematics and Statistics MATH A Test #2 November 03, 2014

Price and Quantity Competition Revisited. Abstract

Spring 2018 Social Sciences 7418 University of Wisconsin-Madison. Transactions and Portfolio Crowding Out

LECTURE 3. Chapter # 5: Understanding Interest Rates: Determinants and Movements

Net Pension Liabilities, Intergener Title and Pension Reforms.

Tests for Two Ordered Categorical Variables

Consumption Based Asset Pricing

Facility Location Problem. Learning objectives. Antti Salonen Farzaneh Ahmadzadeh

Highlights of the Macroprudential Report for June 2018

Taxation and Externalities. - Much recent discussion of policy towards externalities, e.g., global warming debate/kyoto

GOODS AND FINANCIAL MARKETS: IS-LM MODEL SHORT RUN IN A CLOSED ECONOMIC SYSTEM

Risk and Return: The Security Markets Line

Elements of Economic Analysis II Lecture VI: Industry Supply

- contrast so-called first-best outcome of Lindahl equilibrium with case of private provision through voluntary contributions of households

OCR Statistics 1 Working with data. Section 2: Measures of location

Homework 4 Answer Key

EDC Introduction

Trivial lump sum R5.1

Mathematical Thinking Exam 1 09 October 2017

MgtOp 215 Chapter 13 Dr. Ahn

Final Examination MATH NOTE TO PRINTER

SPECIFIC TERMS AND CONDITIONS FOR TERM DEPOSIT-i BASED ON TAWARRUQ (Effective Date: 2 April 2018)

Real Exchange Rate Fluctuations, Wage Stickiness and Markup Adjustments

THE VOLATILITY OF EQUITY MUTUAL FUND RETURNS

Finite Math - Fall Section Future Value of an Annuity; Sinking Funds

SOCIETY OF ACTUARIES FINANCIAL MATHEMATICS. EXAM FM SAMPLE SOLUTIONS Interest Theory

CS 286r: Matching and Market Design Lecture 2 Combinatorial Markets, Walrasian Equilibrium, Tâtonnement

Memo to Bruce Bartlett: Just Do the Math

Cardholder Application Form

Measures of Spread IQR and Deviation. For exam X, calculate the mean, median and mode. For exam Y, calculate the mean, median and mode.

REFINITIV INDICES PRIVATE EQUITY BUYOUT INDEX METHODOLOGY

Labor Market Transitions in Peru

Wages as Anti-Corruption Strategy: A Note

4. Greek Letters, Value-at-Risk

Administrative Services (4510P)

OPERATIONS RESEARCH. Game Theory

Principles of Finance

FM303. CHAPTERS COVERED : CHAPTERS 5, 8 and 9. LEARNER GUIDE : UNITS 1, 2 and 3.1 to 3.3. DUE DATE : 3:00 p.m. 19 MARCH 2013

A MODEL OF COMPETITION AMONG TELECOMMUNICATION SERVICE PROVIDERS BASED ON REPEATED GAME

ECON 4921: Lecture 12. Jon Fiva, 2009

Economic Design of Short-Run CSP-1 Plan Under Linear Inspection Cost

Spring 2010 Social Sciences 7418 University of Wisconsin-Madison. The Financial and Economic Crisis Interpreted in a CC-LM Model

iii) pay F P 0,T = S 0 e δt when stock has dividend yield δ.

Trivial lump sum R5.0

4: SPOT MARKET MODELS

Privatization and government preference in an international Cournot triopoly

STUDY GUIDE FOR TOPIC 1: FUNDAMENTAL CONCEPTS OF FINANCIAL MATHEMATICS. Learning objectives

An Application of Alternative Weighting Matrix Collapsing Approaches for Improving Sample Estimates

5. Market Structure and International Trade. Consider the role of economies of scale and market structure in generating intra-industry trade.

Construction Rules for Morningstar Canada Dividend Target 30 Index TM

Tuition Fee Loan application notes

c slope = -(1+i)/(1+π 2 ) MRS (between consumption in consecutive time periods) price ratio (across consecutive time periods)

ISE High Income Index Methodology

ECONOMETRICS - FINAL EXAM, 3rd YEAR (GECO & GADE)

Political Economy and Trade Policy

Chapter 15: Debt and Taxes

CHAPTER 9 FUNCTIONAL FORMS OF REGRESSION MODELS

Economics 1410 Fall Section 7 Notes 1. Define the tax in a flexible way using T (z), where z is the income reported by the agent.

Actuarial Science: Financial Mathematics

Solution of periodic review inventory model with general constrains

PART-B. a from sources other than from owning race horses and Winnings from Lottery

Transcription:

Taxng Sales under the FarTax What Rate Works? by Paul Bachman Drector of Research, Beacon Hll Insttute, Suffolk Unversty Jonathan Haughton Assocate Professor of Economcs Senor Economst, Beacon Hll Insttute Suffolk Unversty Laurence J Kotlkoff Professor of Economcs, Boston Unversty Research Assocate, Natonal Bureau of Economc Research Alfonso Sanchez-Penalver Economst, Beacon Hll Insttute, Suffolk Unversty Davd G Tuerck Charman and Professor of Economcs Executve Drector, Beacon Hll Insttute Suffolk Unversty September 2006 JEL codes H2, H5 We are ndebted to Davd Burton and Karen Walby for very valuable comments and to Douglas Guffre for hs excellent research assstance

Abstract As specfed n Congressonal bll HR 25/S 25, the FarTax s a proposal to replace the federal personal ncome tax, corporate ncome tax, payroll (FICA) tax, captal gans, alternatve mnmum, self-employment, and estate and gfts taxes wth a sngle-rate federal retal sales tax The FarTax also provdes a prebate to each household based on ts demographc composton The prebate s set to ensure that households pay no taxes net on spendng up to the poverty level Bll Gale (2005) and the Presdent s Advsory Panel on Federal Tax Reform (2005) suggest that the effectve (tax nclusve) tax rate needed to mplement HR 25 s far hgher than the proposed 23% rate Ths study, whch bulds on Gale s (2005) analyss, shows that a 23% rate s emnently feasble and suggests why Gale and the Tax Panel reached the opposte concluson Ths paper begns by projectng the FarTax s 20 tax base net of ts rebate Next t calculates the tax rate needed to mantan the real levels of federal and state spendng under the FarTax It then determnes f an effectve rate of 23% would be suffcent to fund 20 estmated spendng or f not, the amount by whch non-socal Securty federal expendtures would need to be reduced Fnally, t shows that the FarTax mposes no addtonal real fscal burdens on state and local government, notwthstandng the requrement that such governments pay the FarTax when they purchase goods and servces Implementng the FarTax rate of 23% would produce $2,586 bllon n federal tax revenues whch s $358 bllon more than the $2,228 bllon n tax revenues generated by the taxes t repeals Adjustng the base for the prebate and the admnstratve credt pad to busnesses and states for collectng the tax results n a net tax base of $9,355 bllon In 20, spendng at current levels s projected to be $3,285 bllon Revenues from the FarTax at a 23% tax rate, plus other federal revenues, are estmated to yeld $3,209 bllon whch s $76 bllon less than current CBO spendng projectons for 20 The $76 bllon amounts to only 273% of non- Socal Securty spendng ($2,177 $2,101) Ths s a remarkably small adjustment when set aganst the more than 30% rse n the real value of these expendtures snce 2000 Ensurng real revenue neutralty at the federal level, gven the net base of $9,355 bllon, mples a rate of 2382% on a tax-nclusve bass and 3127% on a tax-exclusve bass These and other calculatons presented here gnore a) general equlbrum feedback (supply-sde and demand-sde) effects that could sgnfcantly rase the FarTax base (see, for example, Kotlkoff and Joksch, 2005), b) the possblty that tax evason would exceed the consderable amount automatcally ncorporated here va the use of NIPA data, whch undercount consumpton expendtures due to evason under the current tax system, and c) the roughly $1 trllon real captal gan the federal government would secure on ts outstandng nomnal debt, were consumer prces to rse by the full amount of the FarTax The FarTax redstrbutes real purchasng power from state and local governments to ther state and local ncome-tax taxpayers It does so by reducng factor prces relatve to consumer prces and, thereby, reducng the real value (measured at consumer prces) of state and local ncome tax payments, whch are assessed on factor ncomes (namely, factor supples tmes factor prces) Gale (2005) and the Tax Panel (2005) recognzed ths loss n real state and local government revenues n clamng that these governments need to be compensated for havng to pay the FarTax But what they apparently mssed s that ths loss to these governments s exactly offset by a gan to ther taxpayers Were state and local governments to mantan ther real ncome tax collectons the assumpton made here by ncreasng ther tax rates approprately, ther taxpayers real tax burdens would reman unchanged and there would be no need for the federal government to compensate state and local governments for havng to pay the FarTax on ther purchases The second s that HR 25 does not preclude state and local governments from levyng ther sales taxes on the FarTax-nclusve prce of consumer goods and servces Ths produces sgnfcantly more revenue compared to levyng ther sales taxes on producer prces Moreover, Gale (2005) and the Tax Panel (2005) arrved at a hgher tax rate because they dd not estmate the Fartax rate, but nstead estmated a sales tax of ther own desgn whch had a substantally narrower base

1 Introducton The FarTax plan, as specfed n Congressonal bll HR 25/S25 The Far Tax Act of 2005 proposes to replace most of the exstng federal taxes wth a comprehensve consumpton tax n the form of a natonal retal sales tax, effectve January 1, 20 The Act would repeal the federal ncome tax (ncludng the captal gans tax and the alternatve mnmum tax), the corporate ncome tax, federal payroll taxes, the self-employment tax, and the estate and gft tax The Act s ntended to be revenue neutral, and would replace lost federal revenue wth a natonal retal consumpton tax (the FarTax ) leved at a tax-nclusve rate of 23 percent HR 25 calls for revenue, rather than spendng, neutralty Revenue neutralty commonly means usng dfferent taxes to generate the same number of nomnal dollars But most tax changes have lttle potental to change prces, so nomnal revenue neutralty generally equates to real revenue neutralty, whch, n turn, equates to real spendng neutralty The FarTax has the potental to sgnfcantly change both the prces pad by consumers and those receved by producers Consequently, focusng on nomnal revenue neutralty would beg the queston of what would happen to these prces and, thus, to real spendng levels In ths paper, we focus on real revenue/real spendng neutralty To be precse, we determne what FarTax rate s needed, not only for the federal government but also for state and local governments, to mantan ther real spendng levels after the swtch to the FarTax Focusng on real rather than nomnal neutralty has the decded advantage that one can determne the revenueneutral FarTax tax rate wthout havng to pn down what happens to the prce level As Gale (2005) ponted out and as our math confrms, the formula for the FarTax rate needed to acheve real revenue/real spendng neutralty on a flow bass s ndependent of the prce level Some crtcs of the FarTax argue that the rate needed for ths purpose would be far greater than 23 percent; Gale (2005) argues that t would be at least 31 percent 1 The most mportant fndng of our paper s that the 23 percent called for n HR 25/S 25 s, n fact, very close to the requred rate 2 Indeed, the requste 2382 percent rate s so close to 230 percent that only a 273 percent cut n non-socal Securty federal expendtures from the CBO projected spendng level for 20 s needed to accommodate a 230 percent rate Ths s a remarkably small adjustment when set aganst the more than 30 percent rse n the real value of these expendtures snce 2000 It s mportant to note that theses calculatons are based on the statc assumpton that mplementaton of the FarTax would have no effect on the tax base; n so dong, they gnore the expansve effect that the FarTax could be expected to exert on the base as t elmnates the bas aganst savng nherent n the exstng tax system These calculatons gnore a) general equlbrum feedback (supply-sde and demand-sde) effects that could sgnfcantly rase the FarTax base (see, for example, Kotlkoff and Joksch, 2005, or Tuerck et al, 2006b), b) the possblty that tax evason would exceed the consderable amount of evason automatcally ncorporated n our calculatons gven our use of NIPA data, whch undercount consumpton expendtures due to evason under the current tax system, and c) the 1 See Gale (2005) and Presdent s Advsory Panel on Federal Tax Reform, Smple, Far, and Pro-Growth (November 2005) from http://wwwtaxreformpanelgov/fnal-report/taxreform_intropdf and Gale, The Natonal Retal Sales Tax: What Would the Rate Have to Be? 889-91 2 The dfferent fndngs stem, n part, from the mstaken assumpton by Gale and, we presume, by the Presdent s Tax Panel (whch has not dsclosed ts methodology) that state and local governments need to be compensated for havng to pay the FarTax, n part from our use of updated data, n part from the focus on dfferent years, n part from other methodologcal refnements and choces, and, n part, from our decson n ths study to gnore (modulo some passng remarks) ssues of tax evason, expanson of the tax base due to general equlbrum effects, and captal gans on outstandng government debt

roughly $1 trllon real captal gan the federal government would secure on ts outstandng nomnal debt, were consumer prces to rse by the full amount of the FarTax The next secton measures the sze of the FarTax tax base Secton 3 determnes the tax rate requred to mantan the level of real non-socal Securty federal spendng under the FarTax Secton 4 consders the level of real non-socal Securty federal spendng cut needed to accommodate a 23 percent FarTax rate Secton 5 ndcates that f state and local governments contnue to collect the same real revenues from ther taxpayers, they wll be able to mantan ther real spendng levels, notwthstandng the requrement that they pay the FarTax on ther purchases Secton 6 concludes wth bref dscussons of general equlbrum feedback effects, tax evason, the huge potental captal gan accrung to the federal government from mplementng the FarTax, and what may be the FarTax s most sgnfcant feature ts potental to enhance budgetary dscplne 2 The FarTax Tax Base HR 25/S 25 calls for a tax on all consumpton of goods and servces n the Unted States Ths conssts, for the most part, of what the Natonal Income and Product Accounts defnes as personal consumpton expendtures and government consumpton expendtures Table 1 shows that consumpton, so measured, comprsed approxmately 86 percent of Gross Domestc Product (GDP) n 2005 3 Table 1 GDP and Consumpton, Unted States, 2001-2005 ($ bllons) 2001 2002 2003 2004 2005 Gross Domestc Product 10,128 10,470 10,971 11,734 12,494 Personal Consumpton Expendtures 7,055 7,351 7,710 8,214 8,746 Government Consumpton Expendtures 1,502 1,617 1,737 1,843 1,963 Total Consumpton (personal + government)* 8,557 8,968 9,447 10,058 10,709 As a % of GDP 845 857 861 857 858 Source: US Bureau of Economc Analyss, Natonal Income and Product Accounts [accessed 2006] US Congress, Congressonal Budget Offce, Budget and Economc Outlook for Fscal Years 20 to 2016 (2006) Note: * Totals may not add due to roundng Although Table 1 provdes a rough sense of the base on whch the FarTax would be leved, a number of further adjustments are requred As ndcated n Table 2, the most mportant of these have to do wth the treatment of housng and educatonal expendtures 21 Personal Consumpton Expendtures The FarTax has specal provsons when t comes to taxng housng, educaton, fnancal ntermedaton servces, and travel We also need to make an adjustment for state and local sales taxes 211 Housng 3 The remanng 14 percent conssted of gross prvate domestc nvestment and net exports, nether of whch are part of the FarTax base The FarTax treats exports and mports on a destnaton tax bass It exempts exports and taxes mports Page 4 of 34

Explct rental payments are subject to taxaton under the FarTax Implct rents on exstng owner-occuped housng and farms are not On the other hand, the FarTax mplctly taxes mputed rent on newly constructed housng va a pre-payment approach that leves the FarTax on ther ntal sale 4 Thus, we remove the value of mputed rent for housng and farm dwellngs from the base Snce purchases of new homes are counted as nvestment n new structures n the NIPA accounts, we add these fgures to the base Under the FarTax mprovements to sngle-famly homes, and realtors fees, whch represent payments for servces provded, are also taxable These expendtures are counted as nvestment and not consumpton n the NIPA tables, and they are added to the FarTax base It should be noted that, under the FarTax, there s no tax on the resale of houses or any other property that was prevously subject to the FarTax or that was owned by a consumer on the changeover date 212 Educaton Tuton and job tranng expendtures are treated as an nvestment n human captal and, as such, are excluded from the FarTax base 213 Fnancal Intermedaton The FarTax calls for the taxaton of both explct and mplct fnancal ntermedaton servces that consumers pay to fnancal servces frms Explct fnancal ntermedaton servces nclude fees for brokerage, bankng, loan orgnaton, mutual fund management and other fnancal servces, and are counted n personal consumpton expendtures n the NIPA tables Implct fnancal ntermedaton servces are defned by HR 25/S 25 as the dfference between the basc nterest rate (as defned n secton 805) and the rate pad on an nvestment, account or debt The dfference between actual nterest payments (eg, new home mortgage nterest) and basc nterest payments (the ten-year bond yeld) s taxable Thus, for example, a taxpayer wth a mortgage rate of 7% would have 29% of the mortgage nterest payment subject to tax f the Treasury rate were 5 percent Implct fnancal ntermedaton servces are not ncluded n the accountng of personal consumpton expendtures n NIPA Consequently, we have calculated our own values for mplct fnancal ntermedaton servces for home mortgage, non-proft and personal borrowng 5 4 Accordng to the Natonal Assocaton of Realtors, approxmately 23 percent of newly constructed homes are purchased for nvestment purposes These homes would not be subject to the FarTax when they are newly bult, but the payments made by the renters of these unts would be subject to the FarTax We make an adjustment to account for these purchases If the houses are later sold by the busness to a consumer (e, converted from a busness or nvestment purpose to a consumpton purpose) these sales would be taxed under the FarTax We do not attempt to estmate the revenue from ths provson It could, however, be substantal over tme 5 In Table 2, lne 9, mplct fees are mputed as follows: the excess of the basc nterest rate (as defned n secton 805 of HR 25/S 25) over the rate pad on such nvestment The value for mplct fees for home mortgages s derved by estmatng the prncpal ($6,4819 bllon n 20) by dvdng the total nterest payments lsted n NIPA Table 711, lne 16 ($4654 bllon n 20) by the new-home mortgage nterest rate lsted n table B-73 of the 2006 Economc Report of the Presdent (EROP), whch was 718% n 20 We apply the basc nterest rate defned as the 10-year bond rate lsted n Table B-73 of the EROP to the prncple ($6,4819 bllon x 520% = $3371 bllon) The dfference between total home mortgage payments and the basc nterest payments ($4654 bllon - $3371 bllon = $1283 bllon) s the taxable mplct fnancal ntermedaton fee Ths calculaton s repeated for nonproft nterest usng the new-home mortgage rate The mplct fee for personal nterest pad s calculated by applyng the basc nterest rate (three-year US Treasury securtes rate) from Table B-73, EROP to the Federal Reserve estmate for total outstandng consumer credt (for 20: $2,4149 bllon x 37% = $8935 bllon) Ths fgure s subtracted from the total nterest pad by persons lsted n NIPA Table 711, lne 17 ($244 bllon n 20) to arrve at our estmate of the mplct fnancal ntermedaton servce for personal credt that s subject to the FarTax (for 20: $244 bllon $8935 bllon = $1546 bllon) Page 5 of 34

214 Travel As a destnaton prncple sales tax, the FarTax apples to all retal purchases wthn the Unted States regardless of the natonalty of the purchaser or the orgn of the goods Adjustments to the accounts are necessary to capture purchases made by nonresdents vstng the Unted States and to subtract overseas purchases made by US resdents 6 215 Adjustng for State and Local Taxes The porton of state and local sales taxes that apples to sales at the retal level s deducted n order to avod cascadng or levyng the FarTax on top of state and local sales taxes Snce the FarTax does not apply to ntermedate transactons (busnesses-to-busness sales), the state and local sales taxes that apply to these transactons are automatcally excluded from the base We have adjusted our calculatons to reflect an estmate that 40% of state and local sales taxes apply to busness transactons 7 216 Other Adjustments Food produced and consumed on farms never reaches retal markets and s not subject to the FarTax We subtract ths fgure from the base Fnally, nonproft nsttutons are treated as persons by the NIPA tables, and thus ther consumpton expendtures are ncluded n the prvate tax base The consumpton expendtures of nonproft nsttutons comprse of ther operatng expendtures, ncludng wages and salares of nonproft workers, but do not nclude ther sales of goods and servces to ndvduals The FarTax taxes non-profts sales of goods and servces to ndvduals and ther purchases of goods and servces that are not sold on to ndvduals, ncludng captal goods However, the FarTax does not tax the salares and wages of nonproft workers, and thus an adjustment s needed We remove those salares and wages of nonproft workers that are not nvolved n the producton of goods and servces sold to ndvduals 8 We also remove the captal consumpton allowance, snce t s mpractcal to tax the consumpton of captal 22 Government Consumpton Spendng Government consumpton s ncluded n the FarTax base n order to put personal and government consumpton expendtures on an equal footng 9 Government consumpton 6 Accordng to offcals from the Bureau of Economc Analyss, NIPA Table 255, lne 112: expendtures n the US by non-resdents ncludes travel to the Unted States by non-resdents 7 Raymond J Rng, Jr, Consumers Share and Producers Share of the General Sales Tax, Natonal Tax Journal 52, no 1 (March 1999): 79-90 8 The Personal Consumpton Expendture (PCE) wthn the NIPA accounts ncludes the fnal consumpton of nonproft nsttutons servng households (NIPA Table 29, lne 57, $1837 bllon) and ther sales to households (NIPA Table 29, lne 64, $6768 bllon) We estmate and remove the wage and salary porton of the fnal consumpton expendtures of nonproft nsttutons Frst, we remove the porton of nonproft fnal consumpton expendtures that s attrbutable to educatonal nonproft nsttutons, snce they have already been removed from the base nsttutons (NIPA Table 29, lne 61 mnus lne 67, $52 bllon) Ths leaves the fnal consumpton expendtures at $1317 bllon Next we estmate the rato of wages and salares to total expendtures of non-profts by takng NIPA Table 113, lne 51 and dvdng t by the sum of NIPA Table 29, lnes 58 and 70; the result equals 5165% We apply ths rato to the $1317 bllon to get $68 bllon Ths represents our estmate of the salares and wages of nonproft employees that are not nvolved n the producton of goods and servces that are sold to households 9 Wllam G Gale, Evan F Koeng, Dane Lm Rogers and John Sabelhaus, Taxng Government n a Natonal Retal Sales Tax, Congressonal Budget Offce, Macroeconomc Analyss and Tax Analyss Dvsons, Techncal Paper Seres, No 1999-5 (Washngton DC: CBO, October 1998), 3 Page 6 of 34

expendtures currently nclude payroll taxes pad by government and ncome taxes and payroll taxes pad by ts employees wth respect to government wages They also reflect payroll and ncome taxes pad n the course of producng consumpton goods bought by government from prvate-sector frms The ntent of the FarTax s to substtute a sales tax for all of these taxes Falng to tax government consumpton, whle taxng only prvate consumpton, would make government consumpton expendtures artfcally cheap n comparson to prvate consumpton expendtures, and could cause the provson of some goods and servces to mgrate from the prvate sector to the government sector Actvtes such as trash collecton and transportaton servces are taxed under the FarTax, whether provded by government or the prvate sector 23 The Sze of the FarTax Base Snce the effectve date of HR 25/S 25 s January 1, 20, we estmate the tax base for the FarTax and the federal tax revenues that would be replaced by these proposals for calendar year 20 The Congressonal Budget Offce (CBO) provdes estmates of several mportant economc statstcs and tax revenues for the major federal taxes (see Table 3) 10 As detaled n Appendx B, we use the latest avalable CBO data to form 20 projectons of tax-base components We fnd the 20 FarTax base to be $11,244 bllon Startng wth personal consumpton expendtures of $9,772 bllon, we make adjustments for housng by addng the purchase of new homes and the mprovement of exstng homes The mputed rent for owner-occuped housng and farm dwellngs s removed snce the tax due on the mputed rent wll become prepad when the property s sold as a new dwellng 11 We also adjust for educaton tuton (excluded under the FarTax), taxable nterest and fnancal ntermedaton, foregn travel, and other tems 12 The net effect of these adjustments s to reduce the prvate consumpton base to $9,235 bllon, as Table 2 shows Next, we add government consumpton at the state, local and federal levels to the base We subtract wages pad to government employees who provde educaton and tranng, and we subtract captal consumpton allowances (snce t s mpractcal to tax the consumpton of captal) 13 We add spendng for new buldngs and equpment to the base State and local government consumpton, thus adjusted, equals $1,093 bllon; federal government consumpton equals $916 bllon These amounts sum to $11,244 bllon dollars, representng 81% of 20 US GDP as projected by the Congressonal Budget Offce 14 10 US Congress, Congressonal Budget Offce, Budget and Economc Outlook for Fscal Years 20 to 2016, Washngton DC, Government Prntng Offce, January 2006 11 Table 2, lne 2 accordng to March 2005 report by the Natonal Assocaton of Realtors, 23% of homes purchased n 2004 were for nvestment purposes Also, 79% of homes purchased for nvestment purposes are sngle-famly homes These numbers provde a bass for ths estmate 12 Table 2, lne 8 ncludes Other, (see NIPA 255, lne 110) whch conssts of (1) fees pad to busness schools and computer management tranng, techncal and trades schools, etc, and (2) current expendtures (ncludng consumpton of fxed captal) by nonproft research organzatons and by grant-makng foundatons for educaton and research Gale (1999) ncludes t whle Burton and Mastromarco (1997) exclude t We have chosen to nclude half of ts value 13 Accordng to BEA, government consumpton expendtures nclude the consumpton of fxed captal; gven the mpractcalty of collectng tax on the consumpton of captal, we have removed t from the base n the form of the captal consumpton allowance 14 US Congress, Congressonal Budget Offce, Budget and Economc Outlook for Fscal Years 20 to 2016, Washngton DC, Government Prntng Offce, January 2006, 26 Page 7 of 34

Table 2 Computaton of the FarTax Base, 20 ($ bllons) Lne Taxable Consumpton Categores 20 Source Prvate Consumpton Spendng 1 Personal Consumpton Expendtures 9,772 NIPA 115, lne 2 Housng 2 Purchase of New Homes 394 NIPA 545B, lne 36 3 Purchases of New Moble Homes 9 NIPA 545B, lne 40 4 Improvements to Sngle-Famly Homes 176 NIPA 545B, lne 42 5 Brokers Commssons on Housng 121 NIPA 545B, lne 43 6 Less: Imputed Rent on Housng -1,067 NIPA 245, lne 49 7 Less: Imputed Rent on Farm Dwellngs -15 NIPA 245, lne 51 Educaton 8 Less: Educaton Expendture -221 NIPA 245, lnes 95, 96, and 50% of 97 Fnancal Servces 9 Plus: Taxable Home Mortgage Interest 128 NIPA 711, lne 16, EROP, Table B-73 10 Plus: Taxable Nonproft Interest 5 NIPA 711, lne 18, EROP, Table B-73 11 Plus: Taxable Personal Interest 155 NIPA 711, lne 17, EROP, Table B-73 Travel 12 Plus: Expendture n US by Nonresdents 115 NIPA 255, lne 112 13 Less: Expendture Abroad by US Resdents (non-durables) -8 NIPA 255, lne 111 14 Less: Foregn Travel by US Resdents (servces) -54 NIPA 255, lne 110 (50%) Other 15 Less: Food Produced and Consumed on Farms -06 NIPA 255, lne 6 16 Less: State Sales Taxes -263 NIPA 33, lne 7 (60%) 17 Less: Salares and Wages of Non-Profts NIPA 29, lne 62 mnus lne 68, multpled by -68 52% (% of nonproft wages to total expenses) 18 Plus: Captal Spendng by Non-Profts (net of captal) 58 NIPA 67, lne 8, mnus NIPA 75, lne 20 19 Subtotal, Prvate ) Consumpton Base 9,235 Government Consumpton Spendng State and Local Government 20 State and Local Government Consumpton 1,333 NIPA 33, lne 22 21 Less: Current Educaton Spendng (Wages and Salares) -403 NIPA 63D, lne 94 State and Local Government Investment: 22 Gross Purchases of New Structures 263 NIPA 395, lne 24 23 Gross Purchases of Equpment 63 NIPA 395, lne 25 24 Less: Captal Consumpton Allowance -163 NIPA 33, lne 38 25 Subtotal, State and Local Tax Base 1,093 Federal Government Spendng 26 Federal Government Consumpton 845 NIPA 395, lne 7 27 Subsdes 60 NIPA 32, lne 31 Federal Government Investment: 28 Gross Purchases of New Structures 17 NIPA 395, lne 9 29 Gross Purchases of Equpment and Software 102 NIPA 395, lne 10 30 Less: Captal Consumpton Allowance -108 NIPA 32, lne 44 31 Subtotal, Federal Government Tax Base 916 32 Gross FarTax Base 11,244 33 As a % of GDP 81% 34 Untaxed Federal Government Spendng (GN) 272 NIPA 32, lne 28 (5723%), IRS, SOI Table 14 Note: Totals may not add due to roundng We note that when calculatng the FarTax rate we do not dscount the amount we estmate that federal government would save because of the reduced tax admnstraton and enforcement Page 8 of 34

dutes that t would have under the FarTax Ths reduced spendng would mply a lower tax burden on the prvate sector as well as state and local government, whch would then ncrease ther respectve consumpton levels leavng the FarTax base unchanged 3 The FarTax Rate Gven the base, we can calculate the rate at whch the FarTax must be leved once we know how much tax revenue needs to be rased Two man tems need to be computed, namely the 20 revenue to be replaced and the revenue needed to cover the prebate 31 Replacng Tax Revenue Table 3 detals the amount of revenue currently rased by ndvdual and corporaton ncome taxes, socal nsurance and retrement contrbutons, and estate and gft taxes on a calendar year bass taxes that would be repealed and replaced by the FarTax 15 In calendar year 2005, these taxes yelded $2,059 bllon or 165% of GDP In 20 these taxes are expected to yeld $2,288 bllon or 164% of GDP These fgures are based on CBO estmates that assume that all tax provsons scheduled to expre before 2016, ncludng the tax cuts enacted between 2001 and 2004, do not expre 16 Table 3 Revenue from Income, Payroll and Estate/Gft Taxes, 2003-20 ($ bllons) Actual Estmates Descrpton 2003 2004 2005 2006 20 Indvdual ncome taxes 798 839 945 1,019 1,101 Corporaton ncome taxes 146 212 284 298 290 Socal nsurance and retrement recepts 718 749 804 841 871 Estate and gft taxes 23 25 26 27 26 Total 1,685 1,825 2,059 2,185 2,288 Gross Domestc Product 10,971 11,734 12,494 13,262 13,959 Memo: Taxes as % of GDP 154 156 165 165 164 Sources: NIPA Table 115 Estmates from US Congress, Congressonal Budget Offce, Budget and Economc Outlook for Fscal Years 20 to 2016 (2006) Note: Totals may not add due to roundng It s worth consderng what the FarTax rate would be f t were not for the prebate To calculate the rate before the prebate s ncluded, we would dvde the gross FarTax base (lne 31 n Table 2) by the unadjusted revenues to be replaced, as lsted n Table 3 under the total for 20 to get 2, 288 2035% = 11,244 In the absence of the prebate the FarTax rate would be 2035%, well below that called for n HR 25 32 The Prebate As dscussed n Kotlkoff and Rapson (2005) and Tuerck et al (2006a), the FarTax s prebate makes the FarTax hghly progressve when measured relatve to the economcally meanngful 15 Snce the federal fscal year begns October 1, calendar year 20 contans the last 9 months of fscal year 20 and the frst 3 months of fscal year 2008 We adjusted the fscal year revenue numbers to calendar year 20 by addng 3/4 of the fscal year 20 total revenues to 1/4 of the total revenues for fscal year 2008 16 Congressonal Budget Offce, Budget and Economc Outlook for Fscal Years 20 to 2016, 105 Page 9 of 34

bass of lfetme ncome The prebate s based on the federal poverty gudelnes adjusted to remove any marrage penalty The prebate may be thought of as a rebate, except that t s pad at the begnnng of each month n advance of that month s consumpton expendtures The sze of the monthly prebate provded to a gven household s set at the amount of FarTax that household would pay over the course of the month, were t consumng at the federal poverty lne More precsely, the prebate equals the FarTax tax rate multpled by the famly consumpton allowance dvded by 12, where the famly consumpton allowance s based on the sze of the household 17 An addtonal adjustment s made n the case of marred couples to prevent a marrage penalty snce the poverty level for a famly of two s not twce the poverty level of a sngle person lvng alone Table 4 Computng the FarTax Base Reducton Due to the Prebate for 20 I Sngle Households Household Sze Famly Consumpton Allowance Number of Households Base Reducton 1 $10,016 29,858 $299,049,690 2 $13,490 12,719 $171,584,833 3 $16,965 6,645 $112,727,257 4 $20,440 3,233 $66,092,706 5 $23,915 1,441 $34,464,747 6 $27,390 489 $13,406,258 7 or more $30,864 395 $12,179,087 Subtotal, Sngle Households 54,781 $709,504,577 II Marred Households 2 $20,031 24,991 $500,599,437 3 $23,506 11,489 $270,055,951 4 $26,981 12,980 $350,222,029 5 $30,456 5,775 $175,871,370 6 $33,930 2,009 $68,177,390 7 or more $37,405 1,006 $37,636,330 Subtotal, Marred Households 58,250 $1,402,562,508 Total Prebate Base Reducton $2,112,067,084 Prebate as % of GDP 188% Take, as an example, a famly of four Its 20 famly consumpton allowance s projected to be $26,981, resultng n an annual prebate of $6,205 (023 tmes $26,981) The total famly consumpton allowance or prebate base was estmated by usng the US Department of Health and Human Servces Poverty Level Gudelnes for 2006 and US Census Bureau estmates for the number and sze of households n the Unted States The famly consumpton allowance computed for each famly sze/martal status combnaton was multpled by the number of households n each sze category to compute the total value of the prebate for that category These totals were summed to arrve at the base on whch the prebate would be calculated 33 Tax-Inclusve versus Tax-Exclusve Rates 17 The famly consumpton allowance s the US Department of Health and Human Servces poverty level gudelne plus and addtonal amount to elmnate a marrage penalty Page 10 of 34

At ths pont, we need to clarfy the dfference between tax-nclusve and tax-exclusve sales tax rates An example wll help Suppose a worker named Joe earns $125 and spends all of hs earnngs Suppose further that he pays a tax of $25 If he were subject to an ncome tax, he would earn $125 before tax, $100 after tax and spend $100 at the store Thus, he would need to earn $125 to spend $100 In the case of a sales tax, he would earn $125 and pay $125 at the store Of the $125 pad by Joe at the store, the store would remt $25 n sales tax, meanng that Joe ends up wth just $100 worth of goods and servces We may thnk of the tax rate as $25/$100 = 25%, whch s the tax-exclusve rate (t e ); alternatvely we may report the tax rate as $25/$125 = 20%, whch s the tax-nclusve rate (t ) The 23% FarTax rate set out n HR 25/S 25 s a tax-nclusve rate, as s the current personal ncome tax, whereas most state-level sales taxes are quoted on a tax-exclusve bass For ease of comparson, we report tax rates n both ways n Table 5 34 Determnng the FarTax Tax Rate In ths secton we determne the rate at whch the FarTax would need to be leved n 20 To repeat, we assume that the FarTax would be neutral n the sense that t would permt the same real expendtures by federal, state, and local government as well as cover the costs of the prebate Under current law, the federal budget balance for 20 may be wrtten as Here R1 + R2 + DEF G + TR + GN (1) R1 s the revenue from taxes to be elmnated under the FarTax (ncludng ncome and payroll taxes); R2 s the revenue from federal excse and other taxes that wll contnue to be leved after the FarTax s enacted; DEF s the federal budget defct; G s taxable federal government spendng on goods and servces; TR measures federal transfer payments to ndvduals, ncludng most Socal Securty payments, Medcad and Medcare subsdes, and socal programs such as food stamps, for whch the recpents are not taxed under current law; and GN represents federal spendng and transfers for whch the recpents would not be taxed under the FarTax, but for whch they would be under current law essentally wage and salary costs of educaton, plus nterest payments on the government debt held by the publc plus currently taxable Socal Securty benefts C : Personal consumpton at market value n 20 GS : Taxable state and local government consumpton at market value n 20 Now consder what happens wth the ntroducton of the FarTax Under the FarTax, equaton (1) becomes: (2) R + R2 + DEF = G + TR + GN + PRE + AC Page 11 of 34

In equaton (2) the subscrpt ndcates values under the FarTax, and the components that have the same basc names as n equaton (1) R2, DEF, G, TR and GN represent the same revenue or expendture components as n equaton (1) The three new terms n equaton (2) are: R the tax revenue to be rased by the FarTax n 20; PRE AC the prebate Ths s a new expendture to be fnanced by new tax revenue rased by the FarTax; and the admnstratve credt that the Federal government wll pay vendors and states for collectng the FarTax Unlke the terms n equaton (1), the terms n equaton (2) are not drectly measurable Two ssues arsng n the determnaton of the FarTax values are (1) the reacton of monetary authortes to the swtch to the FarTax and (2) the amount of revenue needed for the FarTax to cover the real expendtures that had prevously been fnanced by the exstng federal taxes Because the FarTax falls on consumpton, there s a queston of how ts mposton would affect the prces of consumer goods 341 Accountng for Changes n Consumer and Producer Prces At a macroeconomc level, prces depend on how the monetary authortes react to changes n tax polcy, macroeconomc condtons and other varables affectng prces In smple terms, the overall prce level must be consstent wth the quantty theory equaton, whereby MV = PY Here M s the money supply, V s the velocty at whch money crculates, P s the prce level, and Y s real ncome For the purpose of ths analyss, we assume that under the FarTax, V and Y would reman unchanged Therefore a rse n the prce level would be possble only f accommodated by an ncrease n the money supply 18 Put another way, wthout monetary accommodaton, prces faced by consumers under the FarTax would not rse Any changes to the level of monetary accommodaton, e ncrease n the money supply, would cause prces to ncrease n the same proporton Let us desgnate α as the percentage (whch could be zero) by whch market prces under the FarTax would exceed expected prces n 20 under current law Assumng that the monetary authortes adjust only to the FarTax n settng polcy for 20, α can take values between 0 and t e, so that 0 α t e, where t e s the tax-exclusve FarTax rate Wth no change n real ncome or the velocty of money, the maxmum amount that prces could ncrease when the FarTax s mposed s the amount of the tax, so the prce would go up by a factor of t e when there s full monetary accommodaton In general the relatonshp between pre- and post-fartax consumer prces, P, and P, s gven by (3) P P ( α ) = + 1 The current consumer prce level, P, has two components: 1 Producer prces (PP): the prces producers receve Ths component ncorporates all unt costs of producton, ncludng unt proft margns 18 In fact, Y would not reman constant, but would rse, owng to the dynamc effects that would arse from replacng the exstng tax system wth the FarTax We dscuss ths further below n connecton wth the evason ssue Page 12 of 34

2 Other federal commodty taxes (PR2): mport dutes, excse taxes and the lke Revenues from these taxes form the R2 component of the federal government revenue mentoned above Under current law ths means that consumer prces are (4) P = PP + PR 2 Snce the FarTax s leved on producer prces as well as on top of other federal commodty taxes, consumer prces under the FarTax satsfy (5) P = ( PP + PR )( 1+ t ) 2 e Now consder how producer prces pre and post mposton of the FarTax are related Ths relaton s gven by (6) = PP ( 1 T )( 1+ α ) PP, where T s the rate by whch producer prces under current law would fall absent any monetary accommodaton Note that ths rate s not necessarly equal to the FarTax rate due to the presence of other commodty taxes 19 Assumng the government adjusts the level of these other commodty taxes to mantan ther real purchasng power, we have Note that (7) = PR2 ( 1+ α ) PR 2 (8) 1 1 + te = 1 t Now, substtutng (3), (6) and (7) n (5): P ( 1+ α ) = [ PP ( 1 T ) + PR2 ]( 1+ t )( 1+ α ) = [ PP ( 1 T ) + PR2 ]( 1 t e ) ( 1 t ) = PP ( 1 T ) PR ( 1 t ) = PP + PR PP T ( t ) = P PP T e P + P + 2 P 2 P PP T = Pt, we get P (9) T = t PP P Lettng γ = we have: PP 19 As we see later, the fact that PR2 s also taxed causes T to be greater than the tax-nclusve FarTax rate, t Page 13 of 34

(10) T = γ t To calculate γ we use consumpton and R2, whch we estmate at $147 bllon n 20 Hence, we have γ Thus (10) becomes C + G + GS 11,244 = 10132 C + G + GS R2 11,244 147 = = (11) T = 10132 t 342 Dealng wth Government Purchases of Goods and Servces Let us now consder the ndvdual components of equaton (2) We start wth nomnal government expendtures G (on the rght-hand sde of the equaton) of goods and servces These expendtures must buy the same real goods and servces under the FarTax as they would under current law, except for those servces of the IRS that would no longer be needed because of the removal of dfferent taxes vald under current law Callng these IRS real savngs IRSS, G = G IRSS + α (12) ( )( 1 ) Nomnal federal transfer payments TR that are not taxed under current law must reman hgh enough to command the same goods and servces under the FarTax as they do under current law Thus, (13) TR = TR ( + α ) 1 343 Treatment of Taxable Transfer Payments and FarTax Tax-Favored Purchases Now consder transfer payments to ndvduals that are subject to ncome taxes under current law Examples here nclude government nterest payments and Socal Securty benefts Mantanng the real purchasng power of these transfer payments before and after the FarTax requres takng nto account that these payments wll no longer be subject to ncome taxaton A smlar ssue arses n the case of government purchase of educatonal servces and other commodtes that would not be subject to the FarTax Assumng the tax break s passed on to purchasers of these commodtes, the government s requred real spendng on such goods and servces wll be reduced Denote by GN the sum of taxable transfer payments plus federal purchases of goods and servces not subject to the FarTax and assume that the average federal tax rate currently beng appled to taxable transfer payments s t fr, then (14) = GN ( 1 T )( 1+ α ) GN Substtutng (11) we can wrte Page 14 of 34

(15) GN GN ( t )( α ) = 1 10132 1 fr + It s possble that some elements of GN would not undergo the once-and-for-all adjustment assumed by equaton (15) For example, HR 25/S 25 requres the ndexaton of Socal Securty benefts, whch mght be nterpreted to mean that the porton of those benefts fallng nto GN would, n practce, be adjusted upward by α but not downward by T For our purpose of mantanng government overall spendng constant n real terms, the ndexng of the Socal Securty payments ncluded n GN would cause the real value of G and/or TR to decrease correspondngly Snce we are nterested n the FarTax rate and not the actual values of G, GN and TR we consder ths approach to be vald 344 The Prebate Nomnal prebate expendtures are calculated by multplyng the total famly consumpton allowance or prebate base, denoted B, by the tax-nclusve rate (t ) and the ncrease n the prce level Hence, (16) PRE B t ( α ) = + 1 345 The FarTax s Admnstratve Credt The admnstratve credt that wll be pad to vendors and state government for collectng the FarTax, AC, s set n HR 25/S 25 at a quarter of 1% (025%) of the revenue collected by the retaler, and another quarter of 1% of the revenue collected by the state and local government The federal government gets no admnstratve credt for collectng any FarTax revenue In order to calculate the admnstratve credt, we must dentfy the sources of collecton, and for ths purpose we separate purchases done at the vendor level, predomnately retalers and professonals, from those done at the government level The latter are wages pad by the dfferent governments to ther employees Sales tax revenue collected at the vendor level ncludes all prvate and government retal purchases Ths comprses prvate consumpton, C, and the non-wage porton of G and GS Ths revenue s frst collected by the vendors, who clam a credt equal to 025% of revenues collected and send the remanng 9975% (= 100% - 025%) to the state government The state government then takes ts 025% of the amount remtted by the vendor, sendng the remander to the federal government The total admnstratve credt for ths type of revenue, as a porton of the revenue, s therefore 0499375% (= 025% + 025% (1 025%) 50% ) It s mportant to consder that federal wages comprse 32% of federal government purchases, and state and local government wages are 41% of state and local government purchases Ths means that the non-wage porton of government purchases relevant to ths type of revenue s 68% of G and 59% of GS respectvely 20 20 For the federal government, NIPA Table 62D, lne 87 (salary and wages) s dvded by the federal government tax base (G)to gve the porton of the tax base that comprses wages and salares Ths percentage s subtracted from 100% to obtan the value of non-wages s the tax base The process s repeated for state and local governments, NIPA 62D, lne 92, except that wages and salares for educaton, lne 94, ($403) are subtracted from total wages and salares snce ths s subtracted from the state and local government tax base Page 15 of 34

The FarTax on state and local government wages s only collected at the state government level, and therefore would earn a credt of only 025% Ths means that for the admnstratve credt we also have to apply a 025% factor to 41% of GS At the same tme, because the federal government wll not clam an admnstratve credt for collectng the FarTax on ts own wage payments, we do not nclude an admnstratve credt for ths porton of FarTax revenues Fnally, the prvate sector ncreases ts consumpton by IRSS on the assumpton that ths reducton n federal government spendng s passed on to taxpayers n the form of a reduced tax burden: { 050% 068 059 025% 041 } ( 1 ) (17) ( ) AC = C + IRSS + G IRSS + GS + GS t + α 346 Revenue Collecton Under the FarTax We now consder the revenue sde of equaton (2), and begn wth R, the revenue rased by the FarTax We know that the tax s leved on consumpton: personal consumpton and the consumpton of federal, state and local governments Therefore: ( ) (18) R = C + G + GS t In the above equaton we have two new terms: C : Personal consumpton at market value n 20 under the FarTax GS : Taxable state and local government consumpton at market value n 20 under the FarTax Assume that there s no monetary accommodaton The FarTax would cause producer prces and, therefore, the tax base for state and local governments to fall Unless some measure s taken, state and local government revenue would fall That would be the equvalent of state and local governments provdng a tax cut to ther taxpayers We assume that state and local governments take the necessary measures to mantan the real value of ther revenues, whch, n ths settng means rasng ther tax rates or expandng ther state sales tax bases by conformng to the FarTax base 21 And ths assumpton mples that these governments wll mantan the real value of ther consumpton purchases We extend ths assumpton to the cost savng enjoyed by the federal government n the form of reduced expendtures on the IRS: the cost savng s passed fully on to consumers Therefore, C = C + IRSS + α, (19) ( )( 1 (20) GS = GS ( + α ) 1 ) 21 States wll have an ncentve to conform ther state sales tax base to the FarTax base because HR 25 provdes that conformng states are allowed to collect state sales taxes on nternet and remote sales to resdents of ther state Other studes have estmated ths to be a potental revenue gan of between $215 bllon and $337 bllon for 2008 Page 16 of 34

Substtutng the relatonshps n equatons (12), (19) and (20) nto equaton (18): ( ) ( ) R = C + IRSS+ G IRSS+ GS t + α 1 (21) R ( C G GS ) t ( α ) = + + + 1 Now consder R2 The revenue n ths category s rased by excse taxes, mport dutes and the lke As we have mentoned prevously, ths revenue must buy the same goods and servces for the government as t dd prevously Therefore the real revenue from these sources under the FarTax must be the same as t would be under the current law Hence (22) = R2 ( 1+ α ) R 2 Let us now consder the defct We assume the defct to be fnanced by prvate savng We contnue to assume that household purchasng power remans fxed In partcular, we assume that wages wll adjust to keep purchasng power constant n real terms Therefore, we further assume savng to be constant n real terms Ths means that the defct n 20 wll be the same under the FarTax, wthout monetary accommodaton, as t would be under the current law Thus (23) DEF DEF ( α ) = + 1 347 The FarTax Tax Rate Formula Substtutng expressons (12), (13), (15), (16), (17), (21), (22) and (23) n equaton (2) gve the equaton for budget balance under the FarTax: (24) ( C + G + GS ) t ( 1+ α) + R2 ( 1+ α) + DEF ( 1+ α) = ( G IRSS)( 1 α) TR ( 1 α) GN ( 1 10132 t)( 1 α) Bt( 1 α) C IRSS ( G IRSS) GS GS t + + + + + + + + { + + + + } ( + α ) 050% 068 059 025% 041 1 We note that (1 + α ) accompanes every term n equaton (24), so t drops from the equaton Ths s mportant snce t mples that the FarTax rate s ndependent of the level of monetary accommodaton Smplfyng equaton (24): (25) [ ] + + ( ) + 09950C 00016IRSS + 09966G + 09960GS t + R2 + DEF G TR GN 1 10132 t B t IRSS = We now group the terms that are multpled by t to get: [ 09950 00016 + 09966 + 09960 + 10132 ] C IRSS G GS GN B t G + TR + GN R2 DEF IRSS = t G + TR + GN R2 DEF IRSS = 09950C 00016IRSS + 09966G + 09960GS + 10132GN B Page 17 of 34

Usng (1), R1 IRSS (26) t = 09950C 00016IRSS + 09966G + 09960GS + 10132GN B Insertng values from Table 5 and solvng gves (27) 2,228 t = = 2382% 9,189 001+ 913+ 1,089 + 276 2,112 The nformaton requred to determne the FarTax rate s set out n Table 5 The FarTax calls for the replacement of federal taxes on personal and corporate ncome, the gft and estate taxes; and the payroll tax We estmate that the revenues rased by these taxes would be $2,288 bllon n 20 under the current law We subtract the cost of the Earned Income Tax and the Chld Tax Credts, whch the federal government counts as spendng, and represents revenue that would not be rased under the FarTax HR 25/S 25 also calls for abolshng the Internal Revenue Servce, snce the states would admnster the FarTax The federal agency that would take responsblty for workng wth the states to coordnate FarTax collectons would need far fewer resources than the IRS today Therefore, we estmate that the federal government would be able to cut $8 bllon from the FY 20 Internal Revenue Servce (IRS) budget of $1101 bllon budget 22 These adjustments reduce the revenues replaced by the FarTax to $2,228 bllon Table 5 Computaton of the 20 FarTax Rate ($ bllons) Revenues to be Replaced Gross Revenue to be Replaced $2,288 Less : EITC and Chld Tax Credt -52 Total Revenue to be Replaced (R1 ) 2,236 IRS savngs (IRSS) -8 Adjusted Revenues to be rased (R1 IRSS) 2,228 Adjusted Tax Base (Inclusve of Tax) Components Personal Consumpton adjusted for Admnstratve Fee (09950C ) 9,189 State and Local Government Consumpton adjusted for Admnstratve Fee (09960GS ) 1,089 Federal Government Consumpton adjusted for Admnstratve Fee (09966G ) 913 Taxed Federal Government Transfers (10132GN ) 276 Less: IRS Savngs Adjustment (00016IRSS) -001 Less: Prebate Base (B) -2,112 Adjusted Tax Base $9,355 Therefore tax rate (t ) s 2,228/9,355, whch equals 2382% Tax-exclusve rate(t e ) s 2,228/(9,355-2,228), whch equals 3127% Note: Totals may not add due to roundng 22 BHI estmates the followng IRS appropratons for FY 20 could be cut: Flng and Account Servces ($1,619 mllon), Shared Servces Support ($1,504 mllon), Complance Servces ($4,497 mllon), Offsettng Collectons- Rembursables ($183 mllon), Exstng User Fees ($100 mllon) and New User Fees ($135 mllon) See US Department of Treasury; Department of Treasury Budget In Bref FY 20, Internal Revenue Servce; avalable at http://wwwrsgov/pub/rs-news/fybudgetnbrefpdf; (Washngton DC) 59-70; Internet; accessed 15 July 2006 Page 18 of 34

As set out n Table 5, the FarTax base needs some adjustments n order to match equaton (26) We have to adjust personal, state and local government and federal government consumptons by the deducton of the admnstratve credt fees We must add the base for the reducton n GN We reduce the base by the net effect of the IRSS n the admnstraton credt Fnally we must deduct the prebate base We, thus, calculate the adjusted base to be $9,355 bllon To rase revenue of $2,228 bllon from a base of $9,355 bllon, the rate that must be mposed s 2382% n tax-nclusve terms, or 3127% n tax-exclusve terms 4 Federal Spendng wth a 23% Rate In the prevous secton, we showed that the FarTax rate requred to keep exstng federal government spendng constant n real terms s 2382% However, HR 25/S 25 calls for a rate of 23% Although there s only a small dfference between the two rates, t would be necessary for the federal government to undergo a reducton n real spendng were the 23% rate to be mplemented Alternatvely, the FarTax could enhance economc growth enough to ncrease the FarTax base three percent, n whch case 23% would be suffcent to avod any spendng reducton (As prevously explaned, ths paper provdes a purely statc analyss, whch gnores the expansve effect that the FarTax could be expected to exert on economc actvty as t elmnates the exstng bas aganst savng In practce, therefore, t would probably be possble to mplement the FarTax at the 23% rate wthout any reducton n federal spendng In the absence of ths expansve effect, however, some reducton n spendng would be necessary) Whle ths reducton s also necessarly small, there s a queston of just how large a reducton would be requred The answer s n part poltcal, nasmuch as every government program has some consttuency that would resst even small budget cuts Here we estmate the percentage reducton n federal government spendng that would be requred under a 23% rate; gven that all spendng, except that for Socal Securty benefts, s avalable for reducton We must take nto account a number of complextes that arse n makng ths calculaton Frst, we must recognze that the avalable pool of spendng depends partly on the rate tself Some spendng (expendtures that fall under the categores of GN, AC and PRE) would be less under a 23% rate than under a 2382% rate Second, we must recall that Socal Securty spendng falls under the TR as well as the GN category Socal Securty payments would make up 2412% of TR and 4796% of GN n 20 We defne: NSS : the amount of non-socal Securty spendng that would be n place under the FarTax; and δ: the percentage of the non-socal Securty spendng (dentfed as NSS ) under a 23% rate that would need to be cut We let (28) NSS = G + 7588 TR + 5204GN + AC + PRE Substtutng ths defnton n equaton (2): Page 19 of 34