a report of the heritage center for data...

A Report

of The Heritage Center

for Data Analysis

A Report

of The Heritage Center

for Data Analysis

214 Massachusetts Avenue, NE • Washington, D.C. 20002 • (202) 546-4400 • heritage.org

NOTE: Nothing written here is to be construed as necessarily reflecting the views of The Heritage Foundation or as an attempt to aid or hinder the passage of any bill before Congress.

TAX INCIDENCE, TAX BURDEN, AND TAX SHIFTING: WHO REALLY PAYS THE TAX?

STEPHEN J. ENTIN

CDA04-12 November 5, 2004

CDA04—01 November 5, 2004CDA04—12 November 5, 2004

TAX INCIDENCE, TAX BURDEN, AND TAX SHIFTING: WHO REALLY PAYS THE TAX?

STEPHEN J. ENTIN

I. INTRODUCTIONWho pays the income tax, the payroll tax, the

estate and gift taxes? Who bears the burden of thegasoline and tobacco taxes? If Congress were toraise this tax rate, or lower that tax deduction,who would gain and who would lose? The out-comes of the political battles over changes in thetax system often hinge on the answers to suchquestions.

To demonstrate who pays current taxes or whowould be the winners and losers from a taxchange, the Joint Committee on Taxation of theCongress (JCT) produces “burden tables” showinghow much money everyone sends, or would send,to the Treasury. Winners and losers are grouped bytheir adjusted gross income class, and the distribu-tional impacts of a tax, or a tax change, are dis-played. Burden tables are also prepared onoccasion by the Treasury and the CongressionalBudget Office, as well as private research groups,using sometimes similar, sometimes differentassumptions and methods of display (such as by“income quintile”). The burden tables are sup-posed to shed light on the tax system or the effectof a new tax proposal, but they often do more toobfuscate than to illuminate the facts.

The true measure of the burden of a tax is thechange in people’s economic situations as a resultof the tax. The changes should be measured as theeffects on everyone’s net-of-tax income after alleconomic adjustments have run their courses. Theburden measure should include not only changesin people’s after-tax incomes in a single year, butthe lifetime consequences of the tax change aswell. Unfortunately, policymakers are not pre-

sented with this type of comprehensive informa-tion on the true burden of taxation and must makepolicy judgments based on incomplete and mis-leading statistics.

One cannot tell the true burden of a tax just bylooking at where or on whom it is initiallyimposed, or at what it is called. Taxes affect tax-payers’ behavior, triggering economic changes thatregularly shift some or even the entire economicburden of a tax to other parties, and alter total out-put and incomes. Taxes reduce and distort the mixof what people are willing to produce in their rolesas workers, savers, and investors. Taxes increasewhat these producers seek to charge for their ser-vices or products. Changes in the prices and quan-tities of output in turn affect people in their rolesas consumers when they try to spend theirincomes. The lost output and other consequencesof taxation impose additional costs on the taxpay-ers that are not reflected in the mere dollaramounts of the tax collections.

The Treasury put these problems well in its1991 study on ending the double taxation of cor-porate income, writing that:

The economic burden of a tax, however,frequently does not rest with the personor business who has the statutory liabilityfor paying the tax to the government. Thisburden, or incidence, of a tax refers to thechange in real incomes that results fromthe imposition of a change in a tax.1

These ultimate effects and burdens of taxationare explored in a corner of the economic literature,but they are nowhere to be found in the “burdentables” that are prepared by the government agen-

2

THE HERITAGE CENTER FOR DATA ANALYSIS

cies and scrutinized in tax debates. Instead, theburden tables are constructed using crudeassumptions and oversimplified rules of thumb toassign various taxes to suppliers of labor or capital,or to consumers. These assumptions and rules areoften adopted more for ease of computation thanfor economic accuracy. In fact, no burden tableever published has been based on how taxes trulyaffect incomes.

What price do we pay for glossing over the trueeconomic burden of a tax? Failure to understandand take account of the economic consequences oftaxation leads to a gross misrepresentation of thedistribution of the tax burden. This in turn has ledto a tax system that, while supposedly promotingsocial justice, is actually harmful to lower-incomeworkers and savers, as well as damaging to thepopulation as a whole. A better understanding ofthe economic consequences and real burdens oftaxation is indispensable to achieving an optimaltax system—one that minimizes the economic andsocial damage associated with financing govern-ment outlays.

A better understanding of the economic conse-quences of taxation would also benefit the Trea-sury and the Congress as they plan the federalbudget and contemplate changes in the tax system.It should lead to more accurate revenue forecast-ing. It might also encourage the adoption of taxbills that are more concerned with increasingnational and individual income and less concernedwith redistributing the existing level of nationalproduct.

This paper will discuss the economic conse-quences of taxation and the factors that influencewhere the burden of various taxes really falls. Itwill review some of the discussions in the eco-nomic literature. Finally, it will suggest that a shiftto a markedly different type of tax system wouldbenefit all players in the economy.

II. SORTING OUT SOME TERMINOLOGYThe terms “tax incidence” and “tax burden” are

thrown around rather loosely in the economic lit-erature and in the popular press. Some authors usethem interchangeably for any of several conceptsof the effect of a tax. Some authors use them forseparate concepts, but different authors do notagree as to which term means which concept. Thispaper will seek to distinguish clearly among sev-eral distinct concepts of “incidence” of a tax and toreserve a single term for each. We define threeconcepts:

• The “statutory” or “legal obligation,” whichrefers to the person on whom the law says thatthe tax obligation falls (which may bear littlerelationship to who actually feels the pain);2

• The “initial economic incidence” (or “inci-dence” for short), which is how the economicsupply and demand conditions in the marketfor the taxed product or service or factor of pro-duction allocate the tax among suppliers andconsumers of the taxed item (which allocationmay be different in the short run and the longrun); and

• The “ultimate economic burden” (or “burden”for short), which measures the changes in peo-ple’s after-tax incomes after all the economicadjustments to the tax have occurred across allaffected markets as consumption behavior,resource use, and incomes shift to their newpatterns.

These definitions distinguish between the terms“incidence” and “burden.” “Incidence” is definedas the partial own-market economic effects of thetax, which may also be thought of as partial equi-librium analysis. “Burden” is defined as the generalequilibrium economic results involving all mar-kets. When the paper quotes other sources thatemploy the terms differently, the reader must per-form the required mental translation.3

1. U.S. Department of the Treasury, Report of the Department of the Treasury on Integration of the Individual and Corporate Tax Systems: Taxing Business Income Once, January 1992, p. 146.

2. Statutory obligation is not the same thing as the obligation to remit, which involves the tax collection laws and pro-cess. A tax is in a sense “paid” by whoever is legally responsible for remitting the money to the taxing authority, whether that is the U.S. Treasury or one of the various state and local tax departments or offices. Most people are sophisticated enough to realize that who sends in the check does not indicate who pays the tax. Income tax withhold-ing is a good example. A worker’s employer by law must transmit income taxes withheld from a worker’s paycheck to the Treasury each pay period, but the tax actually falls according to statute on the worker’s wages, not on the employer’s income. “Remittance” is not the same thing as “statutory obligation.”

3


Chart 1 CDA 04-12

Imposition of a Tax

Quantity

Pric

e

Supply(No Tax)

DemandDemand

Tax

Q

P0

Q1

Supply(With Tax)

Reduction in Value ofEconomic Output =

Loss to Consumer

+

Loss to Producer

Pp

Pc

ResourcesRedirectedto other Activities

E1

E0

III. THE SIMPLE EXAMPLE OF A SELECTIVE EXCISE TAX: STATUTORY OBLIGATION, INITIAL INCIDENCE, ULTIMATE BURDEN

Charting a Simple Excise TaxConsider the imposition of a selective

excise tax, such as the cigarette tax or thegasoline tax. (See Chart 1.) In the absenceof the tax, supply would equal demand atthe equilibrium point E0, with a unit priceof P0 and a quantity of Q0 units.

Imposing a per unit tax of t = (Pc–Pp)drives a wedge between the price paid bythe consumer (Pc) and the price received bythe producer (Pp). As the gross price to thebuyer is driven up, the quantity demandedshrinks (movement along the demandcurve). As the net price received by theseller falls, less is supplied (movement alongthe supply curve). The quantity of outputfalls from its original value (Q0) to its new value(Q1). Market equilibrium shifts from E0 to E1.

Tax revenue is t x Q1 (the shaded area, unit taxtimes quantity). Note that the revenue is not equalto t times the original quantity of the product inthe absence of the tax; it is t times the reduced out-put brought about by the tax. In usual parlance,the upper portion of the revenue rectangle, (Pc–P0) x Q1, is considered to be the share of the taxthat falls on the consumer because he now pays ahigher tax-inclusive price. The bottom portion ofthe rectangle, (P0–Pp) x Q1, is considered to bethe share of the tax that falls on the producer inthe form of a lower net-of-tax price and revenuereceived for selling the product.

The reduction in output deprives the consumerof the value he places on the lost output, the tallertrapezoidal area under the demand curve betweenQ0 and Q1. The reduction in output frees upresources for other uses equal to the shorter trape-zoidal area under the supply curve between Q0and Q1. The shaded triangle between the supplyand demand curves is the dead weight social costof the tax, representing the excess value of the lostproduct over its resource cost, split between theconsumer and the producer.

The imposition of the tax is sometimes illus-trated as a backward shift in the supply curve(shifting the tax-inclusive supply curve to passthrough point E1, labeled “supply with tax” in thediagram). This can be viewed as showing the tax

3. See Don Fullerton and Gilbert E. Metcalf, The Distribution of Tax Burdens, International Library of Critical Writing in Economics, No. 155 (Northampton, Mass.: Edward Elgar Publishing, Inc., March 1, 2003). Fullerton and Metcalf use the term “statutory incidence” to refer to the statutory obligation as defined by the tax law (what is here called “statu-tory obligation”). They use the term “economic incidence” to refer to the changes in people’s economic welfare brought about by the tax, in that the tax changes equilibrium prices, with wide-ranging consequences, what is here called “ultimate economic burden.” For example, a tax on a particular product induces consumers to alter their pur-chases, which in turn affects the prices or returns paid to each input, thereby affecting the welfare of consumers, workers, and suppliers of capital. Two terms are not really enough, however. There is still the need to distinguish between the economic incidence revealed by “partial analysis,” which involves the changes in the price of the taxed product and its effect on that product’s consumers and producers (and which must further be broken down into the short- and longer-run effects), and “general equilibrium analysis,” which must include all the subsequent adjustments as consumers switch to other products and factors shift to other uses, including leisure, or are reallocated between consumption and capital accumulation, altering the capital stock over time and affecting wages throughout the econ-omy. The Fullerton–Metcalf anthology contains many seminal papers on tax incidence that explore these different fac-ets of the analytical spectrum.

4


to be a cost of calling forth the product. Alterna-tively, it is described as a representation of a taximposed on the consumer, emphasizing the highergross price paid as the result of the tax. The taxmay also be drawn as a backward shift in thedemand curve, shifting it to pass through the pointwhere price equals Pp and quantity equals Q1.This is sometimes described as illustrating a taximposed on the producer, emphasizing the receiptby the producer of the lower net-of-tax price.

Whether the tax is described as being paid by theproducer or by the consumer, the outcome is thesame: The rise in the price to the buyer to Pc, thedrop in the price to the seller to Pp, and the dropin production to Q1 are identical whichever viewis taken and depend entirely on the rate of the taxand the slopes (elasticities) of the supply anddemand curves. Elasticity will be discussed ingreater detail below.

Statutory or Legal Obligation of an Excise TaxWho pays a selective excise tax? The legal obli-

gation to pay would depend on the wording of thestatute. It might be called either a consumer-leveltax (e.g., the gasoline excise tax, collected at thepump) or a producer-level tax (e.g., the alcoholand tobacco taxes, collected from manufacturers).

As the diagram shows, the distinction is eco-nomically meaningless and does not reflect theeconomic division of the tax burden. Consumersand producers are both affected to some degree,regardless of the statutory label. How they sharethe incidence of the tax depends entirely on theirresponsiveness to the price changes, the slopes ofthe supply and demand curves, not on whetherthe wording of the statute charges the consumerwith the tax and it is merely collected by the sellerand forwarded to the government, or whether thestatute names the seller as being charged with thetax directly.

Economic Incidence of an Excise TaxThe initial economic incidence is properly cal-

culated as partly falling on consumers, to theextent of the revenues they pay plus their share ofthe deadweight loss triangle, and partly falling onproducers, to the extent of the revenues they payplus their share of the deadweight loss. Producersare the workers who supply labor and the inves-tors who supply capital to a business. What do wemean by saying that part of the excise tax falls onproducers? When a tax is imposed on a final prod-

uct, the reduction in demand for and output of theproduct in turn reduces the demand for the inputsused to produce the product, which reduces work-ers’ wages and investors’ returns on saving.

Note that most consumers are also workers and/or suppliers of capital (unless they are livingentirely on welfare or other transfer payments).The excise tax, insofar as consumers pay it or inso-far as it leads them to reallocate their resources tosecond-best choices, reduces the quantity andvalue of what they can buy with an extra dollar ofincome. The tax devalues their earnings from laboror saving. That is, insofar as the tax is “passed for-ward” to consumers, it is ultimately a tax on theirlabor and capital income. All taxes are ultimatelytaxes on income, which is to say, on producers. Anexcise tax falls either on the labor and capitalemployed in the taxed industry or on the consum-ers, who happen to provide labor and capital ser-vices in other industries.

Incidence and Elasticity. How buyers and sell-ers share the initial incidence of a tax depends ontheir market behavior. The portion of the tax pre-sumed to be paid by the buyer or the seller variesdepending on the responsiveness of the demandfor and the supply of the product or input as theprice changes. In the chart, this is reflected in thesteepness of the demand and supply curves.

“Elasticity” is the percent change in the quantityof a product (or factor of production—labor, capi-tal, land, etc.) supplied or demanded divided bythe percent change in its price (or wage or rate ofreturn). For example, if people are easily discour-aged from buying a particular product (or employ-ing a particular factor) as its price rises, then thatratio will be high, the demand for the product (orfor the factor) is said to be elastic, and the demandcurve is rather flat. If people are unwilling to giveup much of the product (or factor) even if theprice rises sharply, the ratio will be low, thedemand is said to be inelastic, and the demandcurve is steep.

The elasticities of demand and supply tend to begreater in the long run than the short run. It maytake some time for people fully to adapt to a taxchange. For example, in the short run, a rise in thetax on gasoline may encourage people to drivetheir existing cars less by taking fewer trips, by carpooling, or by switching to public transportation.Longer-term, people may replace their existing

5


Chart 2a CDA 04-12

Perfectly Elastic Supply

Quantity

Price

Supply

Demand

Pc

Tax

Q0

P0

Q1

Chart 2b CDA 04-12

Perfectly Elastic Demand

Quantity

Price

Supply

DemandP0

Pp1

Tax

Q1

Q0

cars with models that offer higher fueleconomy or may move closer to theirwork. The long-run demand for gasolineshould be more elastic than the short-rundemand.

Four Extreme Cases of Elasticity.There are four extreme or limiting cases—not generally seen in the real world—thatillustrate the concept of elasticity and itsimplications.

• Perfectly Elastic Supply (Chart 2a). If aproduct is easily reproduced orobtained at the same cost per unit, nomatter how many units are sought,then the supply curve is horizontal andthe net-of-tax price is fixed at that mar-ginal cost. (Example: the supply in asmall town of a commodity soldnationally [say, Budweiser]). If thebuyers in the town are willing to paythe market price, they can get a virtu-ally unlimited supply [or at least all they canhold]. If they are not willing to pay that price,they will get none.) Any tax is borne by theconsumer. Output or availability will fall ifdemand is price-sensitive.

• Perfectly Elastic Demand (Chart 2b). If demandis perfectly elastic, any rise in the price wouldcause a collapse in consumption. (Example:

the demand for beer at one out of 12 conces-sion stands at a stadium. If one stand tries tocharge more than the others, it will lose all itsbusiness to the other stands.) The demandcurve is horizontal, and the market price isfixed. Any tax imposed (on that one beer out-let) will simply lower the net-of-tax price tothe producer, who must bear the whole tax.Output will fall if supply is price-sensitive.

• Perfectly Inelastic Supply (Chart 2c). Ifsupply is perfectly inelastic, the samequantity of product must be offeredregardless of the price. (Example: per-ishable strawberries at a farmers’ mar-ket late in the day.) The supply curve isvertical. The price is fixed by demand(what consumers are willing to pay).Any tax imposed will result in a lowernet-of-tax price to the seller, who mustbear the tax. Output is unchanged.(The strawberries are a short-runexample. Repeated inability to sell thefruit will result in less being grownnext season.)

• Perfectly Inelastic Demand (Chart 2d). Ifdemand is completely insensitive toprice, people insist on the same quan-tity of output regardless of what mustbe paid. (Example: addictive drugs.

6


Chart 2c CDA 04-12

Perfectly Inelastic Supply

Quantity

Price

Supply

Demand

P0

Pp1

Tax

Q

Chart 2d CDA 04-12

Perfectly Inelastic Demand

Quantity

Price

Supply

Demand

Pc1

P0

Tax

Q

Addicts in need of a fix will demandthe drugs up to the full amount oftheir resources.) The demand curve isvertical, and any tax will be borne bythe consumers. Output isunchanged. (Of course, this is tonguein cheek. A dealer in illegal drugs isno more likely to collect and remit ahypothetical sales tax than he is toreport his illegal profits to the IRSunder the income tax. Substituting anational sales tax for the income taxwould not eliminate tax evasion inthe underground economy.)

The Perfect Non-Distorting TaxBase? Politicians eagerly seek these lasttwo situations of perfectly inelastic sup-ply and demand in their quest for theperfect tax base. No matter how highthey might push the tax on such a prod-uct, the tax base would not collapse andrevenues would keep climbing. In particular, poli-ticians like to believe that the demand curves forcigarettes, liquor, and gambling are perfectlyinelastic. They are wrong, but they keep pushingtobacco and alcohol tax rates higher, hoping for amiracle. They also get stingy with the payout ratioson state-sponsored lotteries. In this case, it is thosewho buy lottery tickets who are hoping for a mira-cle. In theory, governments could reduce eco-

nomic distortions and minimize dead weightlosses by putting the highest tax rates on the prod-ucts or inputs that are in most inelastic demand orsupply.

The ultimate example of a non-distorting taxwould be a head tax or poll tax that is owed justfor being alive and is totally unrelated to any incre-mental earnings or the amount of one’s economicactivity. Such a tax, however, might not pass the

“equity” test unless it could be shown thatall parties would share in the resultingimprovement in national output andincome.

Economic Burden of an Excise TaxThe ultimate economic burden of an

excise tax would be found by carrying theanalysis one step further. It is not only theconsumers and producers of the taxedproduct who are affected by the tax.Resources driven from the production ofthe taxed items must seek alternativeemployment and will generally earn lowerreturns in these second-best uses. Theywill compete with and affect resources inthese other uses. For example, land takenout of the production of tobacco becauseof higher cigarette taxes may be used toproduce vegetables instead, lowering theprice of vegetables. Both the displacedtobacco farmers and the existing truck

7


farmers who now face added competition areinjured, while consumers of vegetables benefit.

The impact of the tax may shift over time. Anew tax on wine may simply hit the wineries ini-tially, because their vines, fermenting vats, andbottling machinery are still in place and will earnmore being used than being shut down if thereduced after-tax revenues at least cover the laborcosts. Later, however, the vines may be dug up andthe land shifted to other crops that now yield ahigher return. The machinery may wear out andnot be replaced. As supply falls, the excise tax willbe shifted to consumers longer-term. They willhave to pay more for a bottle of wine. They mayswitch some of their spending to other goods andservices, affecting other industries.

Human capital may bear part of the cost. If a taxon wine causes a vineyard to convert to growingtable grapes or avocados, the vineyard workersmay be kept on to tend and pick the new crops; iftheir skills are transferable, they will face littledamage. It would be different for the technicalexperts responsible for the fermenting, testing, andtasting of the wines; they may have no alternativeuse for these highly specialized skills, whichbecome redundant. Such specialists who areforced into other occupations will lose the wagepremium their skills commanded. The caves inwhich the wines were stored, and the slopes withmicroclimates peculiarly suited to wine produc-tion, will lose their advantage and some of the rentthey commanded in wine production.

The need to consider these economy-wide andlong-term ramifications, called “general equilib-rium” analysis, is not a new idea in tax theory.Alfred Marshall’s classic discussion of the inci-dence of taxation in his Principles of Economics is asvalid today as it was roughly a hundred years ago.Taxes on inputs are borne largely by the suppliersof the inputs if those inputs have no good alterna-tive uses (inelastic supply), but are borne largelyby the consumers of the product if the inputs arereadily shifted to other uses (elastic supply). A newtax imposed on existing capital will be borne bythe capital in the short run but may discouragerenewal of the capital stock as it wears out, causingthe tax to be shifted to the consumers in the longrun (and to any other immobile inputs that wouldhave worked with the lost capital). A nationwidetax may impact producers and consumers of theproduct, but a local tax will simply drive the pro-

ducers to move their inputs to another part of thecountry. In Marshall’s words:

It is a general principle that if a tax impingeson anything used by one set of persons inthe production of goods or services to bedisposed of to other persons, the tax tendsto check production. This tends to shift alarge part of the burden of the tax forwardson to consumers, and a small partbackwards on to those who supply therequirements of this set of producers.Similarly, a tax on the consumption ofanything is shifted in a greater or less degreebackwards on to its producer.

For instance, an unexpected and heavy taxupon printing would strike hard upon thoseengaged in the trade, for if they attemptedto raise prices much, demand would fall offquickly: but the blow would bear unevenlyon various classes engaged in the trade.Since printing machines and compositorscannot easily find employment out of thetrade, the prices of printing machines andwages of compositors would be kept low forsome time. On the other hand, thebuildings and steam engines, the porters,engineers, and clerks would not wait fortheir numbers to be adjusted by the slowprocess of natural decay to the diminisheddemand; some of them would be quickly atwork in other trades, and very little of theburden would stay long on those of themwho remained in the trade. A considerablepart of the burden, again, would fall onsubsidiary industries, such as those engagedin making paper and type; because themarket for their products would becurtailed…. Authors and publishers [and]booksellers…would suffer a little….

[I]f the tax were only local, thecompositors would migrate beyond itsreach; and the owners of printing housesmight bear a larger…proportionate shareof the burden than those whose resourceswere more mobile….

Next, suppose the tax to be levied onprinting presses instead of on printedmatter. In that case, if the printers had nosemi-obsolete presses which they wereinclined to destroy or to leave idle, the tax

8


would not strike at marginal production: itwould not immediately affect the output ofprinting, nor therefore its price. It wouldmerely intercept some of the earnings of thepresses on the way to the owners, and lowerthe quasi-rents of the presses. But it wouldnot affect the rate of net profits which wasneeded to induce people to invest fluidcapital in presses: and therefore, as the oldpresses wore out, the tax would add tomarginal expenses…. [T]he supply ofprinting would be curtailed; its price wouldrise: and new presses would be introducedonly up to the margin at which they wouldbe able…to pay the tax and yet yield normalprofits on the outlay. When this stage hadbeen reached the distribution of the burdenof a tax upon presses would henceforth benearly the same as that of a tax uponprinting….4

Burden Tables Botch Excise Tax “Incidence” and “Burden”

Burden tables use the least meaningful of all theabove concepts of incidence and burden to allocatethe impact of excise taxes. Burden tables assumethat all excise taxes, whether labeled consumers’ ormanufacturers’ excise taxes, are paid entirely by theconsumers of the products (as under the statutoryobligation concept of a consumer-level tax). The“distribution” of the tax across income levels is cal-culated by taking the average amount spent on theproduct by people in various adjusted gross incomeclasses times the tax rate. The tables ignore the splitbetween producers and consumers that must occurin any market with normal elasticities. Further-more, they look only at the revenues collected, t xQ1, and ignore the deadweight loss, so that, evenignoring the split, they do not measure the total ini-tial incidence correctly.

An excise tax analyst at the JCT or Treasury willuse the long-run elasticities of demand and supplyfor the taxed good to estimate the eventual changein consumption (the drop from Q0 to Q1) and willestimate the tax revenue that the Treasury willreceive at the new, reduced level of consumption. Inconstructing a burden table, he will attribute all ofthe incidence of the tax to the consumers. However,

the analyst will assume no loss in total output orefficiency for the economy as a whole, and no lossof revenue from other taxes, because he assumesthat resources driven out of producing the taxedgood find alternative employment at virtuallyunchanged earnings. He ignores any shifting of theeconomic burden to producers as resources areshifted to alternative, lower-paid uses. Burden tableanalysis thus gets both the total and the distributionof excise taxes wrong except in the extreme case of aproduct in absolutely inelastic demand.

IV. EXTENDING THE ANALYSIS: INCOME AND PAYROLL TAXES ON CAPITAL AND LABOR

The same sort of diagram may be applied to anytax. The tax may be a general sales tax, or a payrollor personal income tax on wages or on capitalincome, or the corporate income tax. In the case ofa tax on labor income, the price becomes thewage, and the quantity becomes hours worked orthe level of employment or some other measure ofthe services of labor. In the case of capital services,the price becomes the rate of return on capital,and the quantity is the amount of capital servicesforthcoming from the stock of plant, equipment,structures, and land.

The demand for labor and capital reflects thevalue to the employer of using additional units oflabor and capital. The added output obtained byemploying one more worker or machine is the“marginal product of labor” or “marginal productof capital.” The added physical output times theprice it sells for (marginal value product) is themost that a firm will pay to hire an additionalworker or pay for the services of an additionalmachine or building.

As more of any one factor is added, other factorsheld constant, output rises, but at a diminishingrate. This is the famous “law of diminishingreturns.” The gradual decline in the marginalproducts of labor or capital as more of one of themis employed is why the demand curves for the fac-tors slope downward.

Charts 3 and 4 illustrate the supply and demandconditions generally assumed for broadly definedlabor and capital inputs, respectively, and the dif-

4. Alfred Marshall, Principles of Economics, 8th edition (1920) (Philadelphia, Pa.: Porcupine Press, reprinted 1982), Chap-ter IX, pp. 343–345. The first edition was printed in 1890. Tax incidence and tax shifting are not new notions.

9


Chart 3 CDA 04-12

Hours Worked

Wage

Effect of Tax On Labor

Labor Supply

Net Wage

GrossWage Marginal Product

of Labor (Demand)

Tax

Dropin

Labor

L' L

MPL would rise if there were more

capital to work with, and fall if

capital formation

lagged.

ferent effects one might expect from taxingthese factors.

Labor MarketThe Supply of Labor. The supply of

labor is rather inelastic. It was fashionablein the 1950s and 1960s to assert that thesupply of labor was nearly perfectly inelas-tic with respect to the wage (or after-taxwage). That is, workers did not vary theirlabor supply very much in response tochanges in the after-tax wage. The think-ing was that adult males were the bulk ofthe workforce, and, as their families’ solebreadwinners, they were very attached tothe workforce. Furthermore, they weregenerally employees of corporations orother businesses that set their hours, giv-ing them virtually no option but to work a40-hour week unless there was overtime,which was typically mandatory, or theywere willing to take on second jobs. Withlimited ability to vary their hours worked or par-ticipation in the workforce, such workers wereassumed to bear any taxes imposed on labor,including the income tax and the entire payrolltax, both the employee and employer shares. Thisis the convention still used in burden tables.

Over time, most married women and many teen-agers have entered the workforce, and a growingnumber of “retirees” hold part-time jobs. Many ofthese workers are less tightly “attached” to the work-force than prime-age males. Since the 1980s and1990s, a larger portion of the workforce has becomeself-employed or is seeking to work part-time. Theseworkers have far more flexibility to set their ownhours and display a less rigid attachment to theworkforce than adult males. Also, as two-earner cou-ples have become the norm, men have had moreopportunity to work less, courtesy of their wives’incomes. Although the men may have worked lessas family income rose, the couple may have workedmore, taking both spouses’ efforts together.

One should expect higher elasticities for upper-income workers, whose income and wealth give

them added flexibility to alter their hours whilemaintaining a high living standard. Modern consen-sus estimates of labor force elasticity, while still low,are generally non-zero. For example, a survey of 65labor economists produced estimates of the laborsupply elasticity for men of 0.1 (mean estimate) andzero (median estimate). For women, the surveygave estimates of 0.45 (mean) and 0.3 (median).5

The Demand for Labor. The demand for laboris moderately elastic. Its large share of the nationalincome makes it a major expense for employers,and the marginal product of labor declines onlygradually as the workforce increases. To someextent, capital can be substituted for labor if laborcosts rise. There is also the possibility of shiftinglabor-intensive production abroad to take advantageof lower labor costs if the foreign labor is sufficientlyproductive to make a difference in unit labor costs.

Capital MarketThe Supply of Capital. The supply of capital is

highly elastic. Physical capital (equipment plusindustrial, commercial, and residential structures)can be easily reproduced or expanded (given a bit

5. Victor R. Fuchs, Alan B. Krueger, and James M. Poterba, “Economists’ Views About Parameters, Values, and Policies: Survey Results in Labor and Public Economics,” Journal of Economic Literature, Vol. 36 (September 1998). Some writ-ers believe that the empirical evidence points to a labor supply elasticity of zero or less, which could lead to more work effort at higher tax rates and “reverse” tax shifting. For a number of reasons, that outcome is highly unlikely. For a more sympathetic view, see Jane Gravelle, “Labor Supply Elasticity and Dynamic Scoring,” Congressional Research Service Memorandum, August 21, 2002.

10


Chart 4 CDA 04-12

Desired Amount of Capital

Return to Capital

Effect of Tax On Desired Capital Stock

Net Return

Gross Return

Required Return to Capital (Supply)

Tax

Drop in Capital

K' K

Marginal Product of Capital (Demand)

of time). Furthermore, investors seemwilling to construct and employ addi-tional plant, equipment, and buildingswhenever the after-tax risk-adjusted rateof return approaches about 3 percent(again, given a bit of time).6 Put anotherway, savers will readily finance (buyclaims to the earnings of) capital assets atabout a 3 percent after-tax risk-adjustedrate of return, substituting additional sav-ing for additional consumption.

Thus, the supply of investment goodsand the supply of saving to pay for it areboth fairly elastic over time. Conversely,when rates of return on physical capitalfall below that level, old assets are notreplaced when they wear out. Investorsand savers use a bit more of their incomefor consumption instead, which is, at themargin, virtually as attractive as the fore-gone investment.

The Demand for Capital. The demand for cap-ital is fairly elastic because the marginal product ofcapital declines only gradually as the stockincreases. Years of real-world observations suggestthat it takes a significant rise in the quantity ofcapital and the capital-to-labor ratio to depressreturns and discourage further investment.

Incidence of Taxes on Labor and CapitalIncidence of Labor Taxes. The relatively elastic

demand for labor, coupled with the assumption ofa highly inelastic supply of labor, means that laborbears most of the initial economic incidence oftaxes on labor income. It has become common toassert that all taxes on labor income fall on theworker, including the employers’ share of the pay-roll tax, the employees’ share of the payroll tax, theunemployment compensation tax, and the portionof the income tax that falls on wages and salaries.

However, the modern workforce is seen to dis-play some elasticity of supply; and to that extent, itmust be assumed that workers will respond tohigher tax rates by taking more leisure, and thequantity of labor supplied would fall. A reducedworkforce would lower the productivity of thecapital stock, suggesting that some of the ultimate

burden of a tax on labor would fall on capital own-ers. (Just as the productivity of a given number ofworkers is enhanced if they have more capital towork with, the productivity of a given amount ofcapital is enhanced if there are more workers, par-ticularly more skilled workers, to utilize it. Con-versely, if fewer skilled workers were available, theproductivity of capital would decline. Think ofwhat would happen to the earnings of the fifthtruck at a small trucking company if one of the fivetruck drivers called in sick.) However, the capitalstock may contract in response to a drop in its pro-ductivity and rate of return in order to restore itsformer rate of earnings (see below), which wouldshift the burden back onto the work force.

Incidence of Taxes on Capital Income. Theincidence of a tax on capital income dependsgreatly on the time frame. Physical capital cannotdisappear overnight (in the event of a taxincrease), and it takes time to add to the stock ofplant, equipment, and buildings (in the event of atax reduction). Immediately after a tax increase isimposed on businesses or savers, their after-taxreturns on old assets would be depressed. Finan-cial market adjustments would come swiftly. Bondand stock prices would fall, restoring after-tax

6. Gary Robbins and Aldona Robbins, “Capital Taxes and Growth,” National Center for Policy Analysis, Policy Report No. 169, January 1992, and Gary Robbins and Aldona Robbins, “Eating Out Our Substance (II): How Taxation Affects Investment,” Institute for Policy Innovation, TaxAction Analysis Policy Report No. 134, November 1995.

11


returns for new buyers and forcing new borrowersto offer higher interest rates and rates of return tonew investors.

Over time, investors in physical capital canadapt. The high long-run elasticity of supply ofcapital suggests that a tax imposed on capital willreduce the capital stock until the gross return risesto cover the tax, leaving the after-tax return aboutwhere it was before the tax was imposed. Becauseof the high elasticities of supply and demand forcapital, the reduction in the capital stock may haveto be substantial to increase its return by enoughto cover the tax. As a result, taxes on the earningsof capital assets or on saving may result in sharpreductions in the stock of capital available for pro-duction. Downward adjustments in the physicalcapital stock may take time because capital takessome years to wear out. Eventually, the reductionin the capital stock (or slower than normalgrowth) will bring it back into balance with thegrowth in demand for capital associated with pop-ulation growth.

Adjustment to an adverse shock may take afew years for equipment, a decade or two forstructures. (For example, in the 1988–1990period, Japan instituted an “anti-tax reform” thatsharply raised taxes on capital income, includinginterest and capital gains from stocks, andincreased taxes on buildings and land. The resultwas a particularly severe economic shock that notonly affected the returns to physical capital but

threw much of the Japanese financial sector intochaos as stock and land prices plunged. It hastaken nearly 15 years to sort out the mess. Mostshocks are not that severe, and most adjustmentperiods are not that long.) Positive shocks may beeasier to deal with. New equipment can beordered and placed in service in a few months,new housing constructed within a few quarters,and new commercial or office buildings put upwithin two or three years.

Implications of Incidence for the Tax BaseThe differences in the elasticities of supply and

demand for labor and capital suggest that a taximposed evenly on labor and capital income willreduce the stock of capital by more than the quan-tity of labor supplied. (Compare Charts 3 and 4.)Such a tax is more distorting of economic behaviorthan a tax imposed chiefly on labor income.

This suggests an economic advantage frommoving away from the so-called broad-basedincome tax, which actually taxes income used forsaving and capital formation more heavily thanincome used for consumption, to various taxesthat are saving-consumption–neutral.7 Such neu-tral taxes are often labeled as consumption-basedor consumed-income–based and are often, some-what erroneously, described as taxing labor andexempting returns on capital income. These taxesdo, in fact, tax quasi-rents and other abnormalreturns to capital that exceed the cost of the savingrequired to obtain the assets.

7. Federal and state revenue systems tax income that is saved more heavily than income that is used for consumption. At the federal level, there are at least four layers of possible tax on income that is saved. (1) Income is taxed when first earned (the initial layer of tax). If one uses the after-tax income to buy food, clothing, or a television, one can generally eat, stay warm, and enjoy the entertainment with no additional federal tax (except for a few federal excise taxes). (2) But if one buys a bond or stock or invests in a small business with that after-tax income, there is another layer of personal income tax on the stream of interest, dividends, profits, or capital gains received on the saving (which is a tax on the “enjoyment” that one “buys” when one saves). The added layer of tax on these purchased income streams is the basic income tax bias against saving. (3) If the saving is in corporate stock, there is also the corporate tax to be paid before any distribution to the shareholder or any reinvestment of retained after-tax earnings to increase the value of the business. (Whether the after-tax corporate income is paid as a dividend, or reinvested to raise the value of the business and create a capital gain, corporate income is taxed twice—the double taxation of corporate income.) (4) If a modest amount is left at death (beyond an exempt amount that is barely enough to keep a couple in an assisted living facility for a decade), it is taxed again by the estate and gift tax (the “death tax”). Eliminating the estate and gift tax and the corporate tax would remove two layers of bias. Granting all saving the same treatment as is given to pensions or IRAs, either by deferring tax on saving until the money is withdrawn for consumption (as in a regular IRA) or by taxing income before it is saved and not taxing the subsequent returns (as in a Roth IRA), would remove the basic bias. Saving-deferred taxes, the Flat Tax, VATs, and retail sales taxes are examples of saving-consumption–neutral taxes. For a further explanation of the biases against saving in the current income tax, see Stephen J. Entin, “Fixing the Saving Problem: How the Tax System Depresses Saving and What to Do About It,” Institute for Research on the Economics of Taxation, IRET Policy Bulletin No. 85, August 6, 2001, pp. 15 ff, available at www.iret.org. Also see David F. Bradford and the U.S. Treasury Tax Policy Staff, Blueprints for Basic Tax Reform, 2nd edition, revised (Arlington, Va.: Tax Analysts, 1985).

12


One argument against major reform of the taxsystem (moving to a saving-consumption–neutraltax) is that, if labor is truly in highly inelastic supply,sweeping tax rate reductions would do little toboost labor force participation and hours workedand would have only limited economic benefits.Advocates of the tax status quo, or of higher taxrates on upper-income workers, should be carefulin making such arguments. A highly inelastic sup-ply of labor would also mean that there is a rela-tively small reduction in employment from taxes onlabor income at all levels, which would make suchtaxes relatively non-distorting of economic activity.

In theory, for those public finance graduateswho put great stock on avoiding “economic distor-tion” and maximizing “economic efficiency,” thisshould make labor income the ideal tax base. Onesuspects, however, that people who oppose funda-mental tax reform proposals on the grounds thatthey may appear superficially to be regressive andshift the tax burden from capital income to laborincome would not favor heavy taxes on laborincome as an alternative.

The Ultimate Burden: Further Tax Shifting in a General Equilibrium Framework

Labor and Capital: Complements More thanSubstitutes. Output and incomes are at theirhighest when optimal amounts of labor and capitalwork together to create the goods and services onwhich consumers place the greatest value.Depending on the production process, there maybe some room to substitute labor for capital (orvice versa) or to substitute skilled labor forunskilled labor.

For the economy as a whole, however, and inmost situations, the various skills and talents ofthe workforce, managers, and entrepreneurs andthe services of various types of capital are comple-ments in production, not substitutes. That is, themore there is of any one type of factor, the higherwill be the productivity and incomes of the otherfactors that work with it and gain from its pres-ence. If there is more capital for labor to workwith, wages rise. If an increase in the skilled workforce makes capital more productive, the returnson capital go up.

Taxes Matter “at the Margin.” Taxes affect thewillingness of labor and capital to participate inproduction; or, put another way, taxes affect thecost of labor and capital services, and therefore the

cost of production. Supply decisions are not usu-ally all or nothing. One chooses to work a little bitmore or less, or to save a little more or less, or toemploy a slightly higher or lower number ofmachines, or slightly more or less powerful ormodern ones, on the factory floor. The tax ratesthat affect such decisions are the marginal tax ratesthat apply to the last or next dollar to be earnedfrom small reductions or increases in one’s eco-nomic activity. Taxes that fall at the margin onincremental activity reduce the quantity ofresources available for production. With fewerinputs, there will be less output and income,according to the characteristics of the productionprocess.

Lump-sum taxes, such as a head tax, involve afixed dollar amount owed regardless of income,and so have no impact on decisions about increas-ing one’s earnings. Likewise, one-time retroactivetax hits do not apply to future income, althoughthey may make taxpayers suspicious that they willbe repeated. Such taxes are not “at the margin,”meaning that they do not affect the last or nextdollar earned, and are the only kind of tax thatdoes not reduce incentives and curtail activity.Similarly, rebates of taxes on income of past years,such as President Gerald Ford’s 1975 tax rebate on1974 income tax liability, give no incentive toincrease output in the future.

Taxing One Factor Hurts the Other. If a taxfalls “at the margin,” it depresses the reward to thetaxed factor of production, and less of that factor’sservices will be offered and employed. Because thereis less of that input, all the other factors that workwith it suffer a loss of productivity and income.They, too, bear some of the burden of the tax. Forexample, a tax that reduces the quantity of capitallowers the wages of labor. Labor thus bears much ofthe burden of the tax on capital. (See Chart 5.)

Taxing Capital Hurts Labor a Lot. Insofar assome inputs are more affected by the taxes thanothers, they may withdraw their services to agreater or lesser extent than others do. As someinputs withdraw heavily from the market, theirrelative scarcity affects the productivity, employ-ment, and income of other productive inputs withwhich they would normally work. Because capitalis more sensitive to taxation than labor, a tax oncapital will have a relatively large adverse impacton the quantity of capital, which will then cause arelatively large drop in the marginal product and

13


compensation of labor. Taxes on labor hurtcapital as well, but because labor is lesselastic in supply and withdraws less fromthe market, the effect is less pronounced.

Consider a small trucking companywith five vehicles. Suppose that the rulesfor depreciating trucks for tax purposeschange, with the government demandingthat the trucks be written off over fiveyears instead of three. The owner has hadenough business to run four trucks flat outand a fifth part-time. He is barely breakingeven on the fifth truck under old law. It isnow time to replace one of the trucks.Under the new tax regime, it does notquite pay to maintain the fifth truck. Theowner decides not to replace it, and hisincome is only slightly affected. But whathappens to the wages of the fifth truckdriver? If he is laid off, who bears the bur-den of the tax increase on the capital?

Consider another example, involving humancapital—specifically, medical training. Suppose theimposition of a progressive income tax were to dis-courage the supply of physicians by inducingsome doctors to retire, by causing others to workfewer weeks per year, and by dissuading peoplefrom applying to medical school. One resultwould be fewer jobs available and lower levels ofproductivity and incomes for nurses and supportstaff in medical offices and hospitals. Anotherwould be a rise in the price of health care for con-sumers (including the government).

For example, assume that four doctors havebeen operating separate practices in a large town.Each has been taking off one month a year, duringwhich time the other three cover for him. Follow-ing the tax hike, they decide to merge their offices,with each doctor taking off three months a yearand with a fourth of the support personnel let goas redundant. The doctors prefer an extra twomonths of leisure to the lowered after-tax cashearnings they would have earned by working theirregular work-year. The laid off support staff haveexperienced a less voluntary and potentially more

damaging shock. Doctors’ rates in the region risemarginally. Patients experience longer waits ormust seek out doctors in the next town. Whobears the brunt of the tax on the doctors’ incomes?

Such effects may seem small or unlikely at cur-rent tax rates, but they are certainly pronouncedwhen tax rates are very high. Historical examplesabound. The 1954 tax overhaul in the UnitedStates did little to reduce the top World War II taxrates. The top rate went from 92 percent to 91 per-cent, where it remained until the Kennedy tax ratecuts, which lowered the top marginal rate in stagesto 70 percent in 1964 and 1965. President RonaldReagan often remarked that at such extreme taxrates, it did not pay him to make more that one ortwo movies a year. There were obvious adverseeffects on the U.S. labor markets from the infla-tion-induced “bracket creep” of the 1970s, whichpushed marginal tax rates higher across the board.The top tax rate in Britain before MargaretThatcher’s reforms in 1979 was 98 percent. Theinfamous British “brain drain” was one result.8

In short, taxes on capital reduce the wages oflabor; taxes on labor reduce the rates of return on

8. Another result was conspicuous consumption. That is, saving was affected as well. At the 20 percent interest rates then prevailing in Britain (reflecting high tax rates and high inflation), one could invest £50,000 in a government note, earn £10,000 in interest, pay £9,800 in tax, and have £200 a year left over. Alternatively, one could give up the bond and the interest to buy a Rolls Royce for £50,000 and enjoy the car. Was driving a Rolls Royce worth £200 a year? Many people thought so.

Chart 5 CDA 04-12

Employment

Wage

Labor Supply

MPL (K0)

W

0

N1

MPL (K1)

W1

N0

A Smaller Stock of Capital Reduces Wages

14


capital (at least in the short run, until the capitalstock shrinks); taxes on certain types of laborreduce the wages of other types of labor; taxes oncertain types of capital reduce the returns on othertypes of capital. The repercussions of a tax on onefactor of production on the income of other factors,or of a tax on one sector of the economy on othersectors, are “general equilibrium” effects. Theyoccur outside of the immediate market for the factoror product being taxed and represent impacts thatgo beyond the initial economic incidence of the tax.Such effects are part of the ultimate economic bur-den of the tax and represent some of the shifting ofthe tax burden from the taxed factors or products toother factors and sectors.

Implications of Burden Shifting for the Tax BaseEven for Labor, the Optimal Tax on Capital

Is Zero. Several studies in the economic literatureillustrate that a zero tax rate on capital incomewould raise the after-tax income of labor, inpresent-value terms, even if labor must pick up thetab for the lost tax revenue. That is, a tax on capitalis effectively shifted to labor, which pays morethan the full value of the tax.

In a 1974 paper,9 Martin Feldstein explored theconsequences of a variable capital stock for thedistribution of the tax burden. Previous studiesthat generally assumed no change in the capitalstock had concluded that the burden or benefit ofa tax increase or decrease on capital was borne bycapital. (See the discussion of the corporateincome tax, below.) Feldstein showed the impor-tance of allowing for the capital stock to vary.

Feldstein assumed the tax on capital incomewas eliminated and that on labor was increased ina revenue-neutral manner. He then looked at theleast favorable case for labor, in which people wereeither savers who had no wage income or workerswho did no saving. In a “statutory obligation” orburden table or static sense, the savers would

enjoy all of the benefit from the initial tax cut oncapital income. All workers would face an initialtax increase on wages equal to the dollar amountof the tax cut on capital.

However, Feldstein argued, cutting the tax onthe savers would enable them to save more, at thegiven propensity to save that they display, by leav-ing them more after-tax income. The added savingwould cause the capital stock to rise to a new equi-librium level at which the added saving was justsufficient to cover the added depreciation so as tomaintain the incremental stock.

At the higher capital-to-labor ratio, the produc-tivity of labor and the wage would both be higher(Chart 5 in reverse), leaving the workers withhigher gross wages and more after-tax income inthe steady state despite the higher tax rate onwages. Feldstein showed that, under plausibleassumptions, the present value of the increase infuture after-tax wages due to the rise in grosswages would be greater than the near-term reduc-tion in after-tax wages due to the rise in the taxrate on wages. Workers would be better off inpresent-value terms with no taxation of capital.

A 1986 study by Christophe Chamley showedthat the optimal tax rate on capital is zero in thelong run under a narrow set of assumptions,including a fixed growth rate not affected by taxes,a closed economy, and identical consumers livinginfinite lives.10 Many other studies on the shiftingof taxes on capital to labor have expanded on thiswork by easing a number of Feldstein’s and Cham-ley’s restrictions and using different types of mod-els, showing it to be a more general proposition.11

For example, a 1999 study by Andrew Atkeson, V.V. Chari, and Patrick J. Kehoe demonstrated thatChamley’s result holds under greatly relaxedassumptions, including heterogeneous consumersin overlapping generations, an open economy, anda growth rate that is affected by taxes.12

9. Martin Feldstein, “Incidence of a Capital Income Tax in a Growing Economy with Variable Savings Rates,” Review of Economic Studies, Vol. 41, No. 4 (1974), pp. 505–513.

10. Christophe Chamley, “Optimal Taxation of Capital Income in General Equilibrium with Infinite Lives,” Econometrica, Vol. 54 (May 1986), pp. 607–622.

11. Kenneth L. Judd, “Redistributive Taxation in a Simple Perfect Foresight Model,” Journal of Public Economics, Vol. 28 (October 1985), pp. 59–83. Also see Kenneth L. Judd, “A Dynamic Theory of Factor Taxation,” American Economic Review, Vol. 77 (May 1987), pp. 42–48; N. Gregory Mankiw, “The Savers-Spenders Theory of Fiscal Policy,” American Economic Review, Vol. 90, No. 2 (2000), pp. 120–125; and Casey B. Mulligan, “Capital Tax Incidence: First Impres-sions from the Time Series,” National Bureau of Economic Research, Working Paper No. 9374, December 2002.

15


Speed of Adjustment Is Critical. The resultsin many of these studies are sensitive to the speedof adjustment of the capital stock. In a 1979paper,13 Professor Robin Boadway questioned theconclusion that labor would gain in present valueby eliminating the tax on capital. He suggestedthat a low elasticity of saving could slow the rise inthe capital stock and delay the expected rise inafter-tax incomes. If the added capital formationtook long enough, the higher tax rate on labor inthe not-so-short short run would then outweigh,in present value, the rise in after-tax incomes inthe long run, and workers would be worse off.Similarly, a rise in the tax on capital and a reduc-tion in the tax on labor might make labor better offfor many years before the reduction in the capitalstock lowered workers’ before- and after-tax wagesby enough to make them worse off in presentvalue. Boadway suggested that labor might gainfrom a tax on capital for as long as 65 years beforethe steady state was reached.

Many of these presentations involve stylizedmodels of a highly simplified economy or popula-tion. They achieve the change in national savingand the capital stock solely on the basis ofmechanically moving disposable income fromthose who do not save to those who do, at con-stant propensities to save (fixed rates of saving outof labor and capital income), and let the change insaving, which is only a fraction of the shiftedincome in this approach, determine the change inthe capital stock. By contrast, in the real world, atax change affects the cost of capital and thereturns to saving, which in turn alter the desiredcapital stock and level of saving. These changes insaving and the capital stock can be much largerthan the dollar amounts of the tax change.

N. Gregory Mankiw has illustrated this mechan-ical type of model in a paper aptly titled “The Sav-ers–Spenders Theory of Fiscal Policy.”14 Suchmodels generally assume a closed economy (notopen to trade and international capital flows), lim-iting the supply of saving available to boostdomestic investment. Most assume their elastici-ties without deriving them from a general equilib-rium model tested against actual experience.Hence, they cannot be considered robust picturesof the real world. These studies, of which theBoadway study is a good example, produce undulypessimistic estimates of the length of time it takesto increase the capital stock following a reductionin the tax rate and of the amount by which thecapital stock would rise.

Reality Check. Traditional economists are usedto thinking in terms of a fairly constant “propen-sity to save” and an inelastic supply of saving.They may be skeptical that the quantity of domes-tic saving can increase by enough to allow for astrong burst of capital formation needed to bringabout a rapid adjustment of the capital stock to atax shock. Their focus on the channels by whichthe needed investment is financed is misplaced.They should look first at the speed of adjustmentin the historical record of the real world and thenworry about how it happens rather than declaringan observed phenomenon to be impossible.15

How rapidly the economy will invest or disin-vest to reach the new equilibrium level of capitaldepends on several factors, such as the elasticity ofsaving with respect to the rate of return, the easewith which existing saving flows can be redirectedacross national borders, the elasticity of the globalsupply of investment goods and their resultingcost, and the rate at which existing capital wears

12. Andrew Atkeson, V. V. Chari, and Patrick J. Kehoe, “Taxing Capital Income: A Bad Idea,” Federal Reserve Bank of Min-neapolis Quarterly Review, Vol. 23, No. 3 (Summer 1999), pp. 3–17.

13. Robin Boadway, “Long Run Tax Incidence: A Comparative Dynamic Approach,” Review of Economic Studies, Vol. 46, No. 3 (July 1979), pp. 505–511.

14. Mankiw, “The Savers-Spenders Theory of Fiscal Policy,” p.120.

15. Before van Leeuwenhoek invented the microscope, physicians knew that arteries carried blood from the heart and veins returned it, but they had no way to see the capillaries that connected the arteries to the veins. They were unable to map the full circulatory system, and many people were skeptical of the concept of a circular flow of blood. It would have been logical to assume that it was a single system in flow equilibrium, but that concept had not been invented yet. Today, many economists doubt the country’s ability to finance federal deficits and the investment that is increasing the stock of capital, and to balance saving and investment, because they cannot see where the financing is to come from. They will never be able precisely to predict or trace the flow of trillions of dollars of funds throughout the com-plex world financial system, but the funds do flow nonetheless.

16


Chart 6 CDA 04-12

1

2

3

4

5

6%

1956-I 1960-I 1964-I 1968-I 1972-I 1976-I 1980-I 1984-I 1988-I 1992-I 1996-I

Four Quarter Moving Average

Real After-Tax Return

Economy-Wide Real After-Tax Return to Business Capital

out (in the case of disinvestment).Although these sources of financingand the production streams of physicalcapital are flows, they are part of acomplex stock adjustment process.

One could try to imagine or tomeasure separately how flexible theseflows may be. Alternatively, onecould review the changes in the capi-tal stock that have occurred in thepast following shocks to the after-taxrate of return. The latter approachgives an important reality check. Ifadjustment of the capital stock hasproceeded more rapidly in the pastthan can be accounted for by theflows of saving and investment pre-dicted by some current models, thenthere may be additional or deeperchannels for capital flows in the realworld that are not recognized by themodels. “It’s fine in practice, but it will neverwork in theory!” is an indictment of the theory,not of the real phenomenon.

Rapid Adjustment of Capital Is the Norm.How fast the capital stock adjusts, which is to sayhow quickly the return on capital is restored tonormal levels after a shock, is really an empiricalquestion, not a theoretical one. Many events, suchas technological change, a shift in tax policy, or ashift in inflation, can change the expected returnson capital investment or alter the user cost of capi-tal. The result will be a shift in the desired stock ofcapital, toward which the economy will move overa number of years.

Are changes in the rate of return to capitalmerely consequences of business cycles, or arethey independent factors that drive savers andinvestors to adjust the size of the capital stock toconform to new economic conditions, causingchanges in the rate of investment that generatebusiness cycles? Gary Robbins of Fiscal Associatesand the Heritage Foundation Center for Data Anal-ysis has plotted after-tax rates of return to businesscapital over time. He finds that the movements inthe return to capital, in the desired capital stock,and in the resulting swings in investment activityare seen to lead the business cycle up and down.They are therefore most likely to be a cause, not aresult, of the business cycle. (See Chart 6.)

Robbins also finds that the rates of return havetended strongly to remain in the neighborhood of 3percent. Between 1956 and 2000, the four-quartermoving average rate averaged 2.76 percent and waswithin half a percentage point of this average 60percent of the time. Not only do the returns on cap-ital remain within a fairly narrow band over time,but they tend to revert to the band fairly quickly.This implies that, each time there was a majorshock to the rate of return, whether traceable to tax,inflation, or technological changes, the quantity ofcapital has adjusted rapidly and the rate of returnwas restored soon to its long-run average.16

Robbins has tested the speed of adjustment byrunning regressions looking at implied desiredstocks versus the actual deliveries of capital usingvarious distributed lags. He finds that roughly half ofthe investment in equipment and structures neededto move to the new desired capital stock will occurin the first three years following the shock and thatnearly all of the adjustment is completed within fiveto 10 years (with structures taking a bit longer thanequipment). If the bulk of the increase in the capitalstock occurs in the first decade following the taxchange, as Robbins has found by looking at histori-cal experience, then the case for eliminating the taxon capital is quite strong.

An Open Economy and Flexibility of SavingSpeed the Adjustment of Capital. The observedstability in the real after-tax rate of return in the

17


United States and the speed of adjustment of thecapital stock to shocks make sense because, in aglobal economy, the risk-adjusted rate of return inany sub-region should be kept in rough alignmentwith global returns. Put another way, the size ofthe capital stock in any one country is sensitivenot merely to the innate desired rate of return thathumans display (the “marginal rate of time prefer-ence”), but also to its relative rate of return com-pared to that available on capital abroad. Theelasticity of the capital stock in a region is muchhigher than for the world as a whole.

In a closed economy, net national saving (net ofgovernment dissaving) equals private investment,and the speed of adjustment to a new desiredequilibrium capital stock following a shock is lim-ited by the change in the national saving rate. Inthe case of a tax change in the closed economy, thechange in national saving and investment willdepend on the immediate effect of the tax changeon the government deficit (which is the only effectconsidered in fixed-GDP “static” analysis used bygovernment officials) and on the subsequentdynamic effects of the tax change on the nation’sown domestic private saving, investment, andincome, which in turn depends on the elasticity ofdomestic saving and investment with respect tothe after-tax rate of return. However, the limitationimposed by the flexibility of own-country savingdoes not hold in an integrated world economywith international capital flows.

In today’s world, it would be a great mistake toassert that the progress of any one nation toward anew equilibrium capital stock following a tax ortechnological change is limited by its own savingelasticity or by the static tax-induced change in itsown national saving rate. Changes in the flow ofcapital across national borders can have a major

impact on the speed of adjustment. For example,following the major tax and monetary policychanges of the early 1980s, new U.S. bank lendingabroad dropped from roughly $120 billion in 1982to under $20 billion in 1984. The drop in U.S. capi-tal outflow of $100 billion more than covered the1982–1984 change in the government deficit fol-lowing the 1980 and 1981–1982 recessions and the1981, 1982, and 1984 tax changes. The shift todomestic lending was large enough to finance a largeportion of the increase in private investment in thefirst half of the decade. In addition, the private sav-ing rate increased. There was only a modest rise inforeign capital flows to the United States in thatperiod. (They rose further later in the decade).

Longer time horizons reinforce the importanceof international capital flows and of how a nationtreats foreign investment. From the first Spanishand English settlements in Florida (St. Augustine,1565) and Virginia (Jamestown, 1607) until WorldWar I, a period of over 300 years, the region thatbecame the United States experienced a massiveinflow of population and capital from Europe,Africa, and Asia. The capital inflow allowed thecountry to run current account deficits for most ofthat period. (There was a brief period of currentaccount surplus for about a dozen years after theCivil War, when the U.S. was deflating andimporting gold to restore the dollar to the goldstandard at the pre-war parity. Being money, thegold inflow was not considered an import. If goldwere treated as a commodity, even these surplusesmight have been deficits.) Much of the investmentin the early U.S. canals, railroads, and industrywas financed by foreigners. International capitalflows are not a new phenomenon.

Neither is awareness of the implications of anopen economy for the stock of capital, the wages

16. Unpublished preliminary figures for a forthcoming study from The Heritage Foundation. See Gary and Aldona Rob-bins’s earlier work for the Institute for Policy Innovation, “Eating Out Our Substance (II): How Taxation Affects Invest-ment,” TaxAction Analysis Policy Report No. 134, November 1995, available at www.ipi.org. In the IPI study, using earlier Commerce Department data that have since been revised for the period 1954–1994, the authors found that “the rate of return to new investment, after taxes, depreciation, and inflation, has been remarkably stable over the last forty years. The reason is that investors quickly counter shocks that cause their after-tax return to go up or down by changing their investment behavior. In short, increases in the after-tax return have led to an increase in the rate of cap-ital formation until the return was driven back down to its long-run, economy-wide average of 3.4 percent [old data]. Conversely, decreases in the after-tax return have been followed by a decrease in investment until the after-tax return went back to 3.4 percent. And the adjustment generally takes five years or less. A major source of ‘shock’ is changes in tax policy.” The revisions appear to have affected the level of the rate of return, but not the pattern of year-to-year changes or the conclusion that the public restores its desired rate of return to capital by adjusting the quantity of the capital stock it employs, and does so quickly.

18


of labor, and the revenues of the state. AdamSmith laid out the case for treating capital with kidgloves in The Wealth of Nations:

The proprietor of stock is properly a citizenof the world, and is not necessarily attachedto any particular country. He would be aptto abandon the country in which he wasexposed to a vexatious inquisition, in orderto be assessed to a burdensome tax, andwould remove his stock to some othercountry where he could either carry on hisbusiness, or enjoy his fortune more at hisease. By removing his stock he would put anend to all the industry which it hadmaintained in the country which he left.Stock cultivates land; stock employs labor.A tax which tended to drive away stockfrom any particular country would so fartend to dry up every source of revenue bothto the sovereign and to the society. Not onlythe profits of stock, but the rent of land andthe wages of labour would necessarily bemore or less diminished by its removal.17

In addition to the international flow of capital,one must consider the willingness of savers toincrease saving at the expense of consumption andto alter their investment plans as conditionschange. Since Michael Boskin’s 1978 paper on sav-ing and after-tax returns, people have been a bitmore willing to concede some flexibility in savingbehavior.18

Does Atlas Shrug?, edited by Joseph Slemrod,contains a number of interesting studies describ-ing the taxation of the rich and their responses.19

In Chapter 13, “Entrepreneurs, Income Taxes, andInvestment,” authors Robert Carroll, DouglasHoltz–Eakin, Mark Rider, and Harvey S. Rosenexplored the effect of changes in marginal tax rateson the investment behavior of entrepreneurs. Theyfound that “a five-percentage point rise in marginal

tax rates would reduce the proportion of entrepre-neurs who make new capital investments by 10.4percent. Further, such a tax increase would lowermean capital outlays by 9.9 percent.” They add,“the magnitudes of the estimated response arequite substantial. Our response to the questionposed by the title of this volume is that these par-ticular Atlases do indeed shrug.”20

Progressive Taxes on Human Capital MayAlso Hurt Labor, and a Flat Rate Tax May BeBest. People with particularly high levels of humancapital earn returns well above those available toordinary labor. They may have special talents, suchas athletes and entertainers. They may be peoplewith an unusual ability and willingness to makedecisions and manage risk, such as successful entre-preneurs. They may be people who have acquiredadvanced educations and skills. Such people areamong the highest paid people in the country. Theyearn more, but they also face higher average andmarginal tax rates than most workers.

Because labor is not homogeneous and there aresignificant differences in the skill mix across thepopulation, the relative amounts of skilled andunskilled labor can make a difference in the wagerates earned by each group. Taxing the earnings ofpeople with significant human capital at higherrates than ordinary labor may prove to be counter-productive to workers, just as excessive taxation ofphysical capital appears to be. If people with sig-nificant human capital withdraw that capital fromthe market due to high tax rates, the productivity,wages, employment, and incomes of other peoplewho would have worked with them may be low-ered. The tax on the personal service income of thehighly compensated is then shifted to other work-ers and factors.21

Some studies indicate that high-income workersdo not seem to reduce work effort in the presenceof high tax rates. Several reasons are offered.

17. Adam Smith, An Inquiry into the Nature And Causes of the Wealth of Nations, Chapter II, 1776.

18. Michael Boskin, “Taxation, Saving, and the Rate of Interest,” Journal of Political Economy, Vol. 86, Part 2 (April 1978), pp. S3–S27.

19. Joel B. Slemrod, ed., Does Atlas Shrug? The Economic Consequences of Taxing the Rich (New York, N.Y., and Cambridge, Mass.: Russell Sage Foundation and Harvard University Press, 2000).

20. Robert Carroll, Douglas Holtz–Eakin, Mark Rider, and Harvey S. Rosen, “Entrepreneurs, Income Taxes, and Invest-ment,” Chapter 13 in Slemrod, ed., Does Atlas Shrug? pp. 427 and 442.

21. See Franklin Allen, “Optimal Linear Income Taxation with General Equilibrium Effects on Wages,” Journal of Public Economics, Vol. 17 (1982), pp. 135–143.

19


Upper-income individuals may receive some oftheir compensation in the form of “psychic perks”rather than financial rewards. The tax may beavoided by changing the method of compensation.The tax may be shifted to other factors.

Psychic perks might include the power and pres-tige that are associated with prominent positions inbusiness, sports, or entertainment. These perks areunaffected by high tax rates. Economist HenrySimons, godfather of the progressive income tax,offered this as a justification for not fearing adverseconsequences from steeply progressive taxation.Simons dismissed the concern that highly skilledworkers or entrepreneurs would cut back on theirefforts very much simply because they were taxed,on the grounds that their jobs were interesting—”Our captains of industry are mainly engaged not inmaking a living but in playing a great game.”—andthat the status and power attached to these jobswere rewards enough to encourage continuedeffort.22 This cavalier assumption cannot hold,however, when highly progressive rates reach downto tens of millions of small-business owners andprofessional couples in the middle class.

High tax rates can sometimes be avoided byemploying alternative forms of financial compen-sation that allow the recipients to defer the hightax payments, as with pension plans, or by takingthem in a form, such as capital gains or stockoptions, that is subject to a lower rate of taxationand which also have a deferral feature. There hasbeen a surge in stock options as a form of compen-sation in recent years, spurred in part by the 1993Tax Act. That Act raised the top marginal tax ratesto 36 percent and 39.6 percent from 31 percent. Italso decreed that executive salaries in excess of $1million would be non-deductible businessexpenses, apparently in a misguided effort to dis-courage inequality across the wage scale and topunish corporate boards perceived as being toogenerous to top management. To the extent thatthe marginal product of the affected senior man-agement justified the higher salaries, the meddlingof the law reduced economic efficiency and equityrather than enhancing it. The options explosion,however, altered incentives for senior management

and has been blamed for some recent corporatescandals which, though small in number, havebeen rather spectacular.

Another reason that the rich may not appear tobe stampeding into retirement may be that theyare able to shift the tax to other factors. Such peo-ple’s human capital and talents may be in some-what inelastic demand. If so, with only a smallchange in their numbers, they may be able to trig-ger higher compensation to cover their highertaxes. The burden of the tax would shift to otherworkers and consumers without the appearance ofa large reduction in the hours worked of the rich.In a typical production function, a small distinctfactor of production would typically have a smallerelasticity of demand than larger or more readilysubstitutable factors. As highly paid as some CEOsare, their compensation is generally a small per-cent of a business’s total costs, and their knowl-edge of the business and ability to run it atmaximum efficiency may be very hard to replace,at least in the short run. Their administrative orinventive talents, however, may be transferable toother applications, and they may be more mobile,across companies or across borders, than ordinarylabor. This would suggest a further ability to shifttaxes to other factors.

Neutrality and Economic Efficiency Versus Income Redistribution

Neutral Tax Systems Maximize Income. Thepotential damage to ordinary labor from excessivetaxation of capital, both physical and human, issignificant. It suggests that a saving-consumption–neutral tax with a flat rate would serve every typeof economic actor better than the current tax sys-tem, which includes the graduated comprehensivepersonal income tax, the corporate income tax,and the estate and gift taxes. The alternativesmight include a saving-deferred income tax,23 anational retail sales tax, a value-added tax (VAT),24

a returns-exempt Flat Tax,25 or some combination.

The more familiar comprehensive or broad-based income tax in use today taxes most incomeas it is received, including income used for saving,and taxes the returns on saving as soon as they

22. Henry C. Simons, Personal Income Taxation (Chicago, Ill.: University of Chicago Press, 1938), p. 20.

23. A tax on income less net saving, in which all saving is tax-deferred in the manner that current law allows for limited amounts of saving in an ordinary IRA, 401(k), or pension. This type of tax is also called an inflow-outflow tax, a con-sumed income tax, an individual cash flow tax, or an expenditure tax.

20


accrue (except for capital gains, which can bedeferred until realized). Such taxes fall moreheavily on income used for saving than for con-sumption. The tax bias against saving is madeworse by imposing an add-on corporate tax andtransfer (estate and gift) taxes.26 Any justificationof the comprehensive or broad-based income taxand the additional corporate and death dutiesmust rely on significant non-economic social ben-efits because these taxes impose high economiccosts, including reduced incomes across the board.

Redistribution Lowers Total Income and CanHurt Those It Is Designed to Help. Early advo-cates of redistributionist tax systems acknowledgedsome of the costs. Professor Henry Simons was oneof the most influential early advocates of the broad-based income tax. Simons and Professor RobertHaig defended the use of a definition of taxableincome that includes both income saved and thesubsequent returns on the saving, including capitalgains, interest, and dividends (basically, one’sincome was defined as equal to current consump-tion plus the increase in one’s wealth during theyear). This tax base is sometimes described as “theincrease in the ability to consume.” It results in a taxthat is not saving-consumption–neutral; that is, itfalls more heavily on income used for saving thanconsumption.27 Since the rich save more than thepoor, taxing saving more heavily than consumptionis assumed to be “progressive.” Simons also favoredmaking the marginal tax rate structure graduated(higher tax rates imposed on incremental taxableincome as it exceeds specified levels) to furtherincrease the progressivity of the system.

The pure Haig–Simons definition of income didnot allow for a corporate tax in addition to theindividual income tax, however, because thatwould have been an additional layer of double tax-ation. The professors would have preferred anintegrated tax structure that passed corporateincome on to shareholders for taxation as it wasearned, but were thwarted by practical impedi-

ments. Even for these redistributionists, the degreeof double taxation and distortion inherent in anadd-on corporate income tax went too far.

Professor Simons was well aware that the twindistortions of the tax base and the rate structureinherent in the income tax could lead to a drop insaving, investment, and national income. There-fore, he knew of the possibility of adverse shifts inthe tax burden due to heavy taxation of capitalincome and progressivity. In his magnum opus,Personal Income Taxation, Simons wrote:

The case for drastic progression in taxationmust be rested on the case againstinequality—on the ethical or aestheticjudgment that the prevailing distributionof wealth and income reveals a degree(and/or kind) of inequality which isdistinctly evil or unlovely….

The degree of progression in a tax systemmay also affect production and the size ofthe national income available fordistribution. In fact, it is reasonable toexpect that every gain, through taxation, inbetter distribution will be accompanied bysome loss in production….

[I]f reduction in the degree of inequality isa good, then the optimum degree ofprogression must involve a distinctlyadverse effect upon the size of the nationalincome….

But what are the sources of loss, these costsof improved distribution? There arepossible effects (a) upon the supplies ofhighly productive, or at least handsomelyrewarded, personal services, (b) upon theuse of available physical resources, (c)upon the efficiency of enterpriser activity,and (d) upon the accumulation and growthof resources through saving. Of theseeffects, all but the last may be regarded asnegligible….28

24. Value-added tax, including European-style credit invoice method VATs; goods and services taxes or GSTs (as in Can-ada and Australia); or subtraction method VATs (also called business transfer taxes in the United States, such as is pro-posed in the USA Tax).

25. A returns-exempt tax does not allow a deduction for or deferral of current saving, which must be done on an after-tax basis, but it does not subsequently tax the returns on that after-tax saving. It is the method used for Roth IRAs.

26. See note 7.

27. See note 7.

21


As mentioned above, Simons dismissed the con-cern that highly skilled workers or entrepreneurswould make less effort if highly taxed because theyfound their jobs interesting. Simons took moreseriously the possibility that saving and invest-ment would suffer from his policy prescription:

With respect to capital accumulation,however, the consequences are certain to besignificantly adverse…. [I]t is hardlyquestionable that increasing progression isinimical to saving and accumulation…. Thatthe net effect will be increased consumption…hardly admits of doubt.29

Simons’s remedy was not to do away with pro-gressivity, but to offset its effect on saving by run-ning federal budget surpluses:

The contention here is not that there shouldbe correction of the effects of extreme pro-gression upon saving but that governmentsaving, rather than modification of the pro-gression, is the appropriate method foreffecting that correction, if such correction isto be made.30

The assumption that the government virtuouslywould run large budget surpluses to make up forthe anti-growth consequences of a biased and pro-gressive tax system has proven to be utterly naive.Furthermore, a budget surplus cannot make up forthe adverse effects that high corporate or individ-ual tax rates and unfriendly capital cost recoveryallowances have on the present value of after-taxcash flow from an investment—a calculation thatany business school graduate will undertake indeciding on the feasibility of an investmentproject. Thus, even an offsetting budget surpluswould not prevent a reduction in the equilibriumcapital stock from a reduction in the marginalreturn on investment.

Professor Alfred Marshall, who bowed to thegeneral acceptance of progressivity, nonethelessfavored a more neutral graduated tax on consump-tion over a graduated tax on income:

[T]here is a general agreement that a systemof taxation should be adjusted, in more or

less steep graduation, to people’s incomes:or better still to their expenditures. For thatpart of a man’s income, which he saves,contributes again to the Exchequer until it isconsumed by expenditure.31

As Marshall pointed out, one does not need toadopt a non-neutral income tax to achieve pro-gressivity. Saving-consumption–neutral taxes canbe made progressive as well. In fact, it is not neces-sary to have graduated tax rates to achieve pro-gressivity. A tax which exempts some amount ofincome at the bottom and imposes a flat marginaltax rate on income above that amount is progres-sive because the average tax rate will rise withincome. A graduated consumption-based tax isnot as economically efficient as a flat rate con-sumption-based tax because it increases the taxpenalty at the margin the more productive an indi-vidual becomes and the more effort he or shemakes. Nonetheless, it is far more efficient than agraduated income tax.

The tax bias against saving that was built intothe income tax may have been seen as a way ofputting a kinder face on capitalism and defendingthe free market and private property against theforeign ideologies of fascism, national socialism,and communism that seemed to be sweeping theworld in the 1930s. In retrospect, however, we cansee that the broad-based income tax retards invest-ment, which reduces wages and employment andkeeps people who lack savings and access to capi-tal from getting ahead. Taxes on capital formationhurt the poor more than the rich (who can simplyexchange the pleasures of current consumption forthe future income of similar present value thattheir saving would have generated).

Implication of Dynamic Effects of Taxes for Estimating Federal Revenues

A better understanding of the economic conse-quences of taxation would also benefit the Treasuryand the Congress as they plan the federal budgetand contemplate changes in the tax system. Govern-ment revenue estimators generally ignore the effectof tax changes on the overall level of economic

28. Simons, Personal Income Taxation, pp. 18–20.

29. Ibid., pp. 21–23.

30. Ibid., p. 29.

31. Marshall, Principles of Economics, p. 661.

22


activity, employment, incomes, payroll, profits, divi-dends, and capital gains. This method is known as“static revenue estimation” or “static scoring.”

Static scoring leads to misestimates of the effectof tax changes on revenues. In particular, the reve-nue losses from tax reductions that would promotean increase in economic activity are overstated,and the revenue gains from raising taxes in a man-ner that would retard the economy are overstated.Different tax changes have different effects on theeconomy. Ignoring these effects denies Congressand the Executive important information in choos-ing among tax proposals. Inaccurate revenue esti-mates therefore interfere with budget planning andassessment of proposed tax changes. In particular,they exaggerate the difficulty in achieving funda-mental reform of the tax system.

By contrast, “dynamic scoring” would take intoaccount the effect of tax changes on total incomeand its component parts. Dynamic scoring wouldlead to more accurate revenue forecasting and, onewould hope, to tax bills that are more concernedwith increasing national and individual incomeand less inclined toward redistributing a fixed pie.

V. BURDEN TABLES: AN EXERCISE IN MISDIRECTION

Whenever a change is proposed in the tax sys-tem, one of the first questions asked is, “What isthe distribution of the tax increase or decrease?”That is to say, “If this tax change is enacted, whowill pay more, and who will pay less?” or “Whowill be helped or hurt by the tax change?” Onepossible concern is how the “burden” is distrib-uted among people of different incomes; that is,how the tax change affects the progressivity of thetax system.

Burden Table Assumptions, Methods at Odds with Economic Theory, Reality

Tax analysts in the research community, the JCT,the Congressional Budget Office (CBO), and theOffice of Tax Analysis of the Treasury (OTA)present “burden tables” or textual analysis toanswer these questions. The presentation of theseestimates has considerable political import. There-fore, it is important to remember that, when taxanalysts prepare burden tables or present adescription of tax incidence, they must makeassumptions and apply conventions to assign theincidence of the tax to various economic actors, be

they consumers, workers, savers, etc. Amongother things, they must make assumptions aboutthe responsiveness of labor, capital, and consum-ers to the tax and what time frame to consider inpresenting the burden. Some of these conventionshave more to do with convenience than with accu-racy and are, in fact, highly arbitrary and oftencontrary to economic reality.

Incidence, Not Burden. These “burden tables”or “distribution tables” show how a tax proposalwould alter tax payments of individuals across vari-ous income classes or quintiles in a given year, otherthings held constant. (One such table is the Urban–Brookings Tax Policy Center MicrosimulationModel (version 0304-2), prepared jointly by theUrban Institute and the Brookings Institution andavailable on-line. Other methods of display are pos-sible, such as listing how many tax filers get taxreductions of various amounts, how the tax cut isdistributed among single filers, joint filers, familieswith children, the elderly, etc.)

Such tables are based on existing levels of eachtype of pretax income and the existing distributionof whatever exemptions and deductions are in forceat the time of the tax change. They attribute eachtax either to consumers or producers, or to labor orcapital, with a vague nod to economic theory inwhat would be a limited partial equilibrium analysisof the shifting of the tax within its own market if itwere done consistently. However, they generallyassume that taxpayers’ aggregate incomes andbehavior are not affected by the tax change.

Thus, the analysis is cut short of a full explora-tion of the economic consequences of the tax, andthe ultimate burden of the tax is not described.Consequently, these “burden” tables attempt todemonstrate only the initial incidence of the taxes(and should be renamed “incidence tables”). Theytell us virtually nothing about the distribution ofthe burden of the taxes after people adjust theirbehavior as a result of the levies.

Inconsistent Attribution and Sloppy Theory.Furthermore, the conventions used in tax analysisare often inconsistent from one tax to the next andfail to do a good job of demonstrating even the ini-tial incidence of the taxes. In standard JCT burdentables, and in Treasury and CBO analytical work,consumption taxes are usually assumed to be“passed forward” to consumers in the form ofhigher prices. These taxes include:

23


• Retail sales taxes and value added taxes, and

• Excise taxes (whether imposed on the manu-facturer, the distributor, or at the point of retailsale).

Meanwhile, income taxes and other taxes on fac-tors are assumed to be “passed backwards” toworkers and owners of capital in the form of lowertake-home pay and after-tax incomes from savingand investing. These taxes include:

• The personal income taxes (federal, state, andlocal);

• The corporate income taxes (federal, state, andlocal);

• The payroll tax;

• The estate and gift taxes (federal and state); and

• Property taxes.

Customs fees are an exception to this pattern.They are consumption taxes but are assumed (bythe Treasury) to be borne by the suppliers of theforeign labor and capital that produced them.

Consumption taxes, such as a retail sales tax, aVAT, or excise taxes, whether imposed on consum-ers or on manufacturers, are routinely described asbeing paid by consumers in the form of higherprices because it is assumed that consumers are lessflexible than producers, so that consumer pricesincrease by an amount equal to the tax, with noneof the tax borne by the producers of the taxedgoods. It is as if the supply of goods and serviceswere totally elastic, such that production woulddwindle to zero if there were any reduction in theprice received by the producers, so the consumersmust foot the entire bill.

The personal income tax, however, which fallson labor and capital income of individuals, is rou-tinely described as falling entirely on individualincome earners in the form of lower after-taxincomes, with none borne by the consumers oftheir output. The payroll taxes on wages are simi-larly assumed to be borne entirely by labor. Theestate tax is assumed to fall on the decedents, andthe gift tax, if triggered before death, on thedonors. The distribution of the corporate incometax is so uncertain that it is left out of most burdentables but is thought to be borne mainly by eithershareholders (at least in the short run) or workers(in the long run, as capital adapts). These taxes aredescribed as if workers, savers, and investors

offered their labor and capital in totally inelasticsupply, undiminished in quantity, when the taxcuts their compensation. It is assumed that theymake no demand for an increase in compensationin response to the tax, so they swallow the entireburden of the income and other factor taxes thatthey pay.

In effect, the analysts pretend that producerscan shift consumption taxes onto their custom-ers but must absorb income taxes placed on theirown earnings. Supply is infinitely elastic andinfinitely inelastic at the same time. This is aninconsistent approach to tax shifting that is atodds with both economic theory and real-worldexperience.

In addition, neither approach deals with anyfurther adjustments that occur in the real worldwhen taxes are imposed and resources are shiftedin response from one use to another. Such adjust-ments are the province of general equilibriumanalysis.

These questionable presentations of initial inci-dence unfortunately can have a profound effect onthe prospects for adoption of one or another taxchange. Understanding the shortcomings of theexisting “burden” tables that are really bad effortsat “incidence” tables would improve the policydebate. The goal is not so much to arrive at a bet-ter presentation of “incidence” but to redirectattention from the concept of initial incidence andto refocus the debate on the actual economic con-sequences of tax changes, the ultimate burden oftaxation, and the ultimate economic benefits offavorable tax reform.

Snapshots in Time Rather Than LifetimeImpacts. It is very misleading to display the distri-bution of tax changes as affecting people only inproportion to their current earnings.

A very large share of the income inequality inour economy is due to the fact that more experi-enced and older workers earn more than theiryounger counterparts. Most people will experiencea gradual increase in their real incomes as theyadvance in their careers and their work experiencebuilds, followed by a decline in current earningsupon retirement. Even if everyone had the samelifetime incomes, people currently age 50 wouldprobably display higher incomes than people cur-rently age 20 or currently age 80. It is misleadingto characterize these normal age-related or experi-

24


ence-related changes in income over peoples’ livesas class-based income inequality. That, however, isexactly what the burden tables do when they lumpall ages together.

Similarly, saving behavior and ownership ofassets vary with age. A reduction in the tax rate oncapital gains does nothing this year for someonewho has no capital gains this year but will helphim in the future when he has gains to realize.Suppose Mr. Jones turns 70 this year and decidesto sell his business of 50 years for a $1 milliongain. Mr. Smith is only 69 and will wait to sell hisbusiness until next year. The reduction in the capi-tal gains tax from 20 percent to 15 percent savesMr. Jones $50,000 this year and saves Mr. Smithnothing. Should Mr. Smith feel left out? Hardly.He’ll get his benefit next year. The burden tableswould suggest massive unfairness each yearbecause one (different) person each year gets a$50,000 tax break (in the one year of his life inwhich he has a million dollar gain) and anotherperson the same year gets none.

In this illustration, the capital gains of bothJones and Smith had built up over many years.Should the gain be counted as occurring only inthe year it is taken, boosting the realizer into thetop quintile? Would it not better be counted fordistribution purposes as it is accrued (at an aver-age gain of $20,000 a year), which would make itclear that each man is solidly middle-class?Should it be counted at all, in that the gain ismerely the accumulated reinvestment (saving) ofincome recorded in the gross domestic product(GDP) in the years it was originally earned? Thatmakes it double counting, which is why econo-mists do not count capital gains in nationalincome (and why the capital gains tax is doubletaxation to begin with).32

The Treasury has recently constructed and “aged”a panel of taxpayers whose returns it has followedfor several years, based on a sample of the taxpayingpopulation.33 The panel enables the Treasury toexamine how a tax change would affect a typicaltaxpaying population over time, not just in a singleyear. As an illustration, the authors compared theexpanded distributional analysis of the EconomicGrowth and Tax Relief Reconciliation Act of 2001(EGTRRA) and the Jobs and Growth Tax Relief Rec-onciliation Act of 2003 (JGTRRA) over the span ofthe then-current budget period (2004–2013) to thedistribution calculated at a point in time. Looked atover time, the major provisions of the bill benefittedmany more taxpayers than was indicated by a one-year snapshot.

In the panel study, some taxpayers who lackeddividends income or capital gains in some years ofthe period had dividends or capital gains in otheryears and benefitted from the bills’ reductions inthe tax rates on dividends and capital gains. Sometaxpayers who were in the lowest tax brackets insome years were in higher brackets in others andbenefitted from the reduction in marginal tax ratesin the four highest brackets at some time duringthe period. The authors report that:

For example, in the first year 34.7 percentof taxpayers would benefit from thereduction of tax rates above 15 percent,whereas over ten years 60.7 percent wouldbenefit in at least one year…. In the firstyear, some tax return filers do not benefitfrom any of the major provisions ofEGTRRA because they have no income taxliability under pre-EGTRRA law and do notqualify for the expanded refundability ofthe child credit. But over time, nearly alltaxpayers, 94.4 percent, would benefit.34

32. In a very fundamental sense, taxation of capital gains is double taxation of the future income of an asset. Assets have value because they provide income over time (by providing services over time for which the asset’s owner is paid). In fact, the current market price of an asset is the present value of the expected after-tax future earnings of the asset (the future after-tax returns discounted to the present by an appropriate discount rate). It is the after-tax returns that are relevant because that is the only part of the returns that the owner can expect to keep. An asset will rise in value today if there is an increase in what people expect the asset to earn in the future. If the asset does in fact earn the higher expected income in the future, that higher income will be taxed when it is earned. To also tax the rise in the present value of that increased future after-tax income stream (the present-day capital gain) is to tax the future earnings twice.

33. See Julie-Anne Cronin, Janet Holtzblatt, Gillian Hunter, Janet McCubbin, James R. Nunns, and John Cilke, “Treasury’s New Panel Model for Tax Analysis,” prepared for the session on “Forecasting Government Fiscal Situations,” 96th Annual Conference on Taxation, National Tax Association, Chicago, Ill., November 25, 2003; forthcoming in the pro-ceedings of the conference.

25


Over time, then, the benefits of the bill are farmore widely distributed than is indicated by theordinary one-year snapshot of the distribution ofthe tax reduction.

This research goes far in revealing the flaws inher-ent in standard distribution tables and the distribu-tional objections to growth-oriented tax changes.Nonetheless, it still leaves out entirely the economicadjustments induced by the tax changes, which mayhave an even greater role in spreading the benefits ofa growth-oriented tax change. For example, thereduction in the tax rates on dividends and capitalgains lowers the service price of capital and willinduce more investment, which will lead to higherproductivity and higher wages across the board.Consequently, anyone who works will benefit fromthe higher wages triggered by the bill, even if he orshe never has dividends or capital gains. Even peo-ple living entirely on Social Security will benefitfrom the lower cost structure and more plentifulsupply of goods and services made possible by thelower tax rates on wages and capital income. Theseadditional benefits can only be found by taking intoaccount the shifting of the tax burden and thechanges in people’s economic circumstances that aredue to the economic adjustments to the tax changes.

Measuring Dynamic Responses Essential to a True Burden Table

The burden tables normally produced by theTreasury, the congressional committees, and out-side researchers do not take into account the eco-nomic consequences of taxation and the resultingshifts in incomes and tax burdens. These shifts canhave very large effects on the pre-tax incomes ofworkers, savers, and investors at all income levels,which means that they can have a major effect onthe level and distribution of tax burdens. Becausethe burden tables ignore these effects, they do notaccurately measure the tax burden, either in theaggregate or as to how it is distributed among dif-ferent groups within the population.

A true burden table can only be created byundertaking an assessment of the dynamic effectsof the tax on economic behavior. The informationneeded to produce a true burden table is identicalto that which is required for dynamic revenue esti-

mation (discussed earlier). Government revenueestimators are very reluctant to attempt dynamicscoring of the revenue effects of tax changes,claiming that the process is too difficult and con-troversial. If that is correct, then they need to giveup the pretext that the burden tables that they rou-tinely produce are accurate. If one cannot dodynamic scoring of tax changes for budget pur-poses, one cannot generate accurate burden tables.If burden tables are feasible, then so is dynamicscoring, and it should be adopted forthwith.

VI. ANALYSIS OF SOME SPECIFIC TYPES OF TAXES

The Corporate Income TaxInitial Incidence of the Corporate Income

Tax. No competent student of taxation believesthat corporations pay the corporate income tax.Only people pay taxes. Things and abstractions donot pay taxes. A corporation is, in law, a legal per-son, but that is, in fact, a legal fiction. Therefore,corporations do not really pay the corporateincome tax. Conservative Nobel Prize–winningeconomist Milton Friedman is well known forespousing that view, but liberal economists share itas well. The liberal Nobel economist Wassily Leon-tief told The New York Times 20 years ago:

Corporate income taxes fall ultimately onpeople. Economists have tried but havenever succeeded in finding out how theweight of these taxes is ultimately distrib-uted among income groups. There can belittle doubt that elimination of corporateincome taxes would simplify our tax systemand limit its abuse.35

Ultimate Burden of the Corporate IncomeTax. Tax analysts generally assume that the corpo-rate income tax is borne, at least in the firstinstance, by shareholders. As the Treasury put it,“because corporations are owned by shareholders,corporations have no taxpaying ability indepen-dent of their shareholders. Corporations pay taxesout of the incomes of their shareholders.”36 How-ever, the analysis does not stop there.

Economists also recognize that corporate taxes,though initially coming out of shareholders’

34. Ibid., p. 8.

35. Wassily Leontief, “What It Takes to Preserve Social Equity: Amid Dynamic Free Enterprise,” The New York Times, Feb-ruary 1, 1985, p. A29.

26


incomes, have further economic repercussions thatshift part of the ultimate burden to others. As theTreasury report continues:

Importantly, the burden of the corporateincome tax may not fall on shareholders. Acorporate tax change could induceresponses that would alter other forms ofincome as well. For example, some of theburden may be shifted to workers throughlower wages, to consumers through higherprices, to owners of non-corporate capitalthrough lower rates of return on theirinvestments, or to landowners throughlower rents. This shifting might nothappen quickly, so the short-run incidencecould well differ from the long-runincidence.37

(Note the Treasury’s interchangeable use of theterms incidence and burden, for both the short-run own-market effect and the long-run generalequilibrium outcome.)

In years past, the Congressional Budget Office hasalso suggested that the corporate tax falls about halfon owners of capital and about half on the work-force, arguing that the tax depresses capital forma-tion and therefore depresses productivity and wages,shifting at least some of the burden to labor.

More recently, the Treasury and the CBO haveassumed that the corporate tax is borne by ownersof all capital (corporate capital and competingnon-corporate capital), and none by workers.Most economists believe that the burden of thecorporate tax is borne to some extent by share-holders, workers, and consumers (who are oftenthe same people in different roles), but they do notagree on the division of the burden. Because of theuncertainty in the profession, the JCT has stoppedassigning it to anyone in the official “burdentables.” If the corporate income tax were raised

and individual income taxes were cut by equalamounts, the burden tables would show a reduc-tion in the tax on the population with no loss offederal revenue—an ultimate (and quite impossi-ble) free lunch!

Of course, someone pays the corporate incometax even if the JCT cannot point out who it is. Infact, a modern view of the corporate tax in thecontext of an open, globally integrated economyholds that the burden of the corporate tax falls pri-marily on labor after all adjustments are taken intoaccount.

Varying Views of the Corporate Tax. In 1962,Professor Arnold Harberger produced a seminalarticle on the incidence of the corporate incometax.38 The article did more than analyze the corpo-rate tax; it showed the importance of going beyondnarrow partial equilibrium analysis in looking atthe effects of taxation.

The early Harberger work suggested that thecorporate tax was borne by the owners of all capi-tal, not just corporate capital. Harberger assumeda closed economy with a fixed total capital stock. Thecapital could be allocated either to the corporateor to the non-corporate sectors, which wereassumed to produce somewhat different goods andservices.39 If a corporate tax were imposed, raisingthe tax rate above that of the non-corporate sector,capital would migrate to the non-corporate sector.Gross returns would rise in the corporate sectorand fall in the non-corporate sector to equalizeafter-tax yields between the sectors. Thus, a por-tion of the corporate tax would be shifted to non-corporate capital. There would also be an effi-ciency (dead weight) loss that would make theburden greater than the amount of the tax itself.

In later work, Professor Harberger changed hisassumption that the economy is closed and con-cluded that the corporate tax is borne largely by

36. U.S. Department of the Treasury, Report of the Department of the Treasury on Integration of the Individual and Corporate Tax Systems: Taxing Business Income Once, p. 146.

37. Ibid.

38. Arnold C. Harberger, “The Incidence of the Corporation Income Tax,” Journal of Political Economy, Vol. 70, No. 3 (June 1962), pp. 215–240.

39. If the types of business organizations, corporate and non-corporate, were equally effective in all sectors of the econ-omy, then there would be no cross-sector reallocation due to the tax and no reduction in the returns to the non-corpo-rate sector. Corporate businesses would merely shift the form of their organization to non-corporate, giving up whatever efficiencies (for example, ease of financing or trading ownership in a large business) that had driven them to the corporate form to begin with. They would bear the burden of the tax.

27


domestic labor, at least in the case of a small openeconomy that has little impact on the world rate ofreturn.

Putting a tax on the income from corporatecapital would simply lead to adjustmentswhereby less capital would be at work inthat country…. Where would the capitalgo? It would go abroad…. In realizing thatthe presence of the tax implies thatsignificantly less capital will be combiningwith the same amount of total labor (in thesmall developing country), it should comeas no surprise that the equilibrium wagehas to be lower. But there is an additionaland more critical reason (above andbeyond simple capital labor-substitution)why labor’s wage must fall: the need tocompete with the ROW [rest of the world]in the production of manufactures(corporate tradables). The tax is a wedgethat has been inserted into the pre-existingcost structure. The prices of corporatetradable products cannot go up becausethey are set in the world marketplace; thenet-of-tax return to capital cannot go down(except transitorily), because capital willnot be content to earn less here (in thesmall developing country) than abroad.Some element of cost has to be squeezed inorder to fit the new tax wedge into a coststructure with a rigid product price at oneend and a rigid net-of-tax rate of return tocapital on the other. The only soft point inthis cost structure is wages. If they do notyield, the country may simply stopproducing corporate tradables. Or, if thecountry continues to produce such goods,then wages must have yielded—by justenough to absorb the extra taxes that haveto be paid….40

Harberger goes on to point out that the UnitedStates is a large country, not a small one, so the exitof U.S. capital would somewhat depress the rate ofreturn to capital in the world, which would some-

what mitigate the capital flight and reduce theshare of the tax burden passed on to U.S. labor.Nonetheless, he estimates that U.S. labor wouldstill have to bear seven-eighths of the corporatetax.41 Harberger assumes an unchanged world capitalstock, i.e., that the world stock of capital does not fallto restore after-tax returns to the levels they enjoyedbefore the imposition of the U.S. tax. If one insteadadds the assumption that the world capital stock iselastic over time with respect to the rate of return, theneven this modest offset to the impact of the U.S. corpo-rate tax on U.S. labor would vanish.

Harberger reiterated his analysis in a recentinterview in the IMF Survey conducted by PrakeshLoungani.42

LOUNGANI: The effects of some economicpolicies are better understood thanks toyour academic contributions. You didpath-breaking work on whether capital orlabor bears the burden of the corporateincome tax.

HARBERGER: There are interesting devel-opments to report on that front. In theclosed-economy case that I analyzed in the1960s, the natural result is that capital bearsthe burden of the tax and can easily bearmore than the full burden. But my studentsand I have now analyzed the open-economycase, which is more applicable to today’sglobal economy. The result in this case isthat labor bears the burden and can easilybear more than the full burden.

LOUNGANI: That’s quite a flip. Why doesit happen?

HARBERGER: Think of the so-called“tradable goods” sector of an open economy,the sector that produces goods that aretraded on a world market. The prices ofthese goods are determined in the worldmarket. And, with an open economy, therate of return to capital is largely determinedin the world market, because capital canflow from country to country in search of

40. See Arnold C. Harberger, “The ABCs of Corporation Tax Incidence: Insights into the Open-Economy Case,” Chapter 2 in Tax Policy and Economic Growth (Washington, D.C.: American Council for Capital Formation, 1995), pp. 51–73. Cited lines on pp. 51–52.

41. Ibid., p. 61.

42. IMF Survey, Vol. 32, No. 13 (July 14, 2003).

28


the highest return. Now the governmentgets in there and tries to impose acorporation income tax on capital. Well,who bears the burden? Capital can moveacross national boundaries to try to escapethe tax. So it’s labor, the factor of productionthat can’t easily escape national boundaries,that ends up bearing the burden of the tax.

In this analysis, part of the fixed quantity of U.S.capital relocates abroad, and domestic labor suffersa loss in income and therefore bears the entire cor-porate tax, plus a dead weight loss. One could gotwo steps further in refining the analysis, however.

First, one could note the effect of the shift of U.S.capital abroad on foreign labor and world capitalreturns while retaining the idea of a fixed totalworld capital stock. This would put some of theburden of the corporate tax back on U.S. capital. Ifthe United States were a very small economy, theshift in U.S. assets abroad would have little impacton global rates of return, and the Harberger resultfor the U.S. would follow. Given the size of the U.S.economy, however, there would be some effectsabroad. The tax on domestic U.S. corporationswould drive some investment offshore, but thatinvestment would have to compete harder for avail-able foreign labor. Initially, the foreign capital–laborratio would rise, increasing returns to foreign laborbut reducing returns to foreign capital, consisting ofthe expatriate U.S. capital and the pre-existing for-eign capital. The misallocation of the fixed worldcapital would depress capital returns here andabroad. At least temporarily, all capital, U.S. and for-eign, would suffer some loss of income due to theU.S. tax. Nonetheless, U.S. labor would bear mostof the burden of the tax, which would exceed thetax revenue due to the added dead weight burdenof the economic distortions.

Second, however, one really must relax the (stillpartial equilibrium) assumption of a fixed quantityof domestic and world capital. Capital formationhas been shown to be sensitive to the after-taxreturn. Over time, there would be a reduction inthe quantity of foreign-located capital (whetherforeign- or U.S.-owned) to restore its normal after-tax return, reducing the gains to foreign workers.Foreign returns to capital would not decline signif-

icantly. The reduction in the quantity of U.S. capi-tal would restore its original after-tax return aswell. Capital would bear very little of the burdenof the U.S. corporate income tax. In the long run,one should expect a general equilibrium result thatthe main losers would be U.S. workers.

Other analysts have a different view of the corpo-rate income tax in an open, or partially open, econ-omy. For example, Jane Gravelle and Kent Smettersconstruct a model in which the largest part of thecorporate tax can be borne by domestic capital inspite of trade and capital flows, in effect restoringthe old view of who bears the corporate tax.43 Theyget this result by assuming imperfect substitution ofdomestic and foreign capital (people prefer thestocks and bonds of their home country govern-ments and businesses) and imperfect substitution ofdomestic and foreign goods and services. They alsoassume a fixed total capital stock to abstract fromthe issue of the elasticity of saving.

In their four-sector model, they get the usualresult of a corporate tax shifted mainly to domesticlabor when substitution elasticities are very large:Capital moves abroad, equalizing the domesticand foreign after-tax rates of return. The capitalflight depresses rates of return to foreign capital(“exporting” some of the tax) and raises foreignwages. Wages of domestic labor (the immobile fac-tor) fall. But assuming lower elasticities, which theauthors feel are more plausible, less capital shiftsabroad (because it is assumed to be somewhatimmobile too). People are willing to accept a dropin the after-tax return on capital to own domesticassets, and the tax can open a permanent differen-tial between rates of return at home and abroad.As a result, the bulk of the corporate tax falls ondomestic capital, less on domestic labor. Somecapital is exported, which shifts some of the tax toforeign capital with some gains to foreign labor,but less than in the high-elasticity case.

There are several areas of concern with theGravelle–Smetters approach:

• The assumption of a constant world capitalstock is unrealistic, just as it is in the Harbergeranalysis, and simply throws out the bulk of theadjustment process. The quantity of capital hasbeen seen to vary substantially to restore its

43. Jane Gravelle and Kent Smetters, “Who Bears the Burden of the Corporate Income Tax in the Open Economy?” National Bureau of Economic Research, Working Paper No. 8280, May 2001.

29


after-tax rate of return to normal levels overtime following a tax change. The lower world-wide return on capital post-tax would depressglobal capital accumulation and shift the taxback to labor.

• The assumption of a low substitutability ofdomestic and foreign capital appears to be atodds with observed international flows offinancial and physical investment. Even if sav-ers and investors on average display a homecountry preference, the capital markets actvery “open” if even a few large savers are, at themargin, willing to move capital freely acrossborders. It may be that many people never buyforeign securities and many companies preferto invest at home, reducing the average ratio ofglobal to local assets in domestic portfolios. Atthe margin, however, there are many people,businesses, and institutions that freely arbi-trage across borders. Multinational financialand non-financial corporations send funds anddirect fixed investment all over the world.Consider that the outflow of U.S. capital hasbeen averaging roughly $400 billion a year andforeign investment in the U.S. has been averag-ing over $500 billion a year for some years.The sum of the annual cross-border invest-ment flows has been about $1 trillion—almostas large as total annual investment in theUnited States.

• In the cases where the corporate tax falls ondomestic capital, the Gravelle–Smetters modelimplies that a tax increase can lower the after-tax rates of return on capital for a very longtime and can lead to prolonged differences inthe after-tax rates of return on domestic andforeign capital. This is disturbing on twogrounds. First, in the modern world, returnson global assets of similar risk and quality donot display wide and permanent differentials.Second, taxation of capital has risen drasticallyover the past hundred years with the inven-tions of the corporate and personal nationaland sub-national income taxes, property taxes,and estate and inheritance taxes, yet there hasbeen no correspondingly large change in thereal, risk-adjusted after-tax yields on capital,either financial or physical. It appears that cap-ital, by adjusting its quantity, is able to shift alarge part of the taxes aimed at it onto otherfactors.

The Payroll TaxThe entire Social Security payroll tax on wages

is remitted by employers to the Treasury, butaccording to statute, it supposedly is paid half byemployees and half by employers (“statutory obli-gation”). Most economists would argue that, legis-lative language notwithstanding, the initialincidence and the ultimate economic burden ofthe entire tax is borne by workers. Why? Thewhole tax comes out of gross labor compensationthat could otherwise have gone to labor. Further-more, the supply of labor has been thought bymany to be highly “inelastic.” Consequently, thetax is assumed to be “shifted” almost entirely ontothe worker, not only in its initial incidence, butalso in its ultimate burden.

A more modern view of the labor force suggeststhat the workforce, particularly certain subgroups,such as secondary workers in a family and teenag-ers, does respond to changes in the after-tax wage.A general equilibrium economist would argue thatthis partial elasticity of the supply of labor wouldfurther shift a portion of the ultimate burden ofthe payroll tax to other economic factors, such asconsumers, other types of labor, and any immobileforms of capital such as land, as the labor supplyshrinks in response to the tax. Mobile capital,however, would bear little of the burden, as itcould move abroad or shrink in quantity to restoreits original rate of return.

The Unified Estate and Gift TaxesThe federal unified gift and estate tax (the “death

tax”) is an additional layer of tax on saving. Everycent saved to create an estate has either been taxedor will be taxed under some provision of the incometax. Ordinary saving by the decedent was taxedrepeatedly when the decedent and the companiesshe or he may have owned shares in paid individualand corporate income taxes. Saving by the decedentin a tax-deferred retirement plan will be subject tothe heirs’ income taxes and was subject to the corpo-rate income tax in the case of stock holdings. Thedeath tax is always an extra layer of tax.

Prior to 2001, the estate and gift tax rate toppedout at 55 percent if a parent left money to a childbut could reach almost 80 percent under the gen-eration-skipping tax (GST) if the bequest went to agrandchild or other relative more than one genera-tion removed from the decedent. (The GST rate isequivalent to imposing a 55 percent tax on the

30


estate as if it had gone to a child and then impos-ing another 55 percent rate on the remaining 45percent of the estate as if it had gone from thechild to the grandchild. Congress didn’t want tomiss out on any potential revenue by letting any-one’s death go untaxed!)

If a near-to-retirement couple were thinking ofworking an extra year just to add to an estate, thecombined income, payroll, and estate tax ratescould have exceeded 78 percent, or even 90 per-cent with the GST. That produced quite an incen-tive to retire instead of continuing to work or toreinvest interest or dividends in an estate. The2001 Tax Act reduces the top estate tax rate to 45percent by 2007 and raises the exempt amountsfor the estate and gift tax. It will eliminate theestate tax (but not the gift tax) in 2010, but the taxwill reappear at the old rates in 2011 unless Con-gress votes to make the repeal permanent.

Under the conventions used by the Treasury, theunified estate and gift tax is assumed to be borne bythe decedents (or donors if they exceed exemptamounts before they die). The assumption aboutdecedents is distinctly odd, as they are beyond feel-ing any pain. The heirs are the ones who get lowerbequests due to the tax, and they are a more reason-able choice for victims. However, there are noreadily obtainable data on who the heirs are, so thedecedents are selected by default. This is much thesame rationale as that offered by the drunk wholooks for his lost car keys on the sidewalk under thelamp post, instead of in the parking lot where hedropped them, because under the lamp post is theonly place with enough light to search by.

An even odder form of misrepresentation is thatthis tax is not even called a tax in the NationalIncome and Product Accounts, which instead label itas an innocuous-seeming and voluntary-sounding“asset transfer” from the private sector to the govern-ment. It is not a tax, in NIPAnese, because it falls onthe principal rather than the income of the assets—adistinction without economic meaning or merit.

There is one way in which the decedents couldbe said to have borne the estate tax. If they had arigid goal of how much after-tax bequest they

wished to leave their heirs and trimmed their con-sumption during their lifetimes to save additionalsums or to buy additional life insurance to coverthe added tax cost of leaving an estate, then onecould say that they had borne part of the burdenof the tax. However, it is a fundamental law of eco-nomics that the more expensive you make some-thing, the less people will do of it. The estate andgift taxes seem far more likely to reduce the per-sonal saving and capital accumulation of thepotential donors, rather than their personal con-sumption, and therefore to reduce the inheritancesof their heirs.

The heirs do not bear the full cost of the estateand gift taxes, however. These taxes add to the taxon capital formation and result in a reduced stockof capital. The economic consequences of thereduced capital stock are largely borne by thelabor force.

In spite of (or because of) its horrendously hightax rates, the death tax probably doesn’t raise anynet revenue for the government. Professor B. Dou-glas Bernheim of Stanford estimates that avoidanceof the estate tax by giving assets to children, most ofwhom are in lower income tax brackets than theirparents, costs more in income tax revenue on theearnings of the assets than the estate tax picks up.44

Gary and Aldona Robbins of Fiscal Associates esti-mate that the reduced saving and capital formationlower GDP and wages by so much that the resultingreductions in income and payroll tax collectionsexceed the estate tax take.45 If Bernheim and theRobbinses are each even half right, the tax losesmoney. Estate tax repeal would pay for itself andwould encourage wealth and job creation.

VII. CONCLUSIONCenturies of thought and research have been

devoted to the relationship between taxes and eco-nomic behavior. Classical pioneers explored theprice or incentive effects of taxes on the supply offactors and products over 200 years ago. Micro-economists refined the concepts a century later. Inthe middle of the past century, the Keynesian focuson aggregate demand turned taxes into a demandmanagement tool divorced from price or incentive

44. B. Douglas Bernheim, “Does the Estate Tax Raise Revenue?” in Tax Policy and the Economy, Vol. 1, ed. Lawrence H. Summers (Cambridge, Mass.: MIT Press, 1987), pp. 113–138.

45. Gary Robbins and Aldona Robbins, “The Case for Burying the Estate Tax,” Institute for Policy Innovation, IPI Policy Report No. 150, 1999.

31


effects—a theoretical detour that the monetaristschool and the neoclassical resurgence have largelycorrected.

Today, although more sophisticated work thanever before is being done in the tax field, it appearsthat the original insights of the classical pioneersstill hold true. Strenuous efforts to find exceptionsto the “law of demand” have largely come a crop-per. It is still the best presumption that, if some-thing is made more expensive, people will buy lessof it, and if something is made less expensive, peo-ple will buy more of it. This law still applies towork, saving, and investment and to the trade-offbetween current and future consumption, andbetween consumption of market goods and lei-sure. Increase the tax on effort, and less will besupplied. Reduce the tax on effort, and more willbe offered. Fewer inputs mean less total output.Factors of production are largely complementaryto one another. More of one factor of productionboosts the productivity and income of the otherfactors. Less of a factor limits the productivity andincome of all the other factors.

It is well understood in the economics profes-sion that the current tax system imposes heaviertaxes on income used for saving and investment,and on the formation of human capital, than onincome used for consumption. Today, most econ-omists would agree that these tax disincentives tosave and invest, to work and take risk, have con-sequences. They lead people to undersave andoverconsume and to work less and play more.These modern advances in economic understand-ing strongly urge us to dispose of the currentincome tax structure and replace it with a flat ratetax that is neutral in its treatment of saving andconsumption.

The tax biases against saving and investmentand steeply graduated tax rates were introducedfor the purpose of improving “social equity.” Indecades past, it was assumed that the added layersof tax on income used for capital formation woulddo relatively little economic damage, would incon-venience only the wealthy, and would provide sig-nificant income redistribution. It is becomingapparent, however, that most of the taxes thatseem to fall on those who supply physical capital,intellectual capital, or special talents to the pro-duction process may actually be shifted to ordi-nary workers and lower-income retirees in theform of reduced pre-tax and after-tax incomes.

The adverse economic consequences of non-neutral taxation and graduated tax rates, and theresulting adverse impact on “social equity,” are notdisplayed in the so-called burden tables used toinform the public policy debate or the votes inCongress. With bad information, the public andthe Congress are left with a bad tax system and asub-optimal economy.

A more rational system of calculating and display-ing the real tax burden—one that took full accountof how taxes are shifted—would make it easier toexplain and adopt a more rational tax system. Amore rational tax system, in turn, would maximizethe efficiency of the economy as a whole and wouldenable every individual to maximize his or herpotential lifetime productivity and income.

—Stephen J. Entin is President and ExecutiveDirector of the Institute for Research on the Economicsof Taxation (IRET), a Washington, D.C.-based pro–free market economic public policy research organiza-tion. This CDA Report is slightly adapted from IRETPolicy Bulletin No. 88, September 10, 2004, and ispublished by permission of IRET.

a report of the heritage center for data...

Documents