Tax Analysts®Tax Analysts®

My Subscriptions:

Featured News

August 20, 2013
How Broad Should State Sales Tax Bases Be? A Review of the Empirical Literature
by William F. Fox and LeAnn Luna

Full Text Published by Tax Analysts®

This document originally appeared in the September 4, 2006 edition of State Tax Notes.

William F. Fox is the William B. Stokely Distinguished Professor of Business with the Department of Economics and the director of the Center for Business and Economic Research, University of Tennessee. LeAnn Luna is an assistant professor of accounting with the Department of Accounting and Information Management and a research assistant professor of accounting and information with the Center for Business and Economic Research, University of Tennessee.

* * * * *


In a perfect world, sales tax practice would be relatively straightforward because many concepts are agreed on by most economists and are supported by well-articulated models. However, the reality is that state sales taxing systems in the United States do not accord well with normally espoused statements that are at least purportedly drawn from theory and that vary widely from state to state and even between localities in the same state. States have different goals, which makes the usual economic arguments impossible to universally apply to a particular situation. As a result, the policy change needed to improve one state's overall sales tax system might actually be counterproductive in another situation. Further, the theoretical models do not always follow in the real world, where many conditions necessary for the designed results do not exist. Below we discuss the economic arguments regarding sales taxes, review the limited empirical evidence, and, we hope, provide some fresh perspectives on how those theories and that evidence may be applied in the "second-best" world in which we live.


Two basic admonitions underlie much of the policy counsel -- levy taxes on broad bases with low rates and exempt business inputs.1 Normally, the counsel is to tax all consumption unless a strong justification can be given for exemption on some grounds such as equity or excessive administrative and compliance costs. The premise that more consumption belongs in the base can be supported by economic theory, though the advice will not necessarily stand up to careful theoretical scrutiny in the second-best world where many transactions are untaxed. Indeed, even the case for taxing all goods at a flat tax rate follows only from very specialized conditions, and specifically when all goods are equally complementary with the untaxed good (see Auerbach and Hines, 2002).2

As a corollary to a broad consumption base, legislators are counseled to exempt all business-to-business transactions. That admonition is based on the premise that businesses produce -- they do not consume; therefore, the first rule does not apply to business-to-business transactions. However, theoretical justification for violating that rule may exist when the output (consumption) is otherwise untaxed (for example, services).3

Actual differences from those broad principles are readily apparent. Sales tax bases fall far short of taxing all consumption, with various and frequently granted exemptions -- including for most services,4 food for consumption at home,5 and clothing -- in some states. Also, some transactions are exempted based on who sells the items (such as for exemptions for not-for-profit organizations' sales) and when the purchases are made (such as for sales tax holidays).6 Also, a substantial share of state sales tax revenue arises from taxing business-to-business transactions. One estimate is that $105 billion of 2004 state and local sales tax revenue was collected on business-input purchases, representing about 42.9 percent of sales tax collections.7 That estimate is similar to others of the share levied on business transactions.8

Essentially all taxes distort behavior, so the selection of tax structures on economic efficiency grounds comes down to determining which set of distortions causes the least mischief. The relative importance of the potentially offsetting effects of tax distortions cannot be determined theoretically and must be examined empirically. In this regard, it is useful to separate the discussion into the effects of taxation on the behavior of businesses and of individual consumers. That is an important distinction because the effects of extending or retracting exemptions for various transactions can differ depending on whether the items are purchased by consumers for final consumption or by businesses as intermediate inputs. The distinction can also have important implications for whether the exemption is good policy from an efficiency perspective because as a general rule the admonitions are that business purchases should be exempt and consumer purchases should be taxable.

The next two sections provide general overviews of how sales taxes alter the behavior of businesses and consumers and the empirical evidence on the magnitude of those effects on the size of state economies and the overall well-being of state residents. The empirical literature on many of the questions is somewhat sparse, and many concerns await the clarity that will arise from further research. Researchers seeking topics for future study can find plenty of opportunities in what we do not know. The growing literature does, however, shed light on some of the significant issues, and this report seeks to summarize the results.

Economic Effects of Taxing Consumer

This section examines what we know about how sales taxes affect behavior, with the goal of determining whether broad taxation of consumer purchases is good policy. Analyzing the economic consequences of imposing a broad consumption-based sales tax begins with determining the incidence of the tax -- that is, whether sales taxes are ultimately paid by consumers through higher tax-inclusive prices or borne by business owners, workers, or landowners through lower earnings than might occur if businesses lower their prices to offset some or all of the sales tax. Relatively little empirical research is available regarding who bears the state sales tax, but an article by Besley and Rosen (1999) provides some keen insights. The authors use data from 155 cities to examine whether the tax creates higher gross of tax consumer prices for 12 specific commodities. The research suggests that the tax is forward shifted to consumers, and in many cases the price paid by consumers rises by more than the amount of the tax (that is, the tax is overshifted). In another significant study, Poterba (1996) finds that sales taxes are fully shifted to consumers. Both of those studies support the conclusion that the sales tax is paid by the consumer, as is typically assumed.

The conclusion that consumers pay the sales tax, however, applies to a series of standard consumer items that are likely to be purchased locally, and does not necessarily apply to goods or services commonly sold across state lines or to the tax imposed on business inputs. The tax on purchases can have ramifications for business operations in at least two ways, even if it is forward shifted. First, the research does not indicate that the tax on business inputs is borne by consumers, only that the final levy on a taxed sale is paid by consumers.9 Thus, the tax may raise the cost of doing business in the taxing locale versus in other locales. Second, the higher gross of tax price paid by consumers can raise the excess burdens10 of taxes and harm state economies because they distort consumer decisions. As long as some transactions are exempt from sales tax, granting exemptions (or not) can potentially affect decisions on what to purchase (for example, taxed goods or untaxed services), where to make the purchases (for example, a low- or high-tax jurisdiction), and when to make the purchases (for example, during a sales tax holiday).11

Consider first the effects that sales taxes have on consumers' decisions about whether to buy exempt or taxable items. The effects on behavior and tax revenue depends on how responsive consumers are to the price of the exempt versus that of taxable goods. Merriman and Skidmore (2000) indirectly investigate that question as they study how the sales tax rate has affected the allocation of expenditures between retail activity and service activity between 1982 and 1992. That is a reasonable test of the effect that sales taxes have on exempt versus on nonexempt purchases because many services are exempt in most states and many goods are taxable in most states. Merriman and Skidmore find evidence that the share of the economy in the retail sector fell and the share in the service sector rose in the states with high sales tax rates. Their empirical findings indicate that differential taxation can explain as much as one-third of the relative decline in the retail sector and as much as one-eighth of the relative gain in the service sector. That suggests that, as would be expected, sales taxes alter consumption behavior by increasing the quantity demanded for exempt items compared with that for taxable items. A broader sales tax base should lessen the extent of these behavioral distortions.

Hawkins (2002) examines the responsiveness to price changes (including sales tax rates) of several items that are often exempt from the sales tax. He finds that gasoline, tobacco, and food for consumption at home have the greatest responsiveness to price changes (caused by imposition of the sales tax) and utilities and services have lower responsiveness. The excess burden of taxes rises with the price elasticity, suggesting that exemption of gasoline and food is more costly than exemption of services and utilities. Overall, Hawkins finds that the sales tax's excess burden rises with the narrowing of the tax base. Thus, a narrow base that exempts food for consumption at home, most services, gasoline, and utilities has a 38.5 percent higher excess burden than one that taxes consumption broadly. Baum (1998) also finds that broadening the sales tax base to include food lowers the excess burden of the tax.

Russo (2005) uses a simulation model to study the effects on overall economic activity of a broad-based versus narrow-based sales tax. He finds no relationship between the size of the state's economy and the breadth of the base, but that a broader base results in a small reduction in the excess burden of sales taxes in the state. The broader base increases well-being by permitting a lower tax rate12 that lessens incentives to buy those remaining exempt items and by allowing for a relatively small set of choices between exempt and taxable items. States with broad bases may have less to gain from additional base broadening (and there are relatively few opportunities because the base is so broad), but the reverse effect will still arise -- enactment of additional exemptions raises the excess burden of taxes. It should be noted that Russo observes an even larger gain in well-being for states when they combine taxing all consumption with eliminating taxation of business inputs.

Sales taxes can also change where consumers choose to make purchases. Many goods can be purchased remotely, such as when people purchase online or by mail order, when they travel, or when they cross a border to shop. The sales or corresponding use tax is generally due in a state even if the taxable items are purchased in another jurisdiction. Sometimes remote vendors collect the tax on behalf of states,13 and the sales tax should have no effect on decisions of where to purchase because the tax is imposed regardless.14 The tax creates no differential burden in those cases and will not alter where people shop, but households and businesses have a tax incentive to find vendors that do not collect tax for their state. Use tax compliance by purchasers is poor. Washington state found noncompliance with the use tax by registered businesses, at 27.9 percent, to be the worst of any tax. Individual consumers are broadly seen as essentially noncompliant.15

Two recent studies investigate the effect that sales taxes have on the propensity to shop online.16 Goolsbee (2000) examined the effects sales taxes have on Internet shoppers and found that higher sales tax rates increased the incentive to shop online. His analysis relied on data from 1997, early in the e-commerce buying age, making the results less applicable to today than if the study relied on more recent data. Ellison and Ellison (2006) examine the effects of sales taxes on the online purchasing of memory modules. They conclude that sales at online retailers would fall by about 30 percent in the average state if the sales tax were eliminated at the offline stores. Also, research has been conducted on how tax differentials influence on which side of the border people prefer to shop. See Fox (1986) and Walsh and Jones (1988) for examples. The research finds that people respond to tax differentials by doing relatively more of their shopping on the low-tax side of the border, and the research generally finds large cross-border effects.

Russo (2005) also examined effects of extending the sales tax to Internet sales. He finds that state economies would be slightly larger and the excess burden of taxes would be smaller if all Internet sales could be taxed.17 Presumably that is because the incentives to avoid the tax by purchasing out of state via the Internet are eliminated. The study's result is also consistent with the conclusion that a lower sales tax rate is better for a state's economy because it reduces the incentive to buy outside the state.

Economic Effects of Taxing Business-to-
Business Transactions

This section examines how taxes on business inputs affect state economies, with the goal of determining whether input taxes are harmful to the economy. Economists almost uniformly oppose taxes on business-to-business transactions because of how they think imposition of the sales tax on business purchases will influence business behavior. The reasoning starts with the conceptual basis of the tax -- the sales tax is intended as a levy on consumption. Businesses produce, rather than consume (though they may use inputs in the process of producing). It is reasonable to presume that everything businesses purchase is necessary to produce and sell their product, regardless of whether the firm is a manufacturer, wholesaler, retailer, or service provider, so those transactions do not fit within the conceptual framework of a consumption tax.18

The real concern is that taxes on business inputs have the potential to alter business behavior and to harm the state's economy. First, taxing business-to-business transactions can change the way that businesses operate because companies seek to limit the amount of tax they pay. Companies can substitute nontaxable inputs for taxable ones (for example, they can use untaxed labor rather than taxable purchased inputs) when the taxability differs and input substitution is possible. Alternatively, firms can vertically integrate and bring more production within a single company. For example, printing services are often taxable to businesses. A firm can acquire the equipment and train the necessary personnel to have those services produced in-house, avoiding the sales tax on printing services purchased from outside vendors. Companies should be less profitable when taxes alter the way business is done, because companies would bring those services in-house without the tax, if that were generally the lowest cost way to operate.19 No evidence exists regarding the extent to which firms vertically integrate to lessen their tax burdens, but big firms are presumably in the best position to vertically integrate. Not only are smaller businesses less able to vertically integrate because they have less capacity to spread the capital and personnel costs across their more limited activities, but small businesses as a group also are likely made less profitable as larger companies decrease their outsourcing in response to taxation of transactions between firms.

Second, input taxes raise the cost of producing in a state, which can cause some firms to locate their production in states that impose lower tax burdens on business transactions. There is no empirical research that directly examines the extent to which taxes on business inputs harm a state's economy, though some research does consider whether higher sales tax rates generally harm a state's economy. Unfortunately, the empirical results are somewhat conflicting. Bruce, Deskins, and Fox (forthcoming) find that gross state product falls as states increase their sales tax rates. Further, they argue that the effects of taxes on location are growing because technology makes it increasingly easy for firms to geographically separate their production from their markets. Carroll and Wasylenko (1994) (updating earlier work by Wasylenko and McGuire (1985)) study how numerous fiscal variables, including the sales tax, affect total employment and manufacturing employment in a state. They observe no relationship between sales taxes and total employment. However, they find that states with higher sales tax rates had lower manufacturing employment between 1967 and 1983, but that the effects were no longer present when they studied 1984 to 1988. That suggested that the effects of taxes on business location were diminishing -- the opposite conclusion of Bruce, Deskins, and Fox. Carroll and Wasylenko's study predates recent technology and the Internet, however, and may be less applicable to today's more mobile economy.

None of the research examines the issue addressed here of whether firms alter their production decisions in response to the extent that states tax business-to-business transactions and the rates at which those transactions are taxed. The research examines the effects of taxes on the economy, including the implications for consumers and businesses. For example, high tax rates may lower a state's retail sales as consumers and businesses have the incentive to purchase from remote vendors -- a situation in which decisions are made on the demand, rather than the supply, side of the market. Still, it is reasonable to presume that taxes on business purchases discourage firm locations and production in a state (imposition of taxes certainly would not encourage production in a state). Further, those effects are likely largest for firms purchasing the greatest amount of taxable inputs and firms that can most easily separate their point of production and their markets (such as many firms producing for national or international markets). Thus, the effects are likely to vary across industries and sizes of firms.

Third, taxation of business purchases cascades into higher taxes on the final product. The extent of cascading depends on the complexity of the production process (how many levels of production a good or service goes through), the tax treatment of the various business inputs, and the propensity to vertically integrate in the industry. As a result, the amount of cascading can vary significantly across economic sectors. Assuming that business purchases of capital equipment, communications equipment, utilities, and office supplies are taxable, Hawkins (2002) finds that the sales tax is imposed on inputs equal to 14.7 percent of the revenues of electric producers, 11.2 percent for firms taking fees and admissions, 46.7 percent for producers of shelter, 2.6 percent for gasoline stations, and 11.5 percent for firms providing nonshelter lodging. Derrick and Scott (1993) (with their ultimate goal to examine equity issues) use input-output analysis to investigate the extent to which sales taxes fall indirectly on consumers in Maryland through taxation of business inputs. (The authors presume the tax on business inputs is forward shifted.) They find the weighted average indirect tax rate through inputs is somewhat higher than the direct tax on final goods (suggesting greater cascading than Hawkins's study format). The highest tax on business inputs (as a share of spending on the transaction) occurs in industries such us utilities and housing in which the output is generally exempt. Also, input taxation is below average in all industries in which the output is broadly taxed. But, the relationships are not simple because some industries with low direct taxes on outputs, such as education and household operations, also have low input taxes.

The cascading can have important economic effects because it raises the relative price of some goods and causes people to purchase less of them. Interestingly, Hawkins (2002) finds that the excess burden associated with narrow tax bases is lower with the cascading from taxation of business inputs because the taxes cascade most in areas such as many services that are likely to be exempt. That seems to suggest that exempting all business transactions may not be the best policy in states that choose relatively narrow tax bases, because taxing the business transactions may be an indirect means of taxing the final consumption.

Taxing business-to-business transactions broadens the sales tax base and allows a lower tax rate to raise a specific amount of revenue, given the resulting larger tax base. The lower tax rate reduces the disincentives described above, such as for purchasing untaxed items relative to taxed items (for both businesses and consumers) and for vertically integrating. Lower rates also lessen the disincentive to work caused by the tax being imposed on purchased items.20 Thus, the net effect on a state's economy from taxing business inputs depends on the relative size of benefits from the lower tax rate versus costs from altering business behavior. Russo (2005) finds that eliminating the tax on business inputs results in a small increase in the size of the state's economy and a small reduction in excess burden, even though the tax rate must be higher. This conclusion on excess burden differs from Hawkins's, but may suggest that the effects on excess burden depend on the specific set of exemptions.

Administration and Compliance Costs

Theoretical arguments regarding the sales and use tax system frequently ignore administration and compliance costs.21 In principle, consideration of those costs could affect general conclusions on the desirability of broadly taxing consumption and exempting business inputs. Those costs depend on the number of taxpayers, the number of returns to be filed, tax bases, exemptions, and tax rates. Those costs are problematic with destination taxes such as the sales tax because the good must be followed from the point of sale to the point of final consumption. Remote vendors and other businesses with a nationwide market sell to potentially hundreds or even thousands of different taxing jurisdictions, and the complex task of accurately determining the base and rate in the final destination was, historically, a strong argument for not requiring remote vendors to collect sales and use taxes unless the seller has physical nexus in the destination state. The Streamlined Sales and Use Tax Agreement and new technologies have lessened many of the concerns regarding destination sourcing and other compliance issues, but some problems remain.

Even the deceptively simple case in which all business inputs are exempt from tax presents an array of practical problems. The seller must verify that an exemption certificate is valid and obtain some assurance that the intended final use is a business use.22 Under current laws, most states exempt sales for resale, manufacturing equipment, and intermediate inputs that become component parts of the final product (for example, raw materials for a manufacturer). However, many other business inputs -- such as office supplies, furniture, and computers -- are almost always taxed. A seller of potentially exempt items needs a valid exemption certificate and assurance that the final use will be for a qualified business use (for example, intermediate input of a manufacturing process) or is for resale.23 If the item is not eligible to be sold tax-free, the seller must determine the location of the final consumption for taxable goods. And those compliance burdens become even more difficult with remote vendors.

In addition to compliance burdens, exempting business-to-business transactions can create opportunities for tax evasion. Both businesses and consumers use many goods and services, including computers and vehicles. Individuals can buy items tax-free through a legitimate business (or an inactive business with a still-valid exemption certificate) and convert the items to personal use. Also, businesses can order from a low- or no-tax state, transfer the taxed item to a higher-taxed state, and fail to remit the required use tax to the ultimate destination. That scenario is especially problematic now that more products, such as software, can be delivered over the Internet and transferred easily from jurisdiction to jurisdiction. Businesses that do not intend to comply with the use tax have an incentive to buy otherwise taxable items from a vendor that doesn't have nexus in the purchaser's state and that does not collect sales or use tax at the point of sale.

Governments face several issues when administering the sales tax. The first problem is the sheer volume of businesses -- both in-state and out-of-state -- that are required to file sales and use tax returns. The businesses are often small, and an audit other than a numerical accuracy check is difficult given the large volume of taxpayers. Second, exemptions complicate the process because some sales necessarily are tax-free or tax-advantaged and others are fully taxed. Auditors can verify that the seller obtained a valid exemption certificate, for example, but cannot verify without an audit of the buyer that the ultimate use of the item was for exempt use (for example, for resale). Further, an audit by one state will only deal with sales to that state's residents or businesses. Ensuring complete compliance by a particular seller might require an audit by every state where the business sells. An audit of the buyer can reveal items subject to use tax for which tax was not paid, but uncovering items bought in a low-tax or tax-free jurisdiction and transferred to a higher-tax jurisdiction are more difficult to detect. Tracing an item to its final use is prohibitive in many cases.

A broad base that taxes all sales to final consumers of tangible property, intangible property, and services is easier to administer, for both the seller and the government, than is a system with multiple exemptions or partial exemptions. As with all rules, there are some exceptions to that point. For example, taxation of many personal services (such as lawn care and housekeeping services) may be administratively difficult and expensive. Generally, many services are produced by small providers that entail high administrative and compliance costs per firm. All states, for either administrative or fairness issues, have legislated numerous exemptions to their sales tax. Decisionmakers must be aware that the piecemeal approach to sales taxation in the United States -- with its multiple exemptions, bases, and tax rates -- creates real economic burdens on the businesses that must comply with the tax laws and the governments that must administer them. Those costs must be considered alongside the theoretical arguments in support of and against taxing all consumption and exempting all business-to-business transactions.


What can we learn from this review? One conclusion appears clear: Jointly following the two admonitions of good policy advice -- tax consumer purchases broadly, and exempt business-to-business transactions -- is good for economic growth and economic efficiency. States are clear winners because low rates and few exemptions lessen the distortions in consumer behavior, and limited taxation of business inputs makes states more attractive places to produce. Taxing all consumption goods generally eases compliance and administrative burdens, though exemption for all business-to-business transactions poses some thorny problems. Unfortunately, we are hard-pressed to find states that follow those combined rules. A few states, such as Hawaii and New Mexico, tax consumer purchases broadly, but those states continue to tax many business-to-business transactions.

Are the rules still the best policy advice when followed separately? Empirical literature on the taxation of consumer purchases supports the policy advice offered by analysts -- broad consumption tax bases with low tax rates are more efficient and encourage economic growth even if the states continue to tax many business-to-business transactions. The seemingly continuous current pattern of narrowing bases and rising rates -- the opposite of the policy advice -- increases the distortions of state tax structures because it encourages the purchase of nontaxable items at the expense of taxable items and encourages remote purchases at the expense of Main Street. Moreover, multiple exemptions increase compliance and administrative costs. Legislators should be mindful of those costs as they make decisions about whether to grant additional exemptions of goods and services that are heavily purchased by final consumers and should consider whether the goals of those exemptions (which often include equity) can be achieved through means that are less costly to the economy.

A general conclusion is more difficult to reach about exemption of business-to-business transactions in states with narrow consumption tax bases. The analysis generally supports taxation of some business-to-business transactions when the base of consumer taxation is narrow because taxation of the inputs can partially offset the perverse effects of failing to tax the output. Some states have policies that appear explicitly focused on taxing inputs as a means of indirectly taxing the outputs. That may occur only by coincidence because states are not in a very good position to identify which inputs are best taxed to indirectly tax specific outputs. Further, it may be difficult to impose taxes on inputs, because companies may move production outside the state to avoid the tax and the items must be produced and consumed in the same state to tax the output in the right place. Thus, the best policy is probably to exempt business-to-business transactions except in cases in which taxation of inputs is clearly an effective way to tax the outputs.

      Auerbach, Alan J., and James R. Hines Jr. "Taxation and Economic Efficiency," Handbook of Public Economics: Volume 3, Alan J. Auerbach and Martin Feldstein, eds., Elsevier Press, 2002.

      Baum, Donald N. "Economic Effects of Eliminating the Sales Tax Exemption for Food: An Applied General Equilibrium Analysis," Journal of Economics 24 (1): 125-148, 1998.

      Besley, Timothy J., and Harvey S. Rosen. "Sales Taxes and Prices: An Empirical Analysis," National Tax Journal 52 (2): 157-178, June 1999.

      Bruce, Donald; William F. Fox, and Matthew Murray. "To Tax or Not to Tax: The Case of Electronic Commerce," Contemporary Economic Policy 21 (1): 25-40, 2003.

      Bruce, Donald; John Deskins, and William F. Fox. "On the Extent, Growth and Efficiency Consequences of State Business Tax Planning," forthcoming in Corporate Income Taxation in the 21st Century, Alan Auerbach, James Hines, and Joel Slemrod, eds., Cambridge University Press, forthcoming.

      Carroll, Robert, and Michael Wasylenko. "Do State Business Climates Still Matter -- Evidence of a Structural Change," National Tax Journal 47 (1): 19-37, March 1994.

      Cline, Robert; Thomas Neubig, Andrew Phillips, and William F. Fox. "Total State and Local Business Taxes: Nationally 1980-2004 and by State 2000-2004," State Tax Notes, May 9, 2005, p. 423, 2005 STT 88-2, or Doc 2005-8667.

      Derrick, Frederick W., and Charles E. Scott. "Businesses and the Incidence of Sales and Use Taxes," Public Finance Quarterly 21 (2): 210-226, April 1993.

      Due, John E. and John L. Mikesell. Sales Taxation: State and Local Structure and Administration. The Urban Institute Press, Washington, 1995.

      Edmiston, Kelly, and William F. Fox. "A Fresh Look at the Value Added Tax" in The Challenges of Tax Reform in a Global Economy, James Alm, Jorge Martinez, and Mark Rider, eds., Springer Press.

      Ellison, Glen, and Sara Fisher Ellison. "Internet Retail Demand: Taxes, Geography, and Online-Offline Competition." NBER working paper No. 12242, May 2006.

      Fox, William F. "Implications of the Streamlined Sales Tax Agreement for General Excise Tax Revenues," report prepared for the State of Hawaii Office of the Auditor, Mar. 26, 2006, p. 205.

      Fox, William F. "Tax Structure and the Location of Economic Activity Along State Borders," National Tax Journal 39 (4): 387-401, December 1986.

      Fox, William F.; LeAnn Luna, and Matt Murray. "Issues in the Design and Implementation of Production and Consumption VATs for the American States," State Tax Notes, Jan. 21, 2002, p. 205, 2002 STT 14-31, or Doc 2002-1553.

      Fox, William F., and Matthew Murray. "Sales Taxation in a Global Economy," in James Alm, Jorge Martinez-Vazquez and Sally Wallace, eds., Taxing the Hard to Tax, Elsevier Press, 2004.

      Goolsbee, Austan. "In a World Without Borders: The Impact of Taxes on Internet Commerce," Quarterly Journal of Economics 115 (2): 561-576, May 2000.

      Hawkins, Richard. "Popular Substitution Effects: Excess Burden Estimates for General Sales Taxes," National Tax Journal 55 (4): 755-770, December 2002.

      Merriman, David, and Mark Skidmore. "Did Distortionary Sales Taxation Contribute to the Growth of the Service Sector," National Tax Journal 53 (1): 125-142, March 2000.

      Poterba, James. "Retail Price Reactions to Changes in State and Local Sales Taxes," National Tax Journal 44 (2): 165-76, June 1996.

      Ring, Raymond. "Consumers' Share and Producers' Share of the General Sales Tax," National Tax Journal 52 (1): 79-90, March 1999.

      Russo, B. An efficiency analysis of proposed state and local sales tax reforms, Southern Economic Journal 72 (2): 443-462, October 2005.

      Slemord, Joel. "Michigan's Sales and Use Taxes: Portrait and Analysis," in Michigan at the Millenium, Charles Ballard, Paul Courant, D. Drake, Ronald Fisher, and E. Gerber eds., Michigan State University Press, 2003.

      Walsh, Michael J., and Jonathan D. Jones. "More Evidence on the Border Effect -- The Case of West Virginia, 1979-84," National Tax Journal 41 (2): 261-265, June 1988.

      Wasylenko, Michael, and Therese McGuire. Jobs and Taxes -- The Effect of Business Climate on State Employment Growth Rates," National Tax Journal 38 (4): 497-511, December 1985.

      Watson, Harry. "Excess Burden," in The Encyclopedia of Taxation and Tax Policy, Joseph J. Cordes, Robert D. Ebel, and Jane G. Gravelle, eds. The Urban Institute Press: Washington, pp. 121-122.

      Zodrow, George R. "Optimal Commodity Taxation of Traditional and Electronic Commerce," National Tax Journal 59 (1): 7-31, March 2006. Also, "Should Electronic Commerce Receive Preferential Tax Treatment," State Tax Notes, Aug. 14, 2006, p. 423, 2006 STT 156-2, or Doc 2006-13253.


Joel Slemrod (2003) presents a very clever discussion of the theorems, folk and real, regarding sales tax policy.

2 The untaxed good is leisure when the base includes all goods and services.

3 See Bruce, Fox, and Murray (2003).

4 See for a list of sales taxation of services by every state. Technically, most state sales tax bases are not extended to services, rather than services being specifically exempted by statute.

5 See for state tax treatment of food for consumption at home.

6 See for a list of state sales tax holidays.

7 See Cline, Neubig, Phillips, and Fox (2005).

8 For example, see Ring (1999), who estimated that instate consumers pay 59 percent of the sales tax in the average state with the remainder imposed on businesses and tourists.

9 Of course, Besley and Rosen's finding of overshifting may be the result of forward shifting of the tax on inputs.

10 Watson (2005) provides this definition of excess burden: "Excess burden, also called efficiency cost or deadweight loss, occurs when a tax interferes with a taxpayer's economic decisions and efficient choice but generates no additional benefit to the tax collection agency."

11 Tax holidays can affect decisions to cross state borders to make purchases and decisions of what to buy (because some items are exempt during the holiday), effects that are discussed in general through the rest of this section. Holidays can also affect the timing of purchases, though the economic consequences of very short holidays are almost certainly small and are not considered further in this report.

12 A well-known theoretical result, arguing for low tax rates whenever possible, is that excess burden rises with the square of the tax rate.

13 Based on the U.S. Supreme Court ruling in Quill, Inc. v. North Dakota, firms can be required to collect the tax only in states where they have physical presence. Firms can choose to voluntarily collect and remit the tax for states.

14 Fox (2006) reports that just over one-half of the top e-commerce vendors (on a dollar of sales weighted basis) require consumers to remit the Hawaii general excise tax (GET) when they order.

15 The low Hawaii GET collections from the use tax are one piece of evidence. For example, use taxes for both businesses and individuals are responsible for only 1.7 percent of Hawaii's GET collections.

16 Zodrow (2006) uses a simulation model to examine Internet commerce and concludes that the exemption of Internet sales is unlikely to be optimal, and uniform taxation cannot be overturned on the basis of optimal tax findings.

17 For example, gross state product would rise by about 0.5 percent if the tax was extended to Internet sales of tangible goods and tax relief was provided for business inputs.

18 This statement ignores the propensity to use a company to make purchases of goods that are intended for personal consumption. This can be a form of tax evasion that is intended to lower sales tax liabilities and does not represent the firm operating as a business and producing.

19 Of course, vertical integration is the best business model for some activities in some firms, even without the encouragement from taxes.

20 Specifically, the sales tax is not imposed on leisure time but is imposed on many goods and services purchased by consumers.

21 See Fox and Murray (2004) for a review of administration and compliance costs associated with the sales tax.

22 Many of these issues can be avoided with a value added tax, but the VAT presents its own set of administrative and policy issues. See Fox, Luna, and Murray (2002) and Edmiston and Fox (2006).

23 Sellers often accept the exemption certificate in good faith, but the liability for the tax often remains with the seller.

About Tax Analysts

Tax Analysts is an influential provider of tax news and analysis for the global community. Over 150,000 tax professionals in law and accounting firms, corporations, and government agencies rely on Tax Analysts' federal, state, and international content daily. Key products include Tax Notes, Tax Notes Today, State Tax Notes, State Tax Today, Tax Notes International, and Worldwide Tax Daily. Founded in 1970 as a nonprofit organization, Tax Analysts has the industry's largest tax-dedicated correspondent staff, with more than 250 domestic and international correspondents. For more information, visit our home page.

For reprint permission or other information, contact