Back to Recent Trends in Kentucky State and Local Tax
PolicyBy David E. Wildasin
From Financing State and Local Government
p. 27-38, published 2001
The sales and use tax is one of the major revenue sources for Kentucky, as it is for many states. This tax is imposed on tangible goods but not on intangibles such as services. In this important respect, the tax is far from a comprehensive tax on household consumption. Furthermore, it appears that a substantial portion of sales tax revenue derives from the taxation of transactions between firms. This compounds the tax burden on some categories of goods, as they are taxed at multiple stages in the production process. This system creates high cumulative effective tax rates on some goods and very low tax rates on others. Reforms that would avoid multiple taxation of some goods while including currently untaxed categories of consumption would likely improve the efficiency of the tax system and bring the sales tax closer to a uniform tax on all consumption.
The discussion in Chapter 2 has identified some of the key elements of Kentucky’s fiscal system. Every one of these features of the fiscal system, considered individually and in relation to each other, warrants close attention from the perspective of the principles outlined in Chapter 1. For example, first considering some components of the tax system in isolation, one should ask whether the personal income tax is equitable in its treatment of individual households. What are its efficiency implications for work effort or for savings and investment in Kentucky? Can compliance with and enforcement of this tax be made easier and less costly? Similarly, one can ask whether Kentucky’s corporation income tax improves or detracts from efficiency in resource allocation. Is the corporation income tax too complex, and could it be simplified without sacrificing important policy objectives? Are corporate and noncorporate enterprises treated in uniform or disparate manners, and what are the efficiency and equity implications of doing so? Each of these questions is complex, important, and worthy of serious analysis. The same is true for similar issues that could be raised with respect to the state and local property taxes, local occupational license taxes, school district finance, and other elements of the state’s fiscal system. Equally importantly, one should consider how the elements of this system work together. Has Kentucky achieved the right balance between individual and business taxes? Should the state shift its emphasis away from sales taxation and toward income taxation, or perhaps the opposite? Should local governments play a larger role in raising revenues to finance their policies and programs, or would it be better for the state to assist them with higher levels of transfers?
It is impossible to address all of these issues adequately within the scope of this study. This chapter is devoted to the analysis of just one component of the state’s overall tax system, the general sales tax, and in fact it focuses on just a few features, hitherto relatively neglected, of this tax. A review of the sales tax and a discussion of possible sales tax reforms illustrates the application of some of the basic tax principles discussed in Chapter 1, and of the ways in which these principles can come into conflict, requiring careful balance among them.
As we have seen, the sales and use tax is one of the major revenue sources for Kentucky, as it is for many states. In its present form, Kentucky’s sales tax is generally imposed on retail sales of tangible products, where “retail” means that the taxed commodity is not to be resold. When tangible products are purchased from out-of-state vendors, use tax is to be paid unless the products are resold. There are many specific exemptions from the sales and use taxes, however. When otherwise taxable products are purchased by governmental and nonprofit entities, they are typically exempt from tax. Furthermore, many products purchased for agricultural use are exempt, as are such items as coal used for electricity production, fuels used in energy-intensive industries, motor vehicles, etc. At the household level, perhaps the most important exemption is that for food. In some cases (motor vehicles, coal, and fuels, for example), products that are exempt from sales and use tax are subject to other taxes. Several important issues arise concerning sales and use taxes in Kentucky.
First, because the tax is levied only on tangible products, it exempts services from taxation. The service sector is a large and growing share of the state’s economy, a trend that may accelerate due to the growth of electronic commerce, but that, in any case, has been quite pronounced for some time. Exclusive reliance on sales taxation of tangible goods may therefore limit revenue growth over time, especially in relation to the size of the state’s economy. Moreover, the exemption of services from sales and use tax gives rise to differential tax burdens across sectors, creating fiscal incentives for the economically inefficient expansion of the service sector at the expense of manufacturing and other activities that produce tangible goods.
Second, the administration and enforcement of use taxes is difficult and, at least at the level of households, not very successful. In the absence of effective enforcement of use taxes, in-state vendors are fiscally-disadvantaged relative to out-of-state vendors, and household consumption patterns are distorted in favor of goods that are purchased from outside of the state and away from goods that are locally produced. For both reasons, the difficulties in use tax collection and enforcement cause economic inefficiency: consumption and production decisions are driven, in part, by tax considerations rather than by underlying economic benefits and costs. To the extent that electronic commerce creates greater opportunities for households (and, to a lesser extent, businesses) to purchase goods from out-of-state vendors, this problem is of growing importance. If the sales and use tax base were expanded to include intangible products, use tax enforcement would become an issue of still greater significance, since it can be especially difficult to monitor and tax the electronic delivery of services.
Third, sales and use taxes are imposed not only on final consumption, that is, sales to households, but also on intermediate products, that is, sales to businesses, except when the products are themselves resold. The sales and use tax is therefore something of a hybrid, containing elements of a tax on final consumption together with elements of a turnover tax, that is, a tax on all sales. Turnover taxes, and other taxes on intermediate-goods transactions, are generally harmful to efficient resource allocation in a number of respects. Taxes on intermediate goods encourage firms to use other, untaxed inputs in the production process, thus distorting the efficiency of production. They encourage firms to perform tasks internally rather than to purchase from other firms, since internal transactions are not subject to tax; this gives rise to incentives that lead to increased vertical integration of firms (whether through internal expansion or through mergers and acquisitions) and creates fiscal penalties for small businesses. Taxes on intermediate goods also increase the effective rates of taxation on final consumption because taxes incurred in prior (upstream) stages of the production process are costs of production that must be recovered through higher product prices. The cumulative effective tax on a given category of final consumption can be substantially greater than the statutory rates at the point of final sale because of the cascading of taxes on intermediate stages of production. Moreover, these cumulative effective tax rates will vary across consumption categories, depending on the extent to which taxable intermediate goods are utilized in the upstream production process and on the degree of vertical integration in the production process.
The incidence of the sales tax is also a potential matter of concern. It is sometimes argued that the sales tax is regressive in its incidence because low-income households consume a larger fraction of their incomes than high-income households. The exemptions for food, health-care related items, and other “necessities” are often justified on the ground that they shift the sales tax burden away from low-income toward middle- and higher-income households. Since the tax is imposed at a flat rate on goods rather than on households, it is not possible to differentiate sales tax liabilities in a way that reflects the varying circumstances of individual households, such as income levels, family size, health status, or age. On the other hand, Kentucky already relies heavily on income taxes, and higher tax burdens on better-off households could undermine the state’s ability to compete for skilled workers, entrepreneurial talent, and complementary resources such as business investment, research and development, and the like. This is especially so if taxes imposed on higher-income households are used to finance services that benefit primarily low-income Kentuckians.
Another important equity concern arises from the uneven application of the sales tax across categories of consumption. Since intangibles are not taxed, households that consume such services and those who have invested or who are employed in service-related industries are taxed more lightly than other citizens. Few would wish to argue that these differences among households constitute a sound basis on which to differentiate tax burdens.
As noted above, it is customary to view the sales tax as a tax on final consumption. However, as the tax is actually administered, in Kentucky and elsewhere, this is only partially accurate. A significant portion of the transactions that are subject to sales tax involve sales from one business to another, that is, “intermediate goods” transactions, rather than sales to consumers. Unfortunately, it is very difficult to determine with any accuracy the share of sales and use taxes that fall on intermediate goods in Kentucky. Businesses that collect sales taxes are not required to record whether taxable sales are made to final consumers or to businesses, and it is therefore not possible to measure directly whether sales taxes are collected on transactions involving households or businesses. One can estimate the amount of taxes collected on final consumption by measuring household expenditures and applying the statutory tax rate to those categories of expenditures that are subject to tax; by this procedure, it has been estimated(18) that only about half (52 percent) of sales tax revenues in Kentucky derive from taxation of final consumption. This means that about half of the revenues are obtained from sales to businesses, i.e., from taxation of intermediate goods. By this estimate, then, it is quite misleading to view the sales tax as just a tax on final consumption; about half of the tax is collected in “upstream” transactions as goods pass through the production process en route to sales to consumers. This has important implications for understanding the sales tax in Kentucky in its present form, and for possible reforms of the sales tax.
The problem of cascading of taxes through the production process is easily illustrated with a hypothetical example. Since services are exempt from taxation, suppose that no firms that supply services purchase any taxable tangible products, that the upstream suppliers of these tangible products also do not purchase any taxable products, and that the same is true for all other firms in the upstream supply chain. Suppose also that firms that sell taxable tangible products to consumers also purchase taxable tangible inputs from their upstream suppliers, and that the degree of taxation of intermediate goods in upstream production processes is the same for all producers of taxable final consumption goods. Under these assumptions, the state’s economy is effectively divided into a portion that is completely exempt from tax (service providers and their upstream suppliers), and another portion that is taxable (firms supplying taxable tangible goods to consumers, and the upstream suppliers of these firms). If just half of sales tax revenue derives from a 6 percent tax on final consumption, then the prices of these taxable final consumption goods must also reflect an additional 6 percent of taxes collected at prior stages in the production process. This means that the cumulative effective rate of taxation on these goods must be 12 percent. That is, while tangible goods are taxed at a nominal rate of 6 percent, they are in effect taxed repeatedly in different stages of the production process, with the effective tax cumulating at each stage.
This simple example relies on unrealistic assumptions. In reality, the state’s economy cannot be divided neatly into one part that entirely escapes sales taxation and one part that is uniformly taxed. Even if services escape taxation at the point of sale, service providers sometimes purchase taxed tangible products, or purchase goods and services from firms that themselves purchase taxed tangible products. Some indirect sales tax burden thus falls on firms producing untaxable services or exempt tangible goods. Firms that sell taxable tangible goods do not rely equally on the purchase of taxable inputs, and the extent of taxable transactions in the upstream stages of the production process varies from industry to industry. Thus, it would be more accurate to characterize the sales tax as a tax that produces a haphazard distribution of effective rates of taxation on different commodity categories.
It is not an easy matter to move beyond hypothetical calculations to estimate the actual distribution of tax rates across commodities because of the complexity of the sales tax system itself, with many specific exemptions for particular industries, because of the varying degrees to which different industries purchase inputs from and sell outputs to out-of-state firms and consumers, and because of our uncertain knowledge of industrial structure within the state. For example, there is a 9 percent tax on motor fuels, and automobiles are subject to a 6 percent tax at the time of registration, including registrations incident to title transfers of used vehicles, but neither are subject to the sales tax, proper. One might regard these special taxes as completely separate from the sales tax or they might be viewed as part of a combined tax structure with special features for particular industries. Other examples of specific exemptions are described further below.
The present analysis does not attempt to capture all of these complexities. Rather, for the sake of illustrative calculations, let us assume that industries producing “tangible goods,” including motor vehicles, motor fuels, etc., are generally subject to taxation at a 6 percent rate, and that other industries (services, such as education, health care services, etc.) are exempt from this 6 percent tax. A state input-output table with 52 industries is used to describe the pattern of trade among industries.(19) In addition, in order to take partial account of particular exemptions for particular industries, or for particular uses of the output of some industries, let us allow for total or partial exemptions for the sales of some tangible goods. Table 4 lists the 50 industries (excluding the government sector) that appear in the state input-output table, and, in the second and third columns, the assumed statutory rate of taxation on the sales of each industry’s output.(20)
Table 4: Effective Sales Tax Rates, by Industry
The last column of Table 4 presents estimates of the effective cumulative tax rate that would be borne by consumers of the output of the industries in each row. For example, in the case of lumber, the effective tax rate is 11 percent, reflecting the 6 percent paid directly by consumers at the point of purchase, but also another 5 percent resulting from sales taxes imposed in various “upstream” transactions in the production of lumber. One can see from this table that the cumulative rates are virtually zero for commodities that are not directly taxed when sold either to consumers or to businesses, although even in these cases there may be a very modest tax embedded in the final purchase price because these industries directly or indirectly utilize taxable commodities as inputs in their production process. By contrast, effective tax rates in the 9 to 12 percent ranges can be found for a number of other commodities. The weighted average tax rate for all commodities is approximately 5 percent, so we can see that some forms of consumption are taxed at effective rates approximately twice as great as average, while the effective rates are nearly zero for others.(21)
The effective sales tax rates shown in Table 4 should not be viewed as reliable estimates for all of the goods and services produced within Kentucky. What they do illustrate is that an apparently simple policy with 6 percent taxation of some transactions and 0 percent taxation of others gives rise to a highly complex and widely varying pattern of effective tax burdens across industries. Much of the variation in effective tax rates is evidently attributable to the complex patterns of interindustry trade. Refining these calculations by accounting more precisely for some of the added complexities of Kentucky’s actual tax structure would change the numerical values of the estimates in Table 4. More precise calculations would not, however, overturn the essential finding that Kentucky’s existing tax system produces a complex and arbitrary pattern of taxation among commodity categories.
For the reasons outlined in Chapter 1, wide variations in effective tax rates create fiscal incentives that encourage some types of production and consumption while discouraging others, giving rise to economic inefficiencies. The economic incidence or distribution of the tax burden, though virtually impossible to estimate with precision, would also exhibit similar wide and apparently quite arbitrary variations. To some degree, heavy rates of taxation of some industries may shift tax burdens back to workers and firms engaged in these industries, rather than forward onto consumers, and this also can affect the equity of the distribution of tax burdens. For example, economic activity in some localities or regions within the state may be relatively heavily oriented toward manufacturing industry, whereas services account for a larger fraction of employment and investment in other regions. The figures in Table 4 suggest that the cascading of tax burdens on tangible goods may reduce employment and investment, and possibly wages and some other sources of income, for the former groups relative to the latter.
Kentucky’s sales tax could be reformed in a number of ways. Perhaps the most important potential reforms relate to the definition of the tax base. Should existing exemptions be repealed, or expanded? Would it be a good idea to include intangibles, or should they continue to be excluded? Can use taxes be effectively imposed on consumers, whether for tangible goods or for intangibles?
As we have seen, if the goal of the sales tax is to achieve a reasonable approximation to a uniform tax on all consumption by Kentuckians, the existing sales tax leaves much to be desired. In some respects, the sales tax is too narrow, and should be broadened. In other respects, it is too broad, and should be narrowed.
Broadening is needed in two main respects. First, consumption of intangibles should be subject to tax. This would include health care services and many other items of consumption that some might argue should be left untaxed on equity grounds. It would also include purchases of intangibles from out-of-state suppliers, including electronic content of various forms (online entertainment, software, and consumer services). Second, out-of-state consumer purchases of tangible goods should be taxed. This part of consumption is already taxed, in principle, under the use tax. Practically speaking, however, this part of the tax is widely ignored, preventing uniform application of the sales tax to all forms of consumption.
Narrowing of the tax base is also needed. Businesses that sell tangible goods to in-state businesses are presently subject to tax, except where exempted. These exemptions help to mitigate the cascading of tax burdens on tangible goods, but they apply nonuniformly. The goal should be to remove all effective tax burdens on business-to-business sales.
It is one thing to state these objectives and quite another to achieve them. From the viewpoint of tax administration, compliance, and enforcement, there are several serious obstacles that stand in the way of uniform taxation of consumption. The difficulties involved in enforcing use tax compliance are well known and need not be discussed at length here. Kentucky is certainly not unique in its struggle to tax mail-order, Internet, and other interstate purchases by consumers. The best hope for progress in this area might lie in coordinated efforts by many states to improve, standardize, and simplify use tax reporting and enforcement.
A more difficult problem is that of distinguishing between sales to consumers and sales to businesses so as to focus the burden of the sales tax on the former and to avoid taxation of the latter. There are at least three possible approaches to this problem, two of which involve systematic extensions of existing policy.
At present, a vendor does not (in general) distinguish between sales to consumers and sales to businesses; if a sale is taxable, it is generally taxable regardless of the identity of the purchaser. However, sales to nonprofit institutions, governments, educational institutions, and certain other classes of purchasers are exempt from sales tax. Administratively, one way to allow for these exemptions is to provide tax-exempt purchasers with some means to identify themselves to vendors and thus to escape the imposition of sales tax at the time of purchase. In principle, this procedure could be carried out more widely, by allowing all businesses to identify themselves as tax-exempt purchasers, thereby obviating the sales tax burden on intermediate goods transactions.
Alternatively, rather than providing exemptions by type of purchasers, one could allow for exemptions by type of commodity. This is exemplified, in current policy, by the exemption not only of the sales of services, but of many specific tangible goods, including food, prescription drugs, coal used for the generation of electricity, tombstones, horses, and ratite birds (ostriches, emus, rheas, kiwis, and cassowaries), to name only a few. One could attempt to identify and exempt from taxation those types of goods and services that are sold exclusively or primarily to businesses.
A third approach, widely practiced in the rest of the world, is to insist on the collection of tax on all sales, without regard to the identity of the purchasers, but to allow businesses to reduce the tax liability on their sales by applying a credit for taxes paid on any purchases that they make from other businesses. This is one method of administering a value-added tax (VAT), so-called because it effectively exposes a business to taxation on the value of its sales minus the value of the purchases it makes from other businesses, i.e., the value that it adds in the production process. Although a VAT is imposed, administratively speaking, at every stage of the production process, all the way to final sales to consumers, the crediting for taxes paid at prior stages in the process prevents the tax from cumulating or cascading. In the end, the VAT is economically equivalent to a tax on final consumption.
Each of these approaches presents a number of administrative challenges and none of them could be expected to function perfectly. Exemption of all sales other than to final consumers is perhaps the simplest approach, at least superficially. The implementation of a VAT would represent the boldest departure from existing U.S. practice, but there is considerable administrative experience with VATs around the world. The most challenging administrative problems with a VAT are likely to arise with respect to taxation of interstate transactions where border formalities cannot be permitted to interfere with the free flow of commerce. The exemption of particular commodities, an important characteristic of current policy, does not “cleanly” distinguish between sales to final consumers and sales to businesses. Conceivably, it can be justified as a workable practical solution when applied judiciously to those commodities that are purchased predominantly by businesses. The exemptions found in current policy certainly do not reflect any systematic attempt to identify these commodities, however. Rather, they appear to have accumulated, on a piecemeal basis, in response to representations on behalf of particular interests.(22)
From an administrative viewpoint, perhaps the simplest reform of the sales tax would be to remove all exemptions and to tax, at a uniform rate, all sales of all goods and services to all purchasers. This would seemingly provide the broadest base and would produce the most revenue for any given rate of taxation. Though simple, this reform would likely impose severe efficiency costs on the state’s economy. It would, in effect, convert the sales tax to a turnover tax, with the adverse efficiency and equity consequences noted previously.
The sales tax is an important revenue source not only for Kentucky but for most U.S. states. Because it applies relatively uniformly to a relatively broad base, it permits revenue to be raised with less damage to economic efficiency than some other possible taxes. However, Kentucky’s sales tax is still far from uniformly applied to all consumption. As a result, some types of consumption are effectively taxed at quite high rates, while other types escape most of the burden of the sales tax. Can the sales tax be improved? The answer is surely yes, but reforms must be undertaken with care.
First, attempts to achieve more uniform taxation of consumption must confront significant administrative obstacles. A simple-minded approach to base broadening would remove existing exemptions for services without addressing the problem of cascading of tax burdens through taxation of intermediate-goods transactions. Such an approach could well do more harm than good from the viewpoint of economic efficiency.
Second, more uniform taxation of all consumption, even if administratively feasible, does not necessarily constitute a policy improvement. In particular, more uniform taxation, while possibly efficiency enhancing, might be objectionable on equity grounds. Exemptions for sales taxes on food, for example, are commonly defended on vertical equity grounds, since food expenditures tend to fall as a fraction of income as income rises. Of course, there are other ways to provide tax relief to low-income households (for example, by taxing income rather than consumption), and most careful economic analyses in fact suggest that tax relief for the poor is usually not very effectively promoted by differential taxation of different types of commodities. Exemptions for health care services or prescription drugs might be justified on ability-to-pay grounds or because consumption of these commodities is seen as particularly meritorious. It is worth bearing in mind, however, that alternative and quite possibly superior policy instruments may be available to promote these goals. For example, explicit subsidies for particular kinds of health care services (for cancer treatment, perhaps, but not necessarily for treatment of sports injuries) or for particular kinds of drugs (brand-name vs. generic variants of drugs) might be more effective at promoting health goals than a sales tax exemption, the benefits of which are in any case limited by the sales tax rate.
While vertical equity and ability to pay considerations might be used to justify existing tax preferences in the sales tax, horizontal equity would be promoted by removal of these preferences. Indeed, many of the arguments advanced in favor of specific tax exemptions or other tax preferences, properly understood, actually illustrate the harm that these policies can create. For example, industry-specific exemptions are frequently advocated or defended because of their ostensible impacts on employment in particular industries or on the economies of particular regions. Such effects illustrate precisely how fiscal incentives cause resource misallocation: tax inducements that cause particular industries to expand also cause other industries to contract, and scarce labor, capital, land, and other resources are driven away from their most productive uses into less productive ones. They also illustrate how specific tax preferences can benefit some groups at the expense of others, very possibly creating inequities rather than alleviating them.
The issue of sales tax reform well illustrates how efficiency, equity, and administrative considerations can come into play simultaneously in policy analysis. There is no unambiguous scientific principle that can dictate how these conflicting considerations should be weighed. On balance, since the current system of sales taxation results in considerable and rather arbitrary differences in effective rates of taxation on different commodities, it is reasonable to conclude that efficiency-oriented reforms targeted at more uniform taxation of final consumption deserve further consideration. The development of more complete reform proposals would require additional detailed technical analysis, however.
To view a list of all chapters in this book, click here. To read the chapters in sequential order, please follow the arrows below.
Back to Recent Trends in Kentucky State and Local Tax
Policy
Ahead to Tax Equity in Kentucky
18 R. Ring, “Consumers’ Share and Producers’ Share of the General Sales Tax,” National Tax Journal (March 1999). See also similar estimates by Michael Childress in Chapter 7 of this volume. Return to text.
19 The Governor’s Office of Economic Analysis has kindly provided an input-output table for Kentucky that it uses for studies of the economic impact of different policies. State and local input-output tables are notoriously difficult to construct owing to the lack of collection of primary data, and estimates based on them must be interpreted with care. Return to text.
20 Data on the share of each industry’s output that is subject to sales tax are not collected, so there is no way to determine precisely what figures should appear in the second and third columns of Table 4. Richard Dobson, of the state Revenue Cabinet, kindly provided advice in making the rough estimates reported in this table. Return to text.
21 This weighted average calculation weights each industry category by the amount of the good or service that is used by consumers. Return to text.
22 The Kentucky Revenue Cabinet publishes Kentucky Tax Facts, a summary of all major tax provisions in the state. The section on sales and use taxes lists no fewer than 60 categories of exemptions, many of which, like that for ratite birds, include many more specific commodities. Return to text.