Taxes and Income:Where Does Kentucky Stand?

William H. Hoyt

Much recent debate has focused on the substantial tax “burden” on lower-income households in Kentucky relative to other states and relative to higher-income households in Kentucky. I address these issues by focusing on the “regressivity” of the Kentucky tax system relative to other states and regions in the United States. Two distinct issues are addressed: How much taxes do lower-income households pay within a state relative to higher-income households? And how much taxes do lower-income households pay in absolute terms, that is, what is the average payments in taxes? I analyze these issues by looking at sales, income, and property taxes separately and in total. I find that the Kentucky tax system, while regressive, is less regressive than those of most states, particularly among those in the Southeast. In part this occurs because Kentucky relies more on the most progressive of the taxes, the income tax, and less on the more regressive sales and property taxes than many states. Lower-income households do bear a lighter tax burden relative to higher-income households in Kentucky and compared to most states. But because incomes in Kentucky are lower than in most states, these households still pay relatively high taxes as a percentage of their incomes.

Introduction

Following the re-election of Paul Patton as governor of the Commonwealth of Kentucky, the Lexington Herald-Leader interviewed a number of activists and politicians in the state, asking them what issues they hoped that Governor Patton would consider in his second term.¹ Not surprisingly, several issues related to taxation were brought to the attention of the Governor. One issue that might be of concern, according to University of Kentucky economist Charles Haywood, related to the state’s revenue structure, specifically at how the service sector might be undertaxed. This concern has to do with the efficiency and stability of the tax system.²

The same article, however, mentioned another concern related to taxation: that the poor in Kentucky are penalized, not aided, by the Kentucky tax system. This criticism of the Kentucky tax system is not new and seems to be supported by several recent studies comparing taxes across the states.³ 

Kentucky is by no means alone in the heavy taxation of lower-income households relative to higher-income households. Most state tax systems are regressive, with lower-income households paying a higher percentage of their income in taxes than higher income households. In fact, the Center for Tax Justice (1996) estimates that the state and local tax burden on the highest quintile of income is 29 percent lower, as a percentage of income, than the tax burden on the lowest quintile of income when examined across all states. This regressivity in state and local taxes is a sharp contrast to the significant progressivity of the federal tax system. Despite numerous deductions and income shifting available to higher income households, higher-income households generally pay a substantially higher share of their income in federal taxes than do lower-income households.

That state and local taxes are regressive may not be surprising for two reasons. First, the demand for state and local public services appears to be relatively income inelastic, that is, the amount of spending on state and local services desired by individuals does not increase proportionately with individual income (Craig and Inman (1985) and Gramlich and Rubinfeld (1982)). Second, state and particularly local governments are likely to face a tax base that is much more responsive to higher tax rates. Attempts at increasing the progressivity of taxes may only result in higher-income persons leaving the state or the locality.

In this study, I briefly discuss some evidence regarding the structure of the Kentucky tax system, focusing on how the tax burden varies with income. To address the question of whether Kentucky has a regressive or progressive tax system, it is necessary to have a focus or comparison. Here I compare the taxes paid by Kentucky residents to those for the average for states across the U.S., the states contiguous to Kentucky, and the southeast states. By comparing Kentucky to these other states, we can get a better idea of how low-income households in Kentucky fare and how the tax system might be better designed to assist them.

Tax Structure and Tax Burden

An evaluation of the Kentucky tax system, particularly regarding how tax burden differs with income, only makes sense in reference to the tax systems of other states. How regressive or progressive we are, or could be, may depend on what neighboring states are doing. With this in mind, I compare the Kentucky system to four alternative tax systems: The "average" tax structure in the United States, the tax structure of the contiguous states⁴, the tax structure of the southeast states⁵, and the tax structure of California. California is chosen as an example of a state having an extremely progressive state and local tax system.

Before examining how tax burden varies with income class, I first document how the source of taxes and the amount of taxes differ among the four different regions. In Table 1, we see that the largest source of tax revenue in the United States for combined state and local taxes are sales and excise taxes, the source of 36 percent of tax revenue. This is followed by property taxes with 30 percent of tax revenue and income taxes providing 26 percent of tax revenue.⁶ Kentucky differs from most states in that the income tax is a significantly larger share of its revenues, and the share of tax revenue collected from the property tax is extremely low.

Why might the source of revenue matter for the regressivity of the tax system? Generally, it is argued that sales taxes are the most regressive and the income tax the most progressive of taxes. Income taxes, because of their design and implementation, allow for different rates to be applied to different levels of income. That is, the tax structure is designed to have increasing marginal tax rates. Those with the lowest income levels have the lowest tax rates and are possibly exempt from paying taxes at all. Increasing marginal tax rates, at least in some states, means that higher income households will pay a greater share of their income in taxes.

The sales tax, in contrast to the income tax, is a flat tax assessed on all taxable goods and services. Each dollar of spending on an item or service subject to the state sales tax is taxed at the same rate. Lower- income households, then, will pay the same share of taxes on their spending as high-income households. Despite the exemption of food from the base of most state sales taxes, lower-income households spend a much greater share of their income on goods and services subject to state sales taxes than do higher-income household, and so sales taxes tend to be more regressive.

The property tax falls between the income and sales tax in terms of progressivity. Housing consumption (property value for owners or monthly rent for renters) is the base for this tax. Unlike the goods and services subject to sales taxes, the share of income spent on housing remains relatively constant as income rises.⁷ In addition, numerous states, beginning in the 1960s, offered means-based homestead credits and circuit breakers to reduce the property tax burden on the poor. Chernick (1997) examines how the share of income, sales, and property taxes in state and local tax revenues affects the regressivity of the tax system. The measure of regressivity he uses is the ratio of tax burden (as a percentage of income) for the highest income class to that of the lowest income class. He finds that increasing the share of the income tax in tax collections by 10 percent will, on average, increase this ratio by approximately 10 percent. An increase in sales tax collection by 10 percent would reduce this ratio by 5 percent.

In Table 1 two measures of tax burden are also included for total taxes and the three major taxes. One measure of burden I use is taxes as a percentage of income. Comparing Kentucky to the other regions using this measure, we see that Kentucky has a rate (11.6 percent) that is higher than the national average (11.3 percent) and much higher than the contiguous states (10.5 percent) and the southeast states (10.4 percent).

The second measure I use is per capita tax burden. Here we have a different view of tax burden. While Kentucky’s burden ($2,166) is still higher than that of the southeast states ($2,082), its burden is much lower than the average for the United States ($2,597) and even that of the contiguous states ($2,384).

The explanation for the apparent discrepancies in the two measures is easily reconciled. The last row of Table 1 gives per capita income for the four regions. Since the per capita income in Kentucky ($18,734) is lower than the average in the U.S. ($22,987) or the southeast states ($20,054) the same spending per capita requires a higher tax burden as a share of income. This, I argue later, may be what is really driving the heavy tax burden on lower-income persons in Kentucky.

Comparing Tax Burden Across Income Classes

I now compare the tax burden for households of different income levels for each of the five regions. Data for the calculations here are from Ettlinger (1996). Ettlinger (1996) is an exhaustive and detailed study of determining tax burdens using data on the consumption and spending habits of a large sample of non-elderly married couples. With this large sample of households, he was able to determine how the spending patterns on taxable goods and services and property purchases and rents varied with income. After making assumptions regarding the incidence of the sales and property taxes, he could then estimate taxes paid by the representative household in each state based on that state’s tax system. The personal income taxes were based on federal tax return information. Ettlinger (1996) gives the distribution of tax burden by income quintile. In addition, he segments the top quintile into three separate income groups.

I first consider the property tax burden. In Figure 1 for each of the five regions I give the property tax burden, as a percentage of income, for the income classes. As Figure 1 suggests, the property tax is not a heavy burden on any segment of the Kentucky population and is much less regressive in Kentucky than it is in across the United States. Both the tax burden and the distribution of tax burden in Kentucky look very similar to those of the southeast states.

FIGURE 1
Property Taxes as a Percentage of Income in Kentucky and Comparison Regions         
Source: “Who Pays: A Distributional Analysis of the Tax System in All 50 States,” Citizens for Tax Justice, June 1996.

Figure 2 examines the distribution of tax burden for sales and excise taxes. This figure confirms the regressivity of the sales tax discussed earlier. For Kentucky, the U.S., and the three other regions, the tax burden is substantially higher for households in the lowest income classes than it is for those in the very highest income classes. As Table 1 had suggested, sales taxes are less of a burden for all income classes in Kentucky than in the rest of the U.S. and particularly when compared to the southeast states.

FIGURE 2
Sales and Excise Taxes as a Percentage of Income inKentucky and Comparison Regions

Source: “Who Pays: A Distributional Analysis of the Tax System in All 50 States,” Citizens for Tax Justice, June 1996.

The income tax burden for each of these regions is reported in Figure 3. Here we see that Kentucky is quite different, having a substantially higher burden than the rest of regions with the exception of California at the very highest income levels. Note that in all state regions, the income tax, in contrast to the sales tax, is at least moderately progressive. While Kentucky taxpayers pay more in income taxes than other states, it also appears that they pay more regardless of their income class.

FIGURE 3
Income Taxes as a Percentage of Income in Kentucky and Comparison Regions

Source: “Who Pays: A Distributional Analysis of the Tax System in All 50 States,” Citizens for Tax Justice, June 1996.

This can be seen more clearly in Figure 4, which gives the ratio of the burden for each income class to the burden for middle income classes (3rd quintile). For example, in Kentucky, tax payments as a percentage of income for the lowest income quintile are 41 percent of what they are for those in the middle quintile. Figure 4 again suggests that with the exception of California, the states have moderately progressive tax systems that are relatively similar in terms of relative burden. At low levels of income, Kentucky is very similar to the U.S. average but at upper income levels the average income tax structure in U.S. states becomes more progressive. Given that the maximum marginal tax rate is reached in Kentucky at $8,000 of taxable income, the lack of progressivity at higher income levels is not surprising.

FIGURE 4
Relative Burden of Income Taxes in Kentucky and Comparison Regions

Source: “Who Pays: A Distributional Analysis of the Tax System in All 50 States,” Citizens for Tax Justice, June 1996.

Figure 5 gives the tax burden for all state and local taxes. Here we see that Kentucky has a higher burden at all but the lowest and very highest income brackets. In fact, Kentucky has the lowest burden, as a percentage of income, of any of the five regions for the lowest-income classes. For the 20 percent of households with the lowest incomes, the average tax burden in Kentucky is 10.4 percent in contrast to 12.4 percent for the U.S., and 11.7 percent for the contiguous states.

FIGURE 5
Total Taxes as a Percentage of Income in Kentucky and Comparison Regions

Source: “Who Pays: A Distributional Analysis of the Tax System in All 50 States,” Citizens for Tax Justice, June 1996.

Figure 6 gives a clearer picture of how burden varies with income classes. As with the income taxes and Figure 4, here I give the burden for each income class relative to the burden for the middle income class. Figure 6 suggests that up to the middle income level, Kentucky has a proportional income system with the lowest and second quintiles paying the same share of their income in taxes as the middle quintile. This is in contrast to the regressive tax structure of the other regions at these low income levels. Only when comparing the incomes of the middle quantile to higher income groups does Kentucky display the same regressive nature of total taxes.

FIGURE 6
Relative Burden of Total Taxes in Kentucky and Comparison Regions

Source: “Who Pays: A Distributional Analysis of the Tax System in All 50 States,” Citizens for Tax Justice, June 1996.

Figures 1-3 and 5 displayed both the level of tax burden (in terms of percentage of income paid in taxes) and the relative burden. In Table 2, I focus only on relative burden. In Table 2, I calculate for total as well as the three major component taxes the ratio of the tax burden of the lowest quintile to that of households with incomes in the 80 percent - 95 percent range. For the U.S., households in the lowest quintile pay over twice in much in taxes (as a percentage of income) as those in the 80 percent to 95 percent range. The southeast states and contiguous states have higher ratios of tax burden than the U.S. Kentucky, in contrast to its neighbors, has a lower ratio of tax burden (1.8). While still regressive, this is almost 15 percent lower than the U.S. average.

Further inspection of Table 2 and reference back to Table 1 gives some insights into why Kentucky has a relatively progressive tax structure. Note that the income tax in Kentucky is similar to that of the contiguous states in terms of regressivity, slightly more regressive than that of the U.S., and quite a bit more regressive than that of the southeast states. That Kentucky has more regressive income taxes is due to the fact that a number of counties in Kentucky use payroll taxes, which are only on earned income with no earnings deducted.

Kentucky has a more progressive property tax than the U.S. average that is similar to that of the southeast states and contiguous states. The sales tax is more regressive in Kentucky than the U.S. average but less regressive than the sales taxes operating in the contiguous states and the southeast states.

What is more striking than differences in regressivity across the regions for a single tax is the tremendous differences in the regressivity of the three tax systems. The ratio of tax burdens under the income tax between the lowest income quintile and the highest is 0.26 for U.S. while the same ratio for the sales tax is 5.5. Less technically, the lowest-income households pay only one-fourth as much of their income in state and local income taxes as the highest-income households do while they pay 5.5 times as much of their income in sales taxes as do high-income households.

Then, how is it that Kentucky, which appears to have more regressive income and sales taxes than the U.S. average, has a more progressive overall system? The key to this is found in Table 1, the mix of taxes used in Kentucky compared to the U.S. and the other regions. Kentucky is much more reliant on income taxes than most states and, as suggested by Table 2, regardless of how regressive a state income tax system is relative to other state’s income tax systems, it is far less regressive than a sales tax system. Thus Kentucky’s heavy reliance on the income tax system, typical reliance on sales taxes, and light reliance on property taxes have given it a relatively progressive tax structure.

Conclusion: Are the Poor Paying Too Much in Taxes?

The question of whether the poor are paying too much in taxes is beyond the scope of both this study and is probably a question that could not (and should not) be answered by a professional economists. Here I have suggested that the tax burden on the poor relative to the tax burden on higher- income households is probably slightly lower than in most U.S. states and certainly less than Kentucky’s neighbors and the southeast states.

Why, then, this concern about the poor and taxes in Kentucky? One answer is that taxes for Kentucky’s poor are too high regardless of how other states are taxing their poor. Another possible explanation for this concern is that the poor in Kentucky pay higher absolute taxes than do the poor in other states. As was shown in Table 1, Kentucky residents pay a higher percentage of their income in taxes than the averages of the U.S., the contiguous states, the southeast states, and even California. We find the same when examining Figure 5 as well. Kentucky residents pay a higher percentage of their income in taxes not because spending is higher per capita but because incomes are substantially lower. Thus all taxpayers in Kentucky, poor and rich, pay a greater share of their income in taxes.

There may be reasons to consider modifying the Kentucky tax structure to make it less regressive. Possible modifications might include expanding the sales tax base to include services disproportionately consumed by higher-income classes and increasing the number of tax brackets for the individual income taxes. However, the higher taxes paid by lower-income households in Kentucky appear to have less to do with a regressive tax structure and more to do with a tax base, personal income, that is significantly lower than in other states.

Endnotes

1. Baniak, Peter and Jack Brammer. "Political Observers Await Agenda Patton will Push Using New Clout," Lexington Herald-Leader, Wednesday, November 3, 1999, A1.
2. For a discussion of these concerns of stability and efficiency and the taxation of services, see William H. Hoyt, "The Kentucky Tax System: Problems and Options," Kentucky Annual Economic Report 1995, Center for Business and Economic Research, University of Kentucky.
3. See Lexington Herald-Leader, "Taxing to the Max" (editorial), Wednesday, April 17, 1996.
4. Contiguous states are Indiana, Illinois, Missouri, Ohio, Tennessee, Virginia, and West Virginia.
5. The southeast states used here do not match with the Census definition of the Southeast. Here the southeast states are Alabama, Georgia, Mississippi, North Carolina, South Carolina, and Tennessee.
6. From files provided by the Bureau of Census, Department of Commerce. The totals do not add to 100% because of the omitted categories of taxes on motor vehicles and other taxes.
7. See William H. Hoyt and Stuart S. Rosenthal, "Capital Gains Taxation and the Demand for Owner-Occupied Housing," Review of Economics and Statistics 72 (February 1990): 45-54, for an estimate of the income elasticity of housing.

References

Chernick, Howard. (1997) "On the Determinants of Sub-National Tax Progressivity in the U.S.," Working Paper, Hunter Coolege, CUNY, New York, New York.

Craig, Steven G. and Robert P. Inman. (1985). "Education, Welfare, and the "New" Federalism: State Budgeting in a Federalist Public Economy," Harvey S. Rosen, ed., Studies in State and Local Public Finance, NBER Project Report, Chicago: University of Chicago.

Ettlinger, et. al. (1996). "Who Pays? A Distributional Analysis of the Tax Systems in All 50 States?," Citizens for Tax Justice and the Institute on Taxation and Economic Policy, Washington, D.C.

Gramlich, Edward J. and Daniel L. Rubinfeld. (1982) "Microestimates of Public Spending Demand Functions and Tests of the Tiebout Hypothesis and Median-Voter Hypotheses," Journal of Political Economy, 90:536-60.

Hoyt, William H. and Stuart S. Rosenthal. (1990) "Capital Gains Taxation and the Demand for Owner-Occupied Housing," Review of Economics and Statistics 72: 45-54.

This document is a © of CBER, Center for Business and Economic
Research, located at the University of Kentucky, Gatton College of Business and Economics.