RALEIGH – Because the North Carolina Senate’s latest tax-increase package has already set off debate about the adequacy and fairness of the state’s tax code, I think that now is a good time for me to offer some clarity regarding the question of whether North Carolina’s tax burden is egregiously regressive.

Here’s the clear answer: it depends.

One reason why politicians and political commentators can engage in debate about what would seem to be a simple math problem is that there are a variety of definitions and assumptions embedded in it. The actual computation, in other words, isn’t the issue. The choice of data is.

For example, should we confine the analysis solely to the taxes directly levied by the state and localities in North Carolina? These would include the personal income tax, corporate income tax, sales tax, and property tax – which together comprise the vast majority of state and local tax collections. That sounds reasonable, until one remembers that federal funds make up a significant share of both state and local budgets. Federal taxes fund most of the Medicaid program, for example, which is one of the single-largest chunks of the budget formulated and administered by North Carolina officials.

Given that all North Carolina taxpayers are, by definition, federal taxpayers, too, one could make a case for including federal taxes as part of any distributional analysis of the tax burden. On the other hand, one could argue that North Carolina tax policies deserve a separate spotlight, since North Carolina officials have direct control over them but not over the federal tax code.

That’s a question about the numerator of the fraction – the taxes collected. Here’s a question about the denominator: how should we measure the income of various groups of taxpayers? The income data typically used in tax-burden analysis bear only a modest resemblance to the living standards of American households. Some the quirks include off-the-books income, the value of government benefits, differences in age, differences in household size among quintiles of the population, and various other measurement problems. Dividing taxes paid by household consumption, rather than income, would probably be a better way to assess the progressivity or regressivity of the tax code, but consumption data aren’t easily found by state and year.

There’s still another question to ponder: who really pays a tax? There’s a big difference between the legal liability of a tax and its incidence. For example, if North Carolina were to jack up its cigarette tax by a dollar a pack, as Gov. Beverly Perdue proposed earlier this year, smokers would obviously bear a large amount of the burden of such a tax. But it wouldn’t be 100 percent. Retailers would likely find it difficult to push the entire tax into higher prices, particularly if their stores are near the border of neighboring states with lower excise taxes. Precise estimates would, however, be exceedingly difficult to compute, making it even harder to determine how much income each quintile of the population would lose to the higher excise tax.

There are similar dynamics with income, property, and retail sales taxes. Those with a legal obligation to send checks to the government do not bear the entire incidence of the tax. Taxes commonly thought to sting the poor will hurt some business owners with medium to high incomes. Taxes commonly thought to sting the wealthy, such as the infamous federal luxury tax on yachts, will hurt employees, vendors, and consumers with medium to low incomes.

All that having been said, if for the sake of argument you take the income data at face value, here’s what you will discover.

North Carolina’s tax code is probably a bit regressive if you exclude federal taxes (in my view, the proper remedy for this would be to cut sales and excise taxes, not increase them). If you include federal taxes, the tax burden slopes upward – the higher the income, the higher the share of taxes paid. And if you include in your analysis an estimate of the government spending received by each income quintile, the system is steeply progressive – the bottom 60 percent of households are net receivers of government taxes, while the top 40 percent of households are net payers of government taxes.

Clear enough for you?

Hood is president of the John Locke Foundation