海角大神

This week the rating agency聽聽got a lot of attention for a study that concluded that rising income inequality is damaging state tax revenues.

Well, state tax revenue growth has slowed in recent decades and income inequality has grown. But the story is far more complicated than S&P suggests. Worse, I fear S&P may have flunked Econ 101 and confused correlation with causation. Two things may be true, but it does not mean聽. Margarine does not break up marriages and the number of computer science doctorates doesn鈥檛 lead to higher arcade revenues. (Seriously click the link above. The spurious correlations are hilarious.)

S&P鈥檚 argument: Following their article concluding that inequality is one factor contributing to slower economic growth, they decided to go further and 鈥渇ind that increasing income inequality is undermining the rate of state tax revenue growth.鈥滺igh income households drain potential sales taxes because the rich save so much more than others and, thus consume fewer goods and services relative to their income and since one person鈥檚 consumption is another person鈥檚 income this lowers income tax bills too.

That鈥檚 true as far as it goes, but the paper largely ignores other important trends that are major drivers of falling state tax revenues 鈥� irrespective of changes in the income distribution.

For example, a major reason why sales tax revenues are in trouble is a shrinking tax base鈥攎ore of what we buy is tax-free. And a big cause of sagging income tax revenues over decades is a decline in income tax rates, especially for high-income households. In other words, rich people are paying less of their income in taxes because state lawmakers have cut their taxes.

Let鈥檚 take these one at a time:

A major reason for declining sales tax revenues is a changing economy. More and more of what we consume is in the form of untaxed services. Personal consumption on goods made up about half of all expenditures in 1960 and was down to one-third in 2010. And more of what we buy we purchase on the Internet, sales that are legally taxable but often uncollected.

In 1950, when the S&P data begin, Jeff Bezos wasn鈥檛 even a twinkle in his parents鈥� eyes. Now Amazon and on-line sales are part of our day- to-day life. Thus, while sales tax revenues may be influenced in part by income inequality (high income people do spend less of their income) blaming the slowdown on the growing income of the Top 1% seems a heavy lift.

Similarly states are relying less on income taxes and their taxes are becoming less progressive. Under a progressive income tax, rising inequality tends to聽increase聽revenues since more income is taxed at top rates. But statutory changes at both the federal and state level have offset that effect.

In the 1950s the federal top marginal rate was over 90 percent. By 1980, it had been cut to 70 percent, and by the 2000s, it was half that. State top tax rates have been cut from an average of 8.7 percent in 1980 to around 6.5 percent through most of the 2000s. States have not only cut rates, they鈥檝e also flattened their tax schedules. In more than half of all income-tax states, the top rate goes into effect at $25,000 or below. Thus, high income households pay at the same rate as middle-income families.

Some states are beginning to counter those trends. We鈥檝e seen a recent return to higher top rates in California, New York and other states. At the same time, however, some states such as Kansas are cutting their statutory rates. Many of those changes are so new they are not reflected in the S&P data.

Another big legislative change came in the early 1980s. Before then, a large chunk of the increase in state income tax revenues came from inflation, as households were pushed into higher tax brackets by revenue codes that were not indexed for inflation. But three decades ago that changed as both the federal and most state codes were indexed and inflation rates fell. As a result, the phenomenon known as bracket creep became a less valuable source of tax revenue.

None of this suggests that S&P was entirely wrong. Income inequality may indeed have some effect on state tax revenue. But to know how much, careful researchers need to control for a changing economy and changing state laws. And S&P didn鈥檛 do that.

Based on what we know now, it is impossible to know whether income inequality caused the long-run slump in state tax revenues or whether inequality and lower revenues are just fellow travelers on the same path.

The post聽聽appeared first on聽.