Wednesday, April 11, 2012

Budget Committee Inequality Report: Taxes

 Paul Ryan makes a number of claims about the tax system, below I quote his major points:

The 1986 tax reform, by cutting top rates and broadening the base, caused the average federal income tax rate paid by the top 1 percent to go up; average federal income tax rates for the bottom 80 percent to fall; and the burden of the federal income tax to shift from the bottom 80 percent to the top 20 percent, especially the top 1 percent. At the same time, the incentive effects of lower rates produced a rise in taxable income and an increase in federal income tax revenues.

The results of this successful reform illustrate the folly of looking only at what is happening to tax rates when analyzing the distributional effects of pro-growth tax policy. Lowering tax rates while eliminating shelters that are overwhelmingly used by upper-income taxpayers results in a more progressive federal income tax overall, while creating the kind of economic growth that allows all income groups to prosper while incidentally bringing in more revenue to the government.

 I have a few issues with this story. First of all tax reform didn't involve just income taxes, it impacted all Federal taxes. The appropriate dataset to look at isn't income taxes alone, it's all Federal taxes. Looking at all Federal taxes reveals a rather different story from that Ryan is telling.

The chart below is reproduced from Piketty and Saez, "How Progressive is the U.S. Federal Tax System? A Historical and International Perspective," using average Federal taxes (including income, payroll, estate, and corporate).

Income Groups 1960 1970 1980 1990 2000 2004

Full Population 21.4  23.3  26.6  25.8  27.4  23.4
P20-40              13.9  18.5  16.3  16.2  13.1    9.4
P40-60              15.9  20.2  21.4  21.0  20.0  16.1
P60-80              16.7  20.7  24.5  24.3  23.9  20.5
P80-90              17.4  20.5  26.7  26.2  26.4  22.7
P90-95              18.7  21.4  27.9  27.9  28.7  24.9
P95-99              23.5  25.6  31.0  28.6  31.1  27.2
P99-99.5           34.0  36.1  37.6  31.5  35.7  31.3
P99.5-99.9        41.4  44.6  43.0  33.0  38.4  33.0
P99.9-99.99      55.3  59.1  51.0  34.3  40.2  34.1
P99.99-100       71.4  74.6  59.3  35.4  40.8  34.7



Many of the changes in tax share have to do with changes other than those involving federal taxes, however, you can clearly see that between 1980 and 1990 the average tax rate on the top 1%, and particularly the top 1% of the 1% fell steeply. Since types of income are distributed unequally across the income distribution it is essential to take into account types of income not captured by income taxes, primarily capital taxation, as well as other forms of taxation that fall primarily on the wealthy, such as estate taxes.

I would also point out that lowering income tax rates reduced the incentives to shift income towards types of compensation taxed at lower rates, such as shifting personal income to business income. It is well established that the major impact of tax changes is on shifting taxation between categories and in timing, rather than in creating incentive effects. I find it highly dubious that the majority of the increased income taxation observed by Ryan has to do with incentives instead of simply being a shift in how income is accounted for (such as a reduction in concealing personal income as corporate expenses).

I would also add that there is little evidence that tax changes are all that powerful as explanations for growth. Below is a chart showing US per capita growth rates. Eyeballing it should be enough to show that there aren't any large shifts in growth rates as a result of US tax changes. While tax rates probably have some impact on GDP they are simply not all that important relative to other factors (I would also note that growth relative to changes in overall taxation don't show any particularly strong relationship, but this is beyond the scope of my commentary on Ryan's inequality paper).


Data retrieved from World Bank (http://www.worldbank.org/).

[Tax rates were changed throughout the period, with major reforms in 1969, 1981, 1984, 1990, 1993, and 2001. 1990 and 1993 were tax raises, though they didn't raise rates anywhere close to those of 1980. There were also payroll tax increases up to about 1990 and smaller tax reforms (such as the Taxpayer Relief Act of 1997) throughout the period. Growth rates are noisy, but if taxes were a particularly large determinant of growth you'd think that there would be clear evidence even through the noise with marginal rates collapsing from 87% in 1954 to 28% in 1986, but we don't see any swings in economic growth even broadly comparable to these drastic changes)

Further, I would also note that while decreases in income tax brackets have been fairly stable and shown a long-term downward trend base broadening measures have been far more transient. Michael J. Graetz observes in "Tax Reform Unraveling" that the total number of tax expenditures has grown from 80 in 1986 to 146 in 2006. The cost of tax expenditures has changed from $500 billion in 1986 to $360 billion in 1988 and then increased to $685 billion in 2002 (all in 2004 dollars).

Given this history, while I find it very likely that any tax rate reductions implemented would prove permanent, I find it highly implausible that base broadening measures and loophole closing will prove to be similarly persistent. The trend has always been towards permanent tax decreases through a two step strategy of lowering rates in return for base broadening and loophole closing that make the reform revenue neutral then the addition of new loopholes and base narrowing that result in a permanent reduction in rates. I see no reason why this wouldn't happen again with Ryan's reforms. This is a persistent trend in American tax history, it has been very difficult to maintain a broad tax base and to eliminate tax expenditures. However, it has also been very difficult to raise tax rates once they have been lowered in exchange loophole closing.

Ryan's statements on taxes are highly biased and implausible. They are clearly meant to deceive its readers by cherry picking tax data to tell the story he wants, rather than use the more obviously appropriate data such as overall tax rates. Further, he tells an implausible story about the 86 tax reform, it is well known that the tax loopholes took little time to creep back into the tax code and that per capita growth was less than incredible afterwards. Ryan's entire treatment of the tax changes looks wrong to me and his suggested changes even more so. Until Congress can prove that it can commit to loophole closing I am very strongly opposed to lowering rates in return for these reforms. If history is any guide, we will end up keeping the low rates and having the loopholes open back up. This would be disastrous to the budget. The sensible way forward is to broaden the base and close the loopholes and use the revenue to pay down the debt. Once the debt is far lower than today and Congress has made a credible commitment to keeping loopholes closed then lowering tax rates may be something worth doing. But until then, I consider anyone in government advocating for revenue neutral tax reform to be actually advocating to simply slash tax rates. We know from experience loopholes won't stay closed while rates are likely to remain at permanently lower rates. It's a fools bargain.

Bibliography:
Piketty, Thomas and Saez, Emmanuel. "How Progressive is the U.S. Federal Tax System? A Historical and International Perspective." National Bureau of Economic Research. Cambridge, Massachusetts. 2006. http://www.nber.org/papers/w12404

Graetz, Michael J. "Tax Reform Unraveling." The Journal of Economic Perspectives. Vol. 21. No. 1. 2007. pp. 69 -90.

No comments:

Post a Comment