Wednesday, December 19, 2007

How Volatile are Sales Taxes Compared to Income Taxes?

This is a good question and one on which I have been challenged in the comments. My knowledge of this issue, I have found, is based on evidence that is a bit old and the newest data suggests that the difference is not as great as I had believed. The newest and best evidence I could find comes from an analysis of all states actual tax revenues and is quite interesting to a real wonk. But here are the citation and the relevant results:

"Tax Base Elasticities: A Multi-State Analysis of Long-Run and Short-Run Dynamics"
by Donald Bruce, William F. Fox, and M.H. Tuttle
Southern Economic Journal
Volume 73, Issue 2 (October 2006), pp. 315–341

Table 1: Sales Tax Elasticities
Mean Variance
Long Run Sales Tax Elasticity
0.811 0.048
Short Run Sales Tax Elasticity Above Equilibrium
1.804 7.179
Short Run Sales Tax Elasticity Below Equilibrium
0.149 0.880

Table 2: Income Tax Elasticities
Mean Variance
Long Run Personal Income Tax Elasticity
1.832 0.427
Short Run Personal Income Tax Elasticity Above Equilibrium
2.663 5.014
Short Run Personal Income Tax Elasticity Below Equilibrium
0.217 2.180

These are income elasticities meaning by how much do tax receipts change in percentage terms, from a 1% change in income. Note that sales taxes in the long-run (between year) change quite a bit less than income taxes (0.811 v. 1.832). The authors argue that this is not terribly informative when thinking about volatility. I am not so sure - their point is about maximizing revenue streams not stabilizing revenue collection. Anyway, the short-run (within year) elasticities were found to be quite asymmetric, meaning that they varied a lot depending on if incomes were above equilibrium (basically above average) or below equilibrium. So in good times and bad, in other words. In good times income tax receipts go up quite a bit faster than sales tax receipts (2.663 v. 1.804) and in bad times income taxes fall faster than sales taxes (0.217 v. 0.149) the the difference is small. Notice how small the elasticities are for the bad times. It should be noted that these are based on all states and thus are average effects. States have big differences in their sales tax schemes in terms of what is included and excluded and how much local municipalities tax as well. Most states however, have some exemptions for necessities like non-pre-prepared food.

So the overall picture is that there are higher income elasticities, especially in the long-run, for income taxes than for sales taxes, but that in the short-run the picture looks pretty similar. So is it the long or short-run that we care about? The authors of the study say the short-run is more important because the long-run trend in income is always up. But I disagree, this is exactly what it says: an income elasticity, so when annual aggregate income falls by 5% we should see about a 4% drop in sales tax receipts, but about a 9% drop in income tax receipts. The reverse is true as well. Annual tax revenue is critical in Oregon and this is what these elasticities address. Short term fluctuations in income will not cause the same magnitude of differences, but the differences are there. One thing is clear - income taxes do better in extended periods of income growth, but as we are not worried about growing revenue streams, but lowering volatility, I am not sure this matters much. The conclusion here is probably that it doesn't differ that much though - and going back over other studies that are perhaps not quite as good as this in that they don't utilize as good a data set as this one find that there is little difference in volatility.

Oh, and one other interesting little tidbit of info from the study:

The percent of revenue raised from the sales tax in 2000 varied from 19.5 percent in Virginia to 61.6 percent in Washington. The percent of revenue raised by the personal income tax in 2000 ranged from 16.9 percent in North Dakota to 68.9 percent in Oregon.
I did not know both Oregon and Washington were so extreme and on opposite ends of the spectrum.

1 comment:

darrelplant said...

Washington's study calculates their sales and use tax short-run elasticity at 1.4, 0.2 higher than the figure for their overall tax system.