I had been avoiding posting about Oregon's revenue system for quite a while now because I have a real job and spending time digging into the evidence on tax systems was not a high priority. However, the opinion piece in the Oregonian by Scott Bruun finally prompted me to make a first foray into the debate and I considered a kind of commitment device to finally get my hands dirty. So I decided to start a discussion and over the next month or so I could start to see what the evidence was. After all I have only three readers and they are all pretty patient. Then comes Jeff, who throws gasoline on my nice slow burning fire, and here I am, spending half the night looking at income/sales tax revenue elasticities! Thank you and curse you Jeff Alworth!
So first lesson: a blog is a public forum and so you had better be prepared to engage the public.
But, now that I have looked at a number of studies and have updated my knowledge from my policy student days (gosh, can it really be 15 years ago already?) I am left with two conclusions about the variability of sales taxes.
Second lesson: sales tax revenue is not significantly less volatile than income taxes in the short run and may, in fact, be more volatile.
In a number of studies, short run volatility is actually greater for sales than income tax revenue. Its volatility can change depending on the exemptions and other specifics of the tax, and for some states it is quite stable while in others it is quite volatile, but overall the evidence is simply not there to support the lower volatility claim. What this means is that is the case of short-run income shocks the state is potentially as bad off or worse off with a sales tax.
There is, however strong and consistent evidence that the long-run income elasticity from sales taxes is lower than income. This is not questioned anywhere in the literature, but the interpretations differ. The evidence cited in a previous post show a a fairly significant difference between long-run elasticities for sales and income taxes. This is argued to be largely reflective of the growth rates of the two taxes as incomes in general grow and does not tell us that much about stability. (Which, by the way, is one reason fiscal conservatives don't like income taxes as they grow too fast in their view) In other words, it is argued that this is mostly an indication of a trend and not about variability around that trend. But I am not entirely convinced: it does inform us about, for example, the effects of an extended recession - suppose a three year span of low state income (sound familiar?) - a sales tax should fall less severely in that case.
Third lesson: a sales tax might mitigate the downturn in government revenue during an extended downturn in the economy.
So what does this tell us? Well, it probably says that volatility itself should not be a determining factor in choosing a sales tax. How likely are short-run fluctuations v. long-run downturns and how much each matter seem to be pertinent questions. Oregon has been severely hurt by extended slowdowns of the economy but has also struggled with more short-term economic variation as well.
Forth lesson: with whatever revenue source you have, excepting somewhat property taxes, there is going to be significant volatility.
Whether income or sales are more or less volatile in the short- and long-runs, the essential fact is that they are both volatile and mechanisms for smoothing, rainy-day funds if you will, are necessary if the government really wants to reduce volatility significantly. Sales taxes will not do it. A rainy day fund requires restraint when revenues are high and a reserve to draw on when times are good.
So finally what I am left with is a belief that diversification of revenue sources may help somewhat but that volatility is simply not a particularly compelling case for sales taxes.