Wednesday, January 16, 2008

Evidence on Sales and Income Taxes on State Economic Growth

A while back I clipped this quote from a survey of academic papers on state taxes:

“The Effect of Personal Income Tax Rates on Individual and Business Decisions - A Review of the Evidence”
By Mark Rider

Andrew Young School of Policy Studies
International Studies Program Working Paper 6-15

“…there is a large body of evidence that high state [personal income tax] rates have a negative effect on business and individual decisions and thus slow the growth of state employment and personal income. Consequently, states must use care in setting state [personal income tax] rates to make sure they are not out of line with those of their neighbors and other competitor states.” (page iv)

This was quite intriguing and so I started a process of tracking down that papers cited to support that claim. I was left a bit disappointed in the results. This argument seems to be based on an assumed causal link that is speculative: high personal income taxes cause higher wages to compensate, higher wages deter new investment and dissuade businesses to locate there, and finally this causes slower growth. Overall, I found the claims in the paper not always accurate or reliable. An example of one dubious claim is this: "Helms (1985) reports evidence that state tax burdens significantly reduce state growth rates." This is not correct. Helms finds this to be true only when states use tax revenues to fund transfer payments, it is not true when states use tax revenues to fund public services.

In general there is a consensus that overall tax burdens matter somewhat to state growth, but the degree to which growth is sensitive to overall tax burden is a matter of some debate, mostly because the problems of really testing this in a meaningful way. Two careful papers, the Helms paper referenced above and a subsequent paper from two U of O economists suggest that when tax revenues are spend on public investment the negative effect of state taxes disappear or even reverses ( Mofidi and Stone, 1990). But what really matters to the discussion about Oregon is the evidence on the composition of state taxes on growth. Here the evidence is scant and scattered. for the purposes of this post I will present a few snippets of what I have found so far from my survey of the literature on state taxes and growth. One paper that is a little old but intriguing tests whether specific taxes (corporate, personal income and sales) has an affect on subsequent employment growth. There are some methodological issues, but the results are provocative:



National Tax Journal, Dec85, Vol. 38 Issue 4, pp. 497-511.

"The results here confirm that for the wholesale trade, retail trade and finance industries the size of the personal income tax may have an important indirect effect on employment growth. Moreover, the coefficient on wages is also statistically significant in these three industries. If part of the higher income taxes is borne by firms in the form of higher wages, these personal income tax rates also have a direct effect on employment growth in these three industries." (p. 506)

They also find that corporate and sales taxes have no measurable affect except for sales on employment in wholesale trade.

Three related papers by Carroll, Holtz-Eakin, Rider and Rosen investigate the effect of personal income taxes on entrepreneurship activity in the US, using the 1986 Tax Reform Act as a natural experiment and find that higher personal income taxes suppress investment and employment by entrepreneurial firms. However, this is done in isolation, so again does not specifically address the mix of taxes.

Finally, evidence on out-migration and personal income taxes shows little of what I would consider strong evidence positive correlation, but there does seem to be strong evidence that higher wages compensate for higher tax burdens and there is very good evidence that business location decisions and foreign direct investment decisions are sensitive to local wages.

So what is a lay reader of my blog to conclude? Sadly the received evidence does not say anything too clear about the effect of tax composition. Person income taxes appear to have perhaps a slightly higher correlation with employment growth and lower entrepreneurial activity and less investment from outside the state and country. The extent to which shifting a chunk of that burden to sales taxes would ameliorate the problem is still far from clear. Still, I think there is enough evidence to suggest that we may not be doing ourselves any favors by being an exceptionally high income tax state.


Stevelle said...

"Still, I think there is enough evidence to suggest that we may not be doing ourselves any favors by being an exceptionally high income tax state."

What of property taxes? How do our property taxes stack up nationwide after 17 years under Measure 5?

a lay reader

Patrick Emerson said...

Good question, I'll see what I can find out.