The Tax Foundation Fails to Defend its Methodology

The Tax Foundation has responded to critiques of its index in recent editions of their report, but their response misses the mark. Ten years ago, in the first edition of Grading Places: What do the State Business Climate Indexes Really Tell Us?, I pointed to a number of serious flaws in the State Business Tax Climate Index:

• It bears little or no relation to actual business taxes paid in one state versus another.
• The weights applied to the five components make no sense.
• Important aspects of state corporate income tax laws are left out of the index altogether.
• The index gives much weight to a variety of tax features that are of little consequence, and the report provides no evidence that these features matter to business.
• Despite arguing for tax neutrality, they favor some aspects of taxation that make taxes less neutral.
• The sales tax index is riddled with inconsistencies and components unrelated to economic growth.
• The property tax index makes no attempt to measure the business portion of taxes and gives most weight to wealth taxes falling on individuals, with no justification for their connection to economic growth.

The Tax Foundation has not acknowledged any of these critiques in their annual State Business Tax Climate Index (SBTCI) report, choosing instead to pretend that the main point of Grading Places was merely that the five indexes analyzed in the first edition of that book were inconsistent with one another. The Tax Foundation has responded to Tannenwald’s criticism of their component weights, in particular his suggestion that property taxes should carry much more weight than individual income taxes. To this Scott Drenkard and Joseph Henchman of the Foundation replied: “The main problem with that approach is that other taxes affect economic growth more than property taxes, because of their relative elasticity, and the effects of mobility. Such a weighting scheme would also reward states with low property taxes and high income taxes, which probably isn’t anyone’s idea of good tax policy.”¹ In other words, take our word for it, income taxes are important, and besides, we like our weights better because we hate income taxes more than property taxes.

The Tax Foundation devotes much space in each SBTCI report to defending their position that taxes matter a lot to business decisions. In doing so, they cite research on the level of business taxes and how this affects location decisions and growth, and then leap to the conclusion that this means their index matters greatly, despite having just argued that their index does not attempt to measure business tax levels. Furthermore, their review of the research on the effects of taxes on state growth is quite selective. For a more complete review, which comes to rather different conclusions, see State and Local Business Taxes Are Not Significant Determinants of Growth.

A Study Tax Foundation Uses to Support Validity of their Index Actually Underscores its Irrelevance

The Tax Foundation devotes two paragraphs in the SBTCI report to a 2005 report by Bittlingmeyer et al, which attempted to measure whether state business climate indexes could explain differences in economic growth between counties on either side of a state border.² For example, does the relative ranking of Kansas and Missouri on the SBTCI explain why Jackson County, Missouri grew more rapidly or more slowly than its neighboring county in Kansas? The Tax Foundation claims that the findings of this research “support the argument that taxes impact business decisions and economic growth, and they support the validity of the Index.” But here is what the authors of that research actually conclude:

None of the business climate indexes can explain a large proportion of the variation in growth across counties. The best performing business climate indexes explained at most 5% of the variation in relative growth at the borders, suggesting that most of the variation in economic performance is due to factors not captured by state-level business climate measures. This would seem to suggest that business climate is unimportant in driving relative growth among the states. [emphasis added] We conclude that the majority of the variation in growth is due to local business factors that affect comparative advantage, local policies, or state policies not reflected in the indexes. Undoubtedly, some of the variation in border county growth is also attributable to luck.

Importantly, the study by Bittlingmeyer et al makes no attempt to control for other factors influencing county growth rates. It also needs pointing out that the very modest apparent influence of business climate was found precisely where it should have the largest possible effect: between border counties. This result is consistent with the conclusion of the majority of research that the effects of business taxes on statewide growth are very small. Bittlingmeyer’s research, in other words, hardly constitutes an endorsement of state level tax policies to attract business, which entail substantial loss of revenue throughout the state in the hopes of capturing a small share of growth at the borders, growth that is of less significance precisely because it is within a multi-state labor market and therefore may do little to increase the employment of state residents.

1. Scott Drenkard and Joseph Henchman. “Response to Professor Tannenwald’s Critique of the State Business Tax Climate Index.” March 11, 2013. http://taxfoundation.org/article/response-professor-tannenwalds-critique-state-business-tax-climate-index

2. George Bittlingmayer, Liesl Eathington, Arthur P. Hall, and Peter F. Orazem. “Business Climate Indexes: Which Work, Which Don’t, and What Can They Say about the Kansas Economy?” Prepared for Kansas Inc., 2005.