Comments Guidelines

All comments are pre-moderated. No spam, slurs, personal attacks, or foul language will be allowed.

Friday, December 21, 2012

Conservative ALEC Economic Policies Have No Benefit, Some Risks, for States

The economic policies proposed for states by the conservative American Legislative Council Exchange (ALEC) not only don't work, but carry risks for states' economies, according to new research by the Iowa Policy Project and Good Jobs First.

As we have seen most recently with Michigan's passage of anti-union so-called "right to work" legislation (which lets people free ride on union contracts and makes union organization and representation more difficult), the legislative agenda of ALEC proclaims that states should follow a low-tax, low-wage, non-union route to economic prosperity. Now, you or I might wonder how creating low-wage jobs is supposed to create prosperity, but luckily for us ALEC has ranked the states by how "competitive" their economic policies were beginning in 2007, giving us the chance to see how well their recommended policies have done.

ALEC's 15 recommended factors (p. 18) include taxes, debt service, public employees per 10,000 residents, "quality" of the legal system, "right to work," minimum wage, and workers' compensation costs. As pointed out by earlier commentaries, it says nothing about education or infrastructure, which have clear effects on a state's economy. The new report by economist Peter Fisher with Greg LeRoy and Phil Mattera undertakes a statistical analysis of these policies, using the ALEC ranking of all 50 states as of 2007 to see how well their economies have performed since then. Fisher et al. also highlight the shoddy statistical work by Arthur Laffer in creating the ALEC index and reporting results.

Whereas Laffer frequently makes his points simply by comparing the top vs. bottom 8-10 states, Fisher et al. start with a full comparison of all 50 states via a correlation analysis, then proceed to the necessary addition of holding other potential causes constant in what is know as multiple regression analysis. Beginning with the correlations, here is what the new report finds. Correlation runs from -1 (perfect negative relationship) to +1 (perfect positive relationship); the closer to +1 below, the better the ALEC competitiveness index predicted the following outcomes. All changes are from 2007 to 2011.

ALEC Competitiveness Index ranking correlated with--

State gross domestic product:  .02 (not statistically significant)
Percent change in nonfarm employment: -.09 (not statistically significant)
Percent change in per capita income: -.27 (statistically significant)
Percent change in state and local government revenue: -.16 (not statistically significant)
Percent change in median family income: -.30 (statistically significant)
Change in poverty rate: .21 ("statistically significant" at the .1 level*)

What this tells us is that the states which were following ALEC's preferred policies the most in 2007 saw worse performance in per capita income growth and median family income as well as a worse performance n poverty that we can almost be sure was not due to chance. The only thing ALEC's top states did see as predicted was an increase was in  population (Fisher et al. did not report the correlation coefficient, but their discussion makes it clear that it was statistically significant). However, population growth per se is not an economic outcome, as the report points out.

The concluding regression analysis weakens the case for negative consequences, but provides no support for positive effects of ALEC's state policies. Fisher et al. show that the most important determinants of 2007-11 GDP growth, employment growth, and per capita income growth are the components of a state's economy, with the strongest determinants being the presence of extractive industry (primarily due to the higher price of oil during this period) and a large health care industry. Once these are controlled for, none of the ALEC variables are statistically significant, though the closest is that the top personal income tax rate is associated with higher, not lower, per capita income growth.

If none of ALEC's policies work as advertised for job and income growth, what do they do? They are, in fact, a prescription for economic inequality. So-called "right to work" does not increase growth, but it reduces workers' bargaining power. Reducing taxes on the wealthy increases post-tax inequality. And so on, down the panoply of ALEC policies. Fisher et al. (p. 11) put it well:
The ALEC-Laffer strategies are exclusively those that would lower taxes on corporations and the wealthy, reduce public sector revenues (and hence public investments in education, health and infrastructure), and lower wages by eliminating minimum wages and weakening the bargaining power of workers. Yet the book claims that all of these measures would make states, and their populations, richer.

The report is Selling Snake Oil to the States, and that is precisely what ALEC's policies are.


* Technical note: I'm old school on when we should consider something probably not due to chance. For generations, the standard cutoff was that you have to be 95% certain a result was *not* due to chance to call it statistically significant. In economics, and now increasingly in political science, researchers have sometimes called a result statistically significant using a 90% cutoff instead. In my view, this shift has been due to what is called "publication bias": it is easier to get your study published in an academic journal if you have some statistically significant result. But this is a big problem in areas like minimum wage research, where not finding a statistically significant negative effect from increasing the minimum wage actually tells you a great deal. The key analysis of publication bias in minimum wage research can be found in David Card and Alan Krueger's book Myth and Measurement.

Cross-posted at Angry Bear.