St. Louisans Are Smarter Than the Post-Dispatch Implies
Last month, the St. Louis Post-Dispatch published an editorial critiquing the Show-Me Institute’s study of earnings taxes. It rejected both the type of study and its conclusions — which is unwarranted, on both counts. In taking issue with the model, the Post-Dispatch mentioned an alternate study as “proof” that the institute was wrong:
In 2005, the East-West Gateway Council of Governments actually asked area residents (what a research concept!) about their most important considerations in deciding where to live over the next eight to 10 years.
My graduate research methods class this semester allows me to apply some knowledge: The Post-Dispatch is referring to a qualitative research study, which is typically an initial model for a topic that has not been well studied, or something that cannot be captured with numbers. (One might conduct a qualitative study to get an idea of what needs to be studied more rigorously.) Qualitative studies are subjective, and are influenced by the researcher, the wording of questions, the population interviewed, and how honest the interviewee chooses to be.
Quantitative studies — manipulating data, like in an “abstract model” — are considered higher-tiered studies. That is not to say that quantitative studies are strictly better than qualitative; both have their purposes and their flaws. However, for something like a tax and housing decision, which involves both people choosing whether to live or not to live in an area, a model is better able to predict and explain phenomena. An interview study given to people living in St. Louis city excludes the data points of those people and businesses that have already lived there and chose to locate elsewhere — people that may be difficult to find.
There are some people who would choose to live in the city whether it has a 10-percent earnings tax or no earnings tax at all — but these are not the people at issue in an economic model. A model studies the people on the margin, the limited subset of people for whom that 1 percent is a tipping point between opportunity sets that otherwise have equal attractiveness. Some marginal group of people who weigh the pros and cons of each jurisdiction and find they add up to similar opportunities will choose to live or create a business in a suburb instead of the city, in order to capture that extra 1 percent. (And, although it may seem small, a 1-percent tax over the lifetime of a career adds up significantly.)
Another point of contention for the Post-Dispatch is the amount of money that the tax generates annually ($153.1 million in 2008). The article pointed out that this is similar to the size of the Metropolitan Police Department budget, and then conflated removing the earnings tax with getting rid of the police department — something that the Show-Me Institute never argued. At any rate, the Post-Dispatch seems unaware of our follow-up earnings tax study, which outlined a viable way to replace revenue lost from eliminating the earnings tax — a land value tax that would provide much less distortion of natural economic growth and development in the city, because of its fixed value.
St. Louis County and the entire metropolitan area is proportionally much larger than the city itself; in 2006, the city ranked 51st-largest in the United States, while the entire metropolitan area ranked 19th. People are not avoiding the St. Louis area in general, they are avoiding St. Louis city in particular — and with good reason. Good economists recognize that some people consider finances and tax rates to help them make rational decisions when choosing a residence or a business location. This confidence in people’s ability to weigh the incentives they face is well supported by actual data.