Tuesday, October 23, 2012

Talking In Circles About Business Climate

I’ve always figured that if we are going to talk about the “business climate” - and we seem to be doing a lot of that these days – we ought to know what the term means.

.We can all agree that the business climate is a basket of conditions that affect business performance, and that the more favorable the climate, the better business performance should be. But when it comes to choosing which conditions really make a difference, we’re all over the map: some folks think it’s all about taxes, others point to the quality of the labor force, still others to transportation costs, or the quality of public services, or lifestyle amenities or …well, the list goes on and on. And all too often, how business climate gets assessed has more to do with ideology than jobs, businesses and the economy.

I’ve been mulling this over recently because the Montana Policy Institute, a Bozeman based outfit that calls itself a “free market think tank,” just released a paper listing “business friendliness” scores for 25 Montana cities. The release of the paper was reported and apparently taken seriously by an admiring state press, and that sent me to my computer to find out whether the scores really mean something, or just sound like they do. Because there’s a pretty obvious way we can test the usefulness of any business climate rating, and that’s to determine if it’s actually correlated with business performance.  If it is, great. But if it isn’t, it’s not really telling us what we want to know.

Fortunately, I didn’t have to grub around for data to measure business performance, because the good people at MPI had already calculated an “economic vitality” score – based on median per capita income  and job and population growth – for all 25 cities in their study. So it was just a question of determining whether MPI’s business friendliness and economic vitality scores are correlated with each other. And it turns out they are. Here’s the picture:

Each point in the figure represents a city, and the upward drift in the points as you move to right means that more business friendly cities enjoy greater economic vitality. If you go to the trouble of computing the correlation coefficient (.42) you will find that the relationship between the scores is statistically significant. That means, in essence, that the relationship you see in the scatter diagram is real, and not just a fluke. So MPI has given us a business climate measure that really is related to business performance. Pretty impressive, no?

Well no, not really.

The problem is that MPI bases almost half of its business friendliness calculation on the economic vitality score itself. And that’s okay, as long as you like reasoning in circles. What produces good business performance? A good business climate. And what produces a good business climate? According to MPI, good business performance. So good business performance produces good business performance! True, but not very helpful.

None of this prevents MPI from concluding that cities can become more “business friendly” by “maintaining low tax rates on businesses.” But that has to be true, because MPI uses business taxes, along with business performance, in calculating its business friendliness score. In that case, cities that cut business taxes – which is what MPI, the free market think tank, devoutly hopes they will do - will inevitably raise their friendliness scores. That’s just arithmetic. The real question, however, is whether cutting taxes will improve business performance. Take a look at this next scatter diagram. It plots MPI’s economic vitality score against its tax burden score. Again, each point represents one city.

If you can’t see any relationship between tax levels and business performance in this figure, you’re not the only one. The correlation coefficient between these two variables is -.17, way too low to be statistically significant. In other words, the data just don’t allow us to conclude that lower taxes really lead to better business performance.

What has happened here is that MPI had dreamed up a measure of business climate that as a matter of simple arithmetic improves as business tax rates fall. Then they design their business climate measure so that it is spuriously correlated with business performance. And putting the two together they reach the false conclusion that reducing business taxes will improve business performance.

Somehow, the “think” part of “free market think tank” seems to have gone missing.