I’m generally curious about stated and revealed preferences. I’m even more curious about the ongoing narrative about how great business is in the New South/Sunbelt versus other states (since I live there). So I was naturally eager to explore some of the data behind CEO magazine’s recently released rankings of the best and worst states to do business. For reference, CEO surveyed 550 CEOs for their yearly ranking. The published results are here.
I’ve been thinking about whether the states with a supposedly more favorable business climate are actually seeing this translate into growth. I mean, if 550 CEOs in America decide that Texas is the best state in the union to do business (they did), then that must mean something for the Texas economny.
So in order to take a look, I went to the BEA and decided to grab the GDP data for the top 5 and bottom 5 states in the CEO ranking. I then took the average of the log of real GDP for each cohort of states. So let’s look at a few graphs, and then let’s think a little bit about what they (might) mean.
- North Carolina
47. New Jersey
49. New York
Forgive me for not being overly-attentive to good graph principles (namely the lack of axis labels). I was trying to be quick. I mean, blogging isn’t my day job, exactly. I will say that the Y-axis represents log real GDP.
A couple of things stand out here. One is that, in spite of years of being considered to have a relatively oppressive tax and regulatory regime, the worst 5 states still have on average “stronger” economies than the top 5 states. In a way this makes sense — it is largely states in the New South that are looking to formulate pro-growth policies after decades of economic (and social) stagnation that are leading the top 5. For the last half of the 20th century and still today, the Sunbelt as a whole continues to grow at a faster clip than other regions.
The drop bars are for reference, and confirm that in fact the log GDP’s are slowly but surely converging. Here’s a graph that provides better perspective on this. Note that since we are dealing with logarithms, each unit translates to exponential growth. So don’t read too much into the actual number.
Out of curiosity, I wondered how this relationship bore out when only private industries were counted. Here they are.
Ok so pretty much a wash. I would posit that public industries aren’t affecting the relationship significantly.
Ok, what does all of this mean? Yes, the states that the narrative suggests have a more favorable climate for business are slowly but surely closing the gap — their economies are indeed growing faster than the other states. We sort of knew that already, though, didn’t we? I intend to look at individual well-being through per capita GDP, but will have to follow up later.
What’s most interesting to me is that states like California continue to have massive, robust economies in the face of poor public policy, and political/regulatory uncertainty. The growth in the worst states for business, according to CEO, is not as robust, but it’s not exactly collapsing. With the exception of Michigan, the bottom 5 states all have large, strong economies — especially New York and California. To me, it seems that the disparity in promulgating prudential public policy and the actual growth disparity is a little incongruous. The narrative about these states doesn’t seem to translate 1:1 with economic reality.
One reason (I think) for this is network effects. Even though North Carolina has Research Triangle Park, California is a historic incubator of innovation. Silicon Valley remains the undisputed home for tech entrepreneurship. If you want to start the next Twitter, you NEED to be there. Physically. Not in spirit. Thus there is value for innovators to locate somewhere merely to be in physical proximity of other innovators. It would seem that effects like these can offset the negative effects of poor economic policy. Companies can find ways to pay higher taxes much more easily than finding a substitute for Silicon Valley.
This might be changing, however. Spatial distance is constant, but it is conceptually shrinking with the continued pace of information technology development. There is still not substitue for face to face contact, and there may never be, but there are increasingly desirable substitutes. Karl Smith at Modeled Behavior pondered something similar to this recently when considering exurbanism. Density becomes less obviously beneficial from a human relationships standpoint when we can achieve the same connections with people across longer distances.
One other thought Will Wilkinson posting over at Democracy in America got me chewing on: Maybe businessmen simply think that the top five states they chose are better for business because they tend to lean more conservative? There seems to be some evidence of cognitive bias on this issue among business leaders. And only two states in the bottom five have Republican governors — both leaders of largely liberal states and rarities in recent history. Compare that to the top five states, only one of which (North Carolina) is not led by a Republican. And they are all conservative — North Carolina’s legislature flipped Republican in both houses for the first time in a century this past fall.
Just some things to think about. I’m a libertarian, and so am inclined to support the rankings based on my knowledge of the respective states’ governance. The two questions we should be really asking though are 1) Are we overrepresenting the desirability disparity for doing business in these states? and if so 2) Are we cognitively inclined to be biased on this issue?