Many think tanks produce dashboards of economic indicators to help government officials formulate public policy. While these tools are almost always well-intentioned, sometimes they aren’t well thought out, making them problematic to follow. One example is State New Economy Index, produced by the Information Technology and Innovation Foundation and the Ewing Marion Kauffman Foundation.
This index is designed to provide policy makers with a set of 26 measures to guide efforts to move states to the “new economy,” which the two foundations say is “knowledge-based, globalized, entrepreneurial, IT-driven, and innovation-based.”
The effort is flawed because the designers of the dashboard combine uncorrelated and negatively correlated measures to create overarching indicators. Because combinations of unrelated measures aren’t indicators of anything, the dashboard isn’t useful.
For those who find this point too academic to follow, let me give an example to clarify what I mean. The report on the index says that to adjust to the new economy, states need more “economic dynamism” and offers several measures of what more dynamic places look like. The authors explain that states with a lot of “job churn” (a lot of businesses starting and failing); more “fast growing firms” (a high share of Inc 500 and Deloitte Technology Fast 500 firms); higher value of initial public offerings as a share of worker earnings; and a larger fraction of the population starting businesses (adjusted for how fast the state has been growing), have more economic dynamism, which makes them more successful in the new economy.
At first glance, the economic dynamism measure seems useful. It says that a state needs a lot of people starting businesses, more businesses starting and failing, more high growth companies, and more initial public offerings, to be successful in the new economy.
The problem appears when we look at the measures of economic dynamism. Several of them don’t move in concert. Across states, the job churn measure correlates only 0.03 with the fast growing firms measure and -0.01 with the IPOs measure. This means that states that are high on job churn don’t have a lot of fast growing firms or IPOs. Similarly, the measure of entrepreneurial activity doesn’t correlate very highly with the measure of fast growing firms(0.13) or IPOs (0.11). That is, states with a high share of the population starting businesses don’t have a lot of high growth firms or IPOs.
The job churn measure does correlate reasonably well (0.51) with the indicator of entrepreneurial activity. States that have more new firms starting and failing also tend to have a higher share of their population starting businesses, and vice versa.
If we look at a measure that isn’t part of the economic dynamism index, venture capital – the amount of venture capital invested in the state as a percentage of earnings of workers in the state – the nature of the problem becomes even clearer. The job churn indicator correlates only -0.07 with the venture capital measure and only 0.16 with the indicator of entrepreneurial activity. States that have a lot of businesses starting and failing and a higher share of the population starting businesses don’t have a lot of venture capital.
Which states have a lot of venture capital? The ones with a lot of IPOs (the correlation between the measures of venture capital and IPOs is 0.64) and fast growing firms (the correlation between the indicators of venture capital and fast growing firms is 0.45).
Together these measures show states that have a lot of IPOs also have a lot of venture capital and fast growing firms and states that have a lot of job churn also have a lot of entrepreneurial activity. But states that are high on the second set of factors aren’t high on the first set.
This pattern suggests an important policy issue that is obscured by the State New Economy Index: Whatever factors give states a lot of venture capital, IPOs and fast growing firms are different from those that give states lots of people starting and failing at business creation.
Government officials can’t encourage everything and often have to choose one policy to promote at the expense of another. Given the data pattern described above, which alternative would you hope your state’s leaders would choose: policies that generate more venture capital, IPOs, and fast growing firms or policies that stimulate a lot of new business starts and stops?
Many of us would prefer the former. And that’s where the harm comes from New Economy Index. It obscures the difference between states that have a lot of high growth entrepreneurial activity and states that have a lot of high volume entrepreneurial activity. This lack of clarity leads policy makers to believe that they can get more high growth entrepreneurship by getting more high volume entrepreneurship. Unfortunately, places appear to be strong in just one or the other.