Counterintuitively, some data I’ve seen recently seems to suggest that the answer to the above question may be no.

In recent days, I’ve had the chance to look a bit at two different pieces of Jason Sorens’ work, which I have not yet seen anyone else specifically connect to one another. If anyone has, I’d be interested to see their take on it. Jason Sorens, for those who are not familiar with the name, is the founder of the Free State Project, an organization based on the premise that herding a critical mass of liberty-minded people into a common geographic and political realm will lead to pro-liberty political change.

In addition to be the founder of the Free State Project, Sorens is also a co-author of Freedom in the Fifty States as well as having put forth a statistical analysis of which states have the most libertarians.

Freedom in the Fifty States attempts to rank each of the states from most to least free, resulting in this handy graphic:

Freedom in the 50 States 2013 | Mercatus Center

Freedom in the 50 States 2013 | Mercatus Center

Sorens explains his method for trying to approximate the number of libertarians in each of the fifty states and provides the data he came up with in his blog post here. Reason has compiled a handy graphic reflecting the data:

An engineer by training and previous profession, I was thrilled to see some data by which we can do a sniff test of the hypothesis of the Free State Project experiment— namely that more libertarians means more liberty. Looking at the above data, though, that hypothesis seems somewhat questionable.

Looking at the top 5 most free states according to the Freedom in the Fifty States index:

  • North Dakota and South Dakota are #1 on the Freedom in the Fifty States index, but have only the 11th and 18th most libertarians per capita, respectively.
  • Tennessee is #3 on the Freedom in the Fifty States index, but has only the 26th most libertarians per capita.
  • New Hampshire has the second most libertarians per capita, but is falling on the freedom index, dropping from #1 to #4 as more self-described libertarians have moved there and gained office there.
  • Oklahoma is the #5 most free state according the index, but is at the back of the pack at #47 for number of libertarians per capita.


Looking at the five states with the most libertarians per capita, according to Sorens’ analysis:

  • Montana has the most libertarians per capita, but is #12 and falling on the Freedom in the Fifty States index.
  • New Hampshire, as previously mentioned, has the second most libertarians per capita, but is falling on the freedom index, dropping from #1 to #4 as more self-described libertarians have moved there and gained office there.
  • The #5 state for most libertarians, New Mexico, is a middling #21 on the Freedom in the Fifty States index.

The only three of the top five on either list that seem comparable (and even North Dakota and Alaska are somewhat of a stretch) on the two lists are North Dakota (#1 most free state, #11 most libertarians per capita), Alaska (#3 most libertarians per capita, #13 most free state), and Idaho (#5 most libertarians, #6 most free).

Looking at the bottom 5 most free states according to the Freedom in the Fifty States index:

  • Rhode Island ranks #46 for liberty, but ranks much higher in libertarians per capita up at #25.
  • #47 most free state Hawaii is not ranked on the list of number of libertarians per capita by state due to lack of data.
  • California is the second to last most free state, but ranks almost into the first quartile at #13 for most libertarians per capita.
  • #48 most free state New Jersey and the least free state New York come in the closest to their counterparts, with New Jersey at #44 and New York at #40 for libertarians per capita.


Looking at the five states with the least libertarians per capita, according to Sorens’ analysis:

  • As mentioned earlier, Oklahoma is at the back of the pack at #47 for number of libertarians per capita, but is the #5 most free state according the index.
  • Alabama ranks #46 for libertarians per capita, but comes in above the median at #18 most free state.
  • Connecticut, West Virginia, and Mississippi, numbers 45, 48, and 49 respectively for libertarians per capita, all are reasonably comparable, ranking in the bottom 10 for most free state at numbers 40, 42, and 41 respectively.

Then I did a quick plot of my own, charting the two ranking systems against one another. I left out D.C. and stuck to just the 50 states, of which data for libertarians per capita was missing for two states. Nonetheless, I charted all the states in order of Freedom in the Fifty States index ranking, to get a nice linear plot of those, represented by the red line on the chart. (Note that this line was drawn on rather than charted and may be slightly off as I didn’t want to spend a lot of time trying to figure out how to get a Google Sheet to plot two different data sets on one chart in different formats. The point I’m about to make, however, is not changed by that.) The blue dots represent the rank for libertarians per capita of each state for which data was available.

FreedomRankLibsPerCapita

If libertarians per capita correlates well with liberty, we might expect the rank for libertarians per capita to track somewhat closely with the line for liberty rank, but instead what we see is highly scattered data. This is not encouraging.

What do we take away from this? Are these metrics flawed in how they are measured or defined? Would plotting the actual data rather than the ranking in each category yield a more encouraging picture? I could see how ranking every state in equal increments could potentially skew the picture when the actual figures of merit are not equally spaced- in other words, the difference between State A and State B could be one place in ranking, but 10 times the actual difference between States B and C which also differ by one place in ranking. Does it? Does having libertarians in a particular state not actually have very much to do with liberty in that state? Or… how else might this apparent disconnect be explained?

I would be particularly interested to know if Sorens has published anything that resolves or even addresses without resolving this apparent disparity.