Ranking & Discussion
Ranking & Discussion
By summing the economic freedom and personal freedom scores, we obtain the overall freedom index, presented in table 5. New Hampshire and South Dakota again find themselves in a virtual tie for first. Under unified Democratic governance between January 2007 and January 2009, New Hampshire increased on personal freedom (+0.064) while falling slightly on economic freedom (–0.012), despite a smoking ban. South Dakota improved dramatically on fiscal policy (+0.103), where it was already strong. The next tier of states consists of Indiana, Idaho, Missouri, and Nevada, all roughly tied.
On the other hand, many states perform quite poorly in providing a liberty-friendly environment for their citizens. New York is the least free by a considerable margin. This will surprise few residents of the Empire state. In order from the bottom, New York is followed by New Jersey, California, Hawaii, Massachusetts, and Rhode Island. Unfortunately, these states make up a substantial portion of the total American population. Moreover, these bottom six states have considerable ground to make up even to move off this ignoble list, let alone into a creditable position in the rankings. Individual state profiles provide more information about each state.
Some states improved over the two years since the last study, while others fell significantly. Table 6 lists the five most improved states and five states that slipped the most in terms of freedom. Interestingly, average freedom in the 50 states basically held steady between 2007 and 2009. Oregon takes the prize for the most improved state, due mostly to a big improvement in the quality of its court system, a substantial decline in tax collections (from 9.7 to 8.8 percent of personal income), and the enactment of same-sex civil unions. Wyoming and California vie for the most worsened state, in both cases due partly to deteriorating fiscal situations. In California, almost every area of fiscal policy worsened between 2007 and 2009. California also enacted new regulations, such as mandated paid family leave. In Wyoming, fiscal policy appeared to deteriorate because falling energy prices led to a fall in personal income in this energyrich state, the denominator in the fiscal-policy variables. Nevertheless, Wyoming, which was sixth in overall freedom in 2007 and fell to twenty-first in 2009, remains in much better shape than most states.
Figure 1 is a scatter plot of state economic- and personal-freedom scores. The common libertarian conception about the political spectrum is that leftliberals sit in the upper left corner of this diagram, favoring extensive personal freedom and little economic freedom, while right-conservatives belong in the bottom right corner, favoring economic but not personal freedom. It should be clear from this chart that the truth is much different. Highly liberal states (New York, New Jersey, Rhode Island, Hawaii, Maryland, Massachusetts, and California) seem to cluster much more in the bottom left where both economy and personal life are more regulated. Vermont, Washington, New Mexico, Maine, Oregon, and Alaska seem to conform more to the standard paradigm of left-liberalism. Conservative states divide by region: Southern and Midwestern states (South Carolina, Alabama, Tennessee, Iowa, and the Dakotas) tend to score lower on personal freedom than Western states (Nevada, Idaho, Colorado, and Texas), but there are plenty of exceptions (Mississippi, Indiana, North Carolina, and Missouri are high on personal freedom, while Oklahoma and Montana are mediocre).
Figures 2–4 show the relationship between state percentage of the vote for the Democratic and Green presidential tickets in 2008 and state scores on economic, personal, and overall freedom.21 Democratic/ Green presidential vote share is a rough measure of citizen-opinion liberalism, and these charts allow us to examine the relationship between left–right ideology at the voter level and state policy orientation toward individual freedom. To reiterate previous observations, the relationship between ideology and personal freedom is nearly flat, reflecting the propensity of liberal and conservative states to protect certain freedoms but not others. The relationship between liberalism and economic freedom is more strongly negative, and as a result the relationship between liberalism and overall freedom is modestly negative, but only among the most liberal states. In short, moderate states are no less or more free than conservative states, but liberal states do tend to be less free, particularly on economic issues.
Table 7 provides economic, personal, and overall freedom scores by census division.22 The Mountain division comes in first on personal freedom, while West North Central comes in first both overall and on economic freedom. The Middle Atlantic (New York, New Jersey, and Pennsylvania) is the worst on both economic and personal freedom. The Pacific (which includes Alaska and Hawaii) does very badly on economic freedom but rather well on personal freedom. Southern states do a bit better the more westerly they are, particularly on personal freedom.
Figure 5 maps the states on freedom, rendering the regional patterns visually clearer. The bastions of freedom appear to lie mostly in the heartland, the country lying between the Mississippi and the Rockies. New Mexico is an unusually poor state in this region, while the relative freedom of New Hampshire, Indiana, and Virginia stands out in stark contrast to their neighboring states.
We regressed net internal migration by state from 2000 to 2009 as a percentage of 2000 state population on total freedom scores in 2007 and mean January temperature in each state’s largest city to determine whether freer states tend to attract more people and less free states tend to repel them.23 Both independent variables are statistically significant and positive at the 99 percent confidence level.24 Substantively, the results show that an increase of 0.5 points on the freedom scale, for instance from Connecticut to Iowa, increases net migration by a whopping 5.9 percentage points of 2000 population. This is about twice as large as the effect of a change in mean January temperature of 20 degrees, for instance from Chicago, Illinois, to Birmingham, Alabama. When we regress migration on January temperature, economic freedom, and personal freedom, all variables are again significant at the 99 percent level, and the positive effect of personal freedom is actually more than twice as large as that of economic freedom. A 0.25-unit increase in personal freedom increases migration by 5.0 percentage points of 2000 population, while the comparable effect for economic freedom is 2.4 percentage points (but there is less observed variance across the states on personal than economic freedom). Of course, these results present correlations, which do not necessarily indicate causation, but it seems implausible to us that migration would affect freedom, especially since policy change happens slowly. The most plausible interpretation is that Americans in general are indeed attracted to freedom for its own sake, not just for the economic benefits.
To estimate those economic benefits, we also regressed annualized growth rates in total personal income from 2000 to 2008 on 2007 economic and personal freedom, state and local severance taxes as a percentage of personal income in 2000 (to proxy the state economy’s dependence on mineral extraction), the log of initial per-capita personal income, mean January temperature in the largest city, and an index of educational attainment from Morgan Quitno Press.25 Education, January temperatures, and severance revenues were positively related to income growth and initial per-capita income was negatively related. Economic freedom was associated with more income growth (statistically significant at the 95 percent confidence level), but personal freedom was not. The results show that a 0.25-unit increase in economic freedom increases the average annual growth rate in personal income by about 0.25 percentage points.
If Americans generally prefer freedom as we have measured it, how did some states come to restrict freedom to such a degree? Perhaps the most regulated states on our index have been responding more to interestgroup pressures and politicians’ self-interest than to citizens’ most strongly held preferences.
21. The lines represent the best polynomial fit in two degrees, and the shaded areas are the 95 percent confidence intervals.
22. For the U.S. census divisions, see http://www.census.gov/geo/www/us_regdiv.pdf, accessed July 30, 2008.
23. The source for net internal migration is the U.S. Census Bureau; the source for mean January temperature in each state’s largest city is the National Oceanic and Atmospheric Administration.
24. We ran the regressions with and without robust standard errors with no difference in the results reported.
25. “Results of the 2006 Smartest State Award” (Lawrence, KS: Morgan Quitno Press, 2007), http://www.morganquitno.com/edrank.htm, accessed December 5, 2010