¹ See, for example, Winship, this volume.

² Whenever we refer to “government” we mean (as do the EFNA measures we use) all levels of government: federal, state, and local. For example, government spending in Texas includes all money spent by the federal government within the state of Texas, all money spent by the Texas state government, and all money spent by local levels of government within Texas.

³ For the complete list of the economic variables Fraser uses to construct the index, see Stansel, Dean and McMahon, Fred, Economic Freedom of North America (Vancouver, BC: The Fraser Institute, 2013).Google Scholar

⁴ While we do not report the results in this paper, we have applied our same analysis to the Mercatus Center’s Freedom in the 50 States index. The results are in general agreement with the results obtained using the EFNA index.

⁵ The Fraser Institute has calculated freedom indices for each of the fifty states annually from 1981 to 2009, for a total of 1,450 observations. Most of the socioeconomic outcomes are readily available from 1985 to the present, which results in most of our analyses being based on 1,250 observations from 1985–2009.

⁶ Data source: Ashby, , Bueno, , and McMahon, Fred, Economic Freedom of North America (Vancouver, BC: Fraser Institute, 2011).Google Scholar

⁷ The differences in the means are all statistically significant at the 1 percent level. For the twenty-five more free states over all years, mean = 7.20, s = 0.36, n = 625. For the twenty-five less free states over all years, mean = 6.42, s = 0.41, n = 625.

⁸ Over the period 1985 through 2009, fourteen states always appeared among the least economically free (white), thirteen states always appeared among the most economically free (black), and twenty-three states switched at least once between the least and most economically free (gray).

⁹ In a separate analysis, we controlled for climate, education, natural resources (as proxied by energy production), size of state populations relative to the national population, and state specific fixed effects. A cross-section fixed effects panel data OLS regression of median household income on EFNA, mean temperature, percentage of the population with a bachelors degree, energy production as a fraction of total U.S. energy production, state population as a fraction of total U.S. population, and change in the relative population yields the following results: Observations = 1000, R² = 0.85, D.W. = 2.10, EFNA coefficient = 2441.7, EFNA p-value = 0.000. Additional data sources: U.S. Department of Energy (www.eia.gov/beta/state/), Statistical Abstract of the United States, Table 233, NOAA National Climate Data Center, 2012 (www.currentresults.com/Weather/US/average-annual-state-temperatures.php).

¹⁰ Data source: U.S. Census Bureau, Historical Income Tables: Households, Table H-8 (www.census.gov/hhes/www/income/data/historical/household); Ashby, Bueno, and McMahon, Economic Freedom of North America. 2011.

¹¹ The difference in the means is statistically significant at the 1 percent level. For the more free states, mean = $38,026, s = $9,768, n = 625. For the less free states, mean = $35,348, s = $10,025, n = 625.

¹² In a separate analysis, we controlled for climate, education, natural resources (as proxied by energy production), size of state populations relative to the national population, and state specific fixed effects. A cross-section fixed effects panel data OLS regression of per-capita personal income on EFNA, mean temperature, percentage of the population with a bachelors degree, energy production as a fraction of total U.S. energy production, state population as a fraction of total U.S. population, and change in the relative population yields the following results: Observations = 1000, R² = 0.89, D.W. = 2.00, EFNA coefficient = 1338.5, EFNA p-value = 0.000. Additional data sources: U.S. Department of Energy (www.eia.gov/beta/state/), 2012 Statistical Abstract of the United States, Table 233, NOAA National Climate Data Center (www.currentresults.com/Weather/US/average-annual-state-temperatures.php).

¹³ Data source: Census Bureau, Statistical Abstract of the United States, 1995 Table 713;1996 Table 699; 1997 Table 706; 1998 Table 727; 1999 Table 733; 2000 Table 727; 2001 Table 652; 2002 Table 643; 2003 Table 671; 2004–2005 Table 653; 2006 Table 663; 2007 Table 660; 2008 Table 660; 2009 Table 660; 2010 Table 666; 2011 Table 680; 2012 Table 681; Ashby, Bueno, and McMahon, Economic Freedom of North America, 2011.

¹⁴ The difference in the means is statistically significant at the 1 percent level. For the more free states, mean = $24,985, s = $7,349, n = 575. For the less free states, mean = $23,307, s = $7,044, n = 575. The difference in the means is statistically significant at the 1 percent level. For the more free states, mean = 0.0517, s = 0.0169, n = 625. For the less free states, mean = 0.0580, s = 0.0173, n = 625.

¹⁵ In a separate analysis, we controlled for climate, education, natural resources (as proxied by energy production), size of state populations relative to the national population, and state specific fixed effects. A cross-section fixed effects panel data OLS regression of unemployment on EFNA, mean temperature, percentage of the population with a bachelors degree, energy production as a fraction of total U.S. energy production, state population as a fraction of total U.S. population, and change in the relative population yields the following results: Observations = 1000, R² = 0.55, D.W. = 2.20, EFNA coefficient = −0.007, EFNA p-value = 0.000. Additional data sources: U.S. Department of Energy (www.eia.gov/beta/state/), 2012 Statistical Abstract of the United States, Table 233, NOAA National Climate Data Center (www.currentresults.com/Weather/US/average-annual-state-temperatures.php).

¹⁶ The difference in the means is statistically significant at the 1 percent level. For the more free states, mean = 0.0517, s = 0.0169, n = 625. For the less free states, mean = 0.0580, s = 0.0173, n = 625.

¹⁷ Data source: Census Bureau, Statistical Abstract of the United States, 1995 Table 664; 1996 Table 650; 1997 Table 658; 1998 Table 683; 1999 Table 686; 2000 Table 680; 2001 Table 606; 2002 Table 600; 2003 Table 628; 2004–2005 Table 610; 2010 Table 628; 2011 Table 628; Ashby and McMahon, Economic Freedom of North America, 2011.

¹⁸ U.S. Bureau of Labor Statistics (BLS): Current Population Survey (CPS) [Household Survey — LNS11000000]. See www.census.gov/hhes/www/poverty/about/overview/measure.html. For poverty thresholds, see www.census.gov/hhes/www/poverty/data/threshld/index.html.

¹⁹ In a separate analysis, we controlled for climate, education, natural resources (as proxied by energy production), size of state populations relative to the national population, and state specific fixed effects. A cross-section fixed effects panel data OLS regression of the poverty rate on EFNA, mean temperature, percentage of the population with a bachelors degree, energy production as a fraction of total U.S. energy production, state population as a fraction of total U.S. population, and change in the relative population yields the following results: Observations = 950, R² = 0.59, D.W. = 1.94, EFNA coefficient = −0.021, EFNA p-value = 0.000). Additional data sources: U.S. Department of Energy (www.eia.gov/beta/state/), 2012 Statistical Abstract of the United States, Table 233, NOAA National Climate Data Center (www.currentresults.com/Weather/US/average-annual-state-temperatures.php).

²⁰ See www.census.gov/hhes/www/poverty/about/overview/measure.html. For poverty thresholds, see www.census.gov/hhes/www/poverty/data/threshld/index.html.

²¹ Data source: Census Bureau, Statistical Abstract of the United States, 1992 Table 723; 1995 Table 749; 1996 Table 735; 1997 Table 741; 1998 Table 761; 1999 Table 765; 2000 Table 759; 2001 Table 684; 2002 Table 673; 2003 Table 705; 2004–2005 Table 688; 2006 Table 692; 2007 Table 690; 2008 Table 690; 2009 Table 687; 2010 Table 693; 2011 Table 708; 2012 Table 708; Ashby, Bueno, and McMahon, Economic Freedom of North America, 2011.

²² The difference in the means is statistically significant at the 1 percent level. For the more free states, mean = 0.119, s = 0.034, n = 575. For the less free states, mean = 0.136, s = 0.038, n = 575.

²³ In a separate analysis, we controlled for climate, education, natural resources (as proxied by energy production), size of state populations relative to the national population, and state specific fixed effects. A cross-section fixed effects panel data OLS regression of the Gini coefficient on EFNA, mean temperature, percentage of the population with a bachelors degree, energy production as a fraction of total U.S. energy production, state population as a fraction of total U.S. population, and change in the relative population yields the following results: Observations = 200, R² = 0.56, D.W. = 1.93, EFNA coefficient = −0.004, EFNA p-value = 0.062. Additional data sources: U.S. Department of Energy (www.eia.gov/beta/state/), 2012 Statistical Abstract of the United States, Table 233, NOAA National Climate Data Center (www.currentresults.com/Weather/US/average-annual-state-temperatures.php).

²⁴ The American Community Survey began recording Gini coefficients at the state level in 2006.

²⁵ The difference in the means is not statistically significant. For the more free states, mean = 0.447, s = 0.020, n = 100. For the less free states, mean = 0.450, s = 0.019, n = 100.

²⁶ Data source: Census Bureau, American Community Survey, B19083 (factfinder2.census.gov/faces/nav/jsf/pages/searchresults.xhtml?refresh=t#none); Ashby , Bueno, and McMahon, Economic Freedom of North America, 2011.

²⁷ See Winship’s Section III in his essay in this volume for a more detailed discussion of economic mobility.

²⁸ In a separate analysis, we controlled for climate, education, natural resources (as proxied by energy production), size of state populations relative to the national population, and state specific fixed effects. A cross-section fixed effects panel data OLS regression of the uninsured rate on EFNA, mean temperature, percentage of the population with a bachelors degree, energy production as a fraction of total U.S. energy production, state population as a fraction of total U.S. population, and change in the relative population yields the following results: Observations = 750, R² = 0.45, D.W. = 1.83, EFNA coefficient = −0.009, EFNA p-value = 0.000. Additional data sources: U.S. Department of Energy (www.eia.gov/beta/state/), 2012 Statistical Abstract of the United States, Table 233, NOAA National Climate Data Center (www.currentresults.com/Weather/US/average-annual-state-temperatures.php).

²⁹ Data source: Census Bureau, Statistical Abstract of the United States, 1997 Table 172; 1998 Table 186; 1999 Table 190; 2000 Table 178; 2001 Table 145; 2002 Table 138; 2003 Table 153; 2004–2005 Table 140; 2006 Table 143; 2007 Table 145; 2008 Table 147; 2009 Table 147; 2010 Table 150; 2011 Table 152; 2012 Table 156; Ashby, Bueno, and McMahon, Economic Freedom of North America 2011.

³⁰ The difference in the means is not significant. For the more free states, mean = 0.140, s = 0.040, n = 375. For the less free states, mean = 0.144, s = 0.039, n = 375.

³¹ In a separate analysis, we controlled for climate, education, natural resources (as proxied by energy production), size of state populations relative to the national population, and state specific fixed effects. A cross-section fixed effects panel data OLS regression of population growth on EFNA, mean temperature, percentage of the population with a bachelors degree, energy production as a fraction of total U.S. energy production, state population as a fraction of total U.S. population, and change in the relative population yields the following results: Observations = 1000, R² = 0.21, D.W. = 2.09, EFNA coefficient = 0.006, EFNA p-value = 0.000. Additional data sources: U.S. Department of Energy (www.eia.gov/beta/state/), 2012 Statistical Abstract of the United States, Table 233, NOAA National Climate Data Center (www.currentresults.com/Weather/US/average-annual-state-temperatures.php).

³² The difference in the means is statistically significant at the 1 percent level. For the more free states, mean = 0.013, s = 0.012, n = 575. For the less free states, mean = 0.008, s = 0.009, n = 575.

³³ Data source: Census Bureau, PE-5, compiled by Moody Analytics (www.economy.com); Ashby, Bueno, and McMahon, Economic Freedom of North America, 2011.

³⁴ Data source: IRS, SOI Tax Stats (www.irs.gov/taxstats/article/0, id=212702,00.html); Ashby, Bueno, and McMahon, Economic Freedom of North America, 2011.

³⁵ The difference in the means is statistically significant at the 5 percent level. For freedom ratios above the median, mean = 1.031, s = 0.374, n = 2,448. For freedom ratios below the median, mean = 1.012, s = 0.268, n = 2,452.

³⁶ Government Finance and Employment Classification Manual, Census Bureau, 2006 (www2.census.gov/govs/pubs/classification/2006_classification_manual.pdf).

³⁷ The difference in the means is statistically significant at the 1 percent level. For the more free states, mean = 0.153, s = 0.022, n = 425. For the less free states, mean = 0.187, s = 0.032, n = 425.

³⁸ The difference in the means is statistically significant at the 1 percent level. For the more free states, mean = 0.182, s = 0.025, n = 425. For the less free states, mean = 0.218, s = 0.043, n = 425.

³⁹ Data source: Census Bureau, Census of Governments, 1997, 2002, 2007 (www.census.gov//govs/cog/); Bureau of Economic Analysis, GDP and Personal Income Regional Data (www.bea.gov/iTable/iTable.cfm?ReqID=70&step=1); Ashby, Bueno, and McMahon, Economic Freedom of North America, 2011. Revenue data from missing years is interpolated.

⁴⁰ In a separate analysis, we controlled for climate, education, natural resources (as proxied by energy production), size of state populations relative to the national population, and state specific fixed effects. A cross-section fixed effects panel data OLS regression of state and local debt as a fraction of state GDP on EFNA, mean temperature, percentage of the population with a bachelors degree, energy production as a fraction of total U.S. energy production, state population as a fraction of total U.S. population, and change in the relative population yields the following results: Observations = 850, R² = 0.24, D.W. = 1.86, EFNA coefficient = −0.023, EFNA p-value = 0.000. Additional data sources: U.S. Department of Energy (www.eia.gov/beta/state/), 2012 Statistical Abstract of the United States, Table 233, NOAA National Climate Data Center (www.currentresults.com/Weather/US/average-annual-state-temperatures.php).

⁴¹ The difference in the means is statistically significant at the one percent level. For the more free states, mean = 0.147, s = 0.037, n = 425. For the less free states, mean = 0.165, s = 0.044, n = 425.

⁴² Data source: Bureau of Economic Analysis, GDP and Personal Income Regional Data (www.bea.gov/iTable/iTable.cfm?ReqID=70&step=1); Ashby, Bueno, and McMahon, Economic Freedom of North America, 2011. Revenue data from missing years is interpolated.

⁴³ In a separate analysis, we controlled for climate, education, natural resources (as proxied by energy production), size of state populations relative to the national population, and state specific fixed effects. A cross-section fixed effects panel data OLS regression of the growth in the number of firms on EFNA, mean temperature, percentage of the population with a bachelors degree, energy production as a fraction of total U.S. energy production, state population as a fraction of total U.S. population, and change in the relative population yields the following results: Observations = 650, R² = 0.15, D.W. = 2.09, EFNA coefficient = 5.892, EFNA p-value = 0.000. Additional data sources: U.S. Department of Energy (www.eia.gov/beta/state/), 2012 Statistical Abstract of the United States, Table 233, NOAA National Climate Data Center (www.currentresults.com/Weather/US/average-annual-state-temperatures.php).

⁴⁴ The difference in the means is statistically significant at the one percent level. For the more free states, mean = 37.038, s = 124.918, n = 175. For the less free states, mean = 16.541, s = 26.548, n = 175.

⁴⁵ Data source: Census Bureau, Statistical Abstract of the United States, 1989 Table 864; 1990 Table 864; 1991 Table 864; 1992 Table 846; 1993 Table 862; 1994 Table 847; 1998 Table 878; 1999 Table 885; 2000 Table 876; 2006 Table 745; 2007 Table 744; 2008 Table 741; 2009 Table 742; 2010 Table 748; 2011 Table 766; Ashby, Bueno, and McMahon, Economic Freedom of North America, 2011.

⁴⁶ Data source: Census Bureau, Statistical Abstract of the United States, 1989 Table 864; 1990 Table 864; 1991 Table 864; 1992 Table 846; 1993 Table 862; 1994 Table 847; 1998 Table 878; 1999 Table 885; 2000 Table 876; 2006 Table 745; 2007 Table 744; 2008 Table 741; 2009 Table 742; 2010 Table 748; 2011 Table 766; Ashby, Bueno, and McMahon, Economic Freedom of North America, 2011.

⁴⁷ The difference in the means is statistically significant at the 10 percent level. For the more free states, mean = 1.714, s = 1.971, n = 250. For the less free states, mean = 1.439, s = 1.149, n = 250.

⁴⁸ In a separate analysis, we controlled for climate, education, natural resources (as proxied by energy production), size of state populations relative to the national population, and state specific fixed effects. A cross-section fixed effects panel data OLS regression of state GDP growth on EFNA, mean temperature, percentage of the population with a bachelors degree, energy production as a fraction of total U.S. energy production, state population as a fraction of total U.S. population, and change in the relative population yields the following results: Observations = 1000, R² = 0.43, D.W. = 2.14, EFNA coefficient = 0.009, EFNA p-value = 0.000. Additional data sources: U.S. Department of Energy (www.eia.gov/beta/state/), 2012 Statistical Abstract of the United States, Table 233, NOAA National Climate Data Center (www.currentresults.com/Weather/US/average-annual-state-temperatures.php).

⁴⁹ The difference in the means is statistically significant at the 1 percent level. For the more free states, mean = 0.058, s = 0.034, n = 600. For the less free states, mean = 0.048, s = 0.036, n = 600.

⁵⁰ Data source: Bureau of Economic Analysis, GDP and Personal Income Regional Data (www.bea.gov/iTable/iTable.cfm?ReqID=70&step=1); Ashby, Bueno, and McMahon, Economic Freedom of North America, 2011. Revenue data from missing years are interpolated.