In deciding how to weight personal freedoms, we started from the bottom up, beginning with the freedom we saw as least important in terms of saliency, constitutional implications, and the number of people affected, and working up to the most important. For us, gaming and gambling laws fall at the bottom. All states—even Nevada—regulate gaming to some extent, and while gaming is a popular leisure activity among Americans, it is hardly critical to the foundations of the Republic. Gaming laws are worth 6.2 percent of the overall personal freedom index. Within this subcategory, we rated state and local gaming revenues as a percentage of personal income as threetenths (the only type of tax that gets a positive score in our index!), each type of gaming permitted as onetwentieth, whether “aggravated gambling” is a felony as one-fifth, legalization of Internet gaming as one-sixth, and laws prohibiting social gaming—which are prima facie highly intrusive but we suppose almost never enforced—as one-thirtieth of the total.

The next-lowest-weighted subcategory in paternalism is alcohol regulation, worth 50 percent more than gaming laws. Because of the more apparent ubiquity of alcohol consumption, we weighted regulations that make it more expensive to consume alcohol higher than those that restrict gambling.14 Alcohol regulations include state control of alcohol distribution (an index of wholesale and retail control of beer, wine, and spirits), worth just under one-third of the total; taxes on beer, wine, and spirits, worth just over two-fifths; “blue laws” that prohibit Sunday sales, worth just over one-sixth; and keg regulations (registration/bans), mandatory server training, and happy-hour laws worth the remaining one-tenth.

The next tier from the bottom consists of auto and road regulations and tobacco regulations, each worth 12 percent of personal freedom. Libertarians generally support rules of the road that facilitate optimal traffic flow and prohibit reckless and intoxicated drivers from imposing risks of harm on others, but they oppose laws that punish private behavior that does not violate the rights of others. For that reason, seatbelt laws and sobriety checkpoints count as notable infringements on individual liberty. Bicycle- and motorcycle-helmet laws are also restrictions that impose undue costs (though any harm that comes to people from not wearing helmets should be fully internalized by the individuals in question). Mandatory automotive personal-injury and underinsured driver insurance are paternalistic restrictions on individual choice. Open-container bans are included as a minor nuisance (we support drunkdriving laws, however), and cell-phone driving bans are also included. In theory, cell-phone driving bans could be justified on libertarian grounds as necessary for public safety, but it is better to subsume such bans under a general “distracted driving” penalty, as some states have, which we do not code as a restriction on freedom. It is also worth noting that research disputes the effect of cell-phone usage on crash rates,15 and research also shows that text-messaging bans do not reduce crashes, probably because they are difficult to enforce.16

Tobacco laws receive a higher weight than alcohol regulations because they tend to go much further. The taxes are higher and the limitations on consumption are much more onerous. Smoking bans make up over half of this subcategory, with cigarette taxes making up most of the rest. Regulations on vending machines and Internet purchases round out this subcategory.

Campaign-finance regulations are worth just slightly more, at 12.9 percent of personal freedom. We gave this fairly technical area a relatively high weight because of these laws’ First Amendment implications. Of primary importance are regulations on individual donations to candidates and parties and grassroots political action committee (PAC) donations to candidates and parties. Secondarily, we consider restrictions on corporate and union contributions, but these receive just over half the weight of the other restrictions for two reasons. First, we suspect that corporations and unions are often lobbying for an agenda that restricts freedom in some way. Second, corporations may even prefer restrictions on what they can give to candidates so that politicians cannot shake them down for more funds. Finally, a public financing of elections index is worth almost one-fifth of the subcategory (see appendix for coding details).

At the next level we have asset-forfeiture rules. Asset forfeiture, when perpetrated without a conviction of the owner, is an egregious violation of both property rights and the Fourth Amendment. Unfortunately, only a minority of states have reformed assetforfeiture rules to put the burden of proof on the government and require owner involvement in criminal activity for forfeiture.

Marriage and civil-union laws and arrests for victimless crimes are each worth 16.2 percent of personal freedom. We have measured arrests for victimless crimes in three ways. The first focuses on drug arrests, which constitute the majority of victimlesscrimes arrests. The drug-arrest index divides per capita, over-18 drug arrests by the ratio of the over-18 population that reports having used drugs in the past month to examine how ferociously drug laws are enforced. The other two variables use arrests of adults over 18 for liquor laws, gambling, and prostitution in the numerator. One measure uses state population in the denominator, while the other uses total arrests. The first variable thus measures the likelihood of being arrested for engaging in a victimless crime, while the second captures the focus of police in the state: are they going after real criminals or just people engaged in vice? The latter two variables are equally weighted and, combined, equal the weight of the drug-arrests index.

Marriage and civil-union laws are coded equally with arrests for victimless crimes because of the high salience of the issue. State attempts to enhance the ability of same-sex partners to make voluntary contracts that affect life or death decisions unequivocally enhance individual liberty. (One could argue that states should get the government out of marriage licensing altogether and offer streamlined “life-partnership contracts” to all sorts of families and households, not just heterosexual and homosexual two-partner relationships.) Our main variable in this subcategory simply indicates whether the state recognizes some form of domestic partnership, civil union, or marriage for same-sex couples (states do not get extra points for moving up that scale). Also in this subcategory, although worth far less in the index, are blood-test requirements and marriagelicense waiting periods, which libertarians would deem unnecessary for consenting adults.

The mala prohibita and civil liberties subcategory is a miscellany of paternalistic additions to the criminallaw and civil-liberty issues. Almost a third of the whole subcategory has to do with the high-salience issue of legalized prostitution, which only Nevada and, to a lesser extent, Rhode Island (where brothels are banned, but not sex for money) can boast. Next comes physician-assisted suicide, which only Oregon and Washington permitted as of the closing date on our index (Montana now allows it as well, via court decision). Whether police can take DNA from criminal suspects comprises a little over a tenth of the subcategory. Religious freedom restoration acts (RFRAs), statewide trans fat bans, and two-party consent laws for recording public officials (which police and prosecutors interpret extremely broadly) are weighted the same. The other two policies, worth about a tenth together, are fireworks bans and prohibitions on raw-milk sales.

Marijuana policies are a high-profile issue and are worth over a tenth of the paternalism index. Full legalization of the production and sale of marijuana is the optimal policy choice, which will enhance freedom not only by allowing adults to engage in consensual behavior but also by reducing the harmful consequences of the drug war and the related and more dangerous activities that result from drugs’ being illegal (such as incentives for gang involvement). Unfortunately, every state criminalizes the production and sale of marijuana for nonmedical purposes.17 Some states have decriminalized possession of small amounts of marijuana, legalized medical marijuana, or moderated sentencing criteria for marijuana offenses. These policies should be understood as “humanitarian” measures rather than real alternatives to prohibition. Nevertheless, they do enhance freedom for some people in an important way. We consider medical-marijuana exceptions to be the least important marijuana policy in our dataset, partly because fewer people are affected by this exception than by marijuana laws in general, but mostly because laboratory-developed cannabinoids are rendering the medical argument for legalization less compelling over time. Anything that increases individuals’ fundamental freedom in their own bodies is positive, but the other marijuana policies— legalization of low-level possession (only in Alaska), decriminalization of low-level possession, making high-level possession a misdemeanor rather than a felony, making low-level cultivation a misdemeanor rather than a felony, mandatory minimum sentences for low-level cultivation or sale, and maximum possible sentences for a single marijuana offense (some states allow life in prison for a single marijuana charge)—are more important. We also include in this subcategory policies toward Salvia divinorum, which some states have banned. This variable receives a very low weight because of the plant’s unpopularity for consumption.

Gun control is worth just under a seventh of the full paternalism category. It is worth slightly more than marijuana policies because variance in state policies is so much greater and because state and federal constitutions explicitly protect the right to keep and bear arms.18 Illinois allows municipalities to ban possession of handguns altogether, although McDonald v. Chicago (2010) has recently overturned such laws, while 26 states allow anyone to wear a handgun openly on the hip without a permit of any kind.19 The appendix describes our construction of the guncontrol variable in more depth. Essentially, this variable captures a wide range of policies, from concealed- and open-carry regulations to assaultweapons bans, waiting periods, gun-show and private-sale regulations, licensing of gun owners, registration of firearms, trigger locks, and more.

Education is the most important subcategory under paternalism, worth roughly twice as much as marriage and civil unions, asset forfeiture, and arrests for victimless crimes and worth more than three times as much as alcohol regulations. It represents almost one-twelfth of the total freedom index. Besides taxing and spending, which are each worth one-eighth of the overall index, education is the most important subcategory. The reason we consider education regulations so important is that they affect the future course of liberty by affecting how and what the next generation is taught. Education regulations lie within the paternalism category because they are fundamentally justified on the claim that parents do not know how or where best to educate their own children. Politically, of course, the regulations probably exist to serve the interests of school administrators and teachers’ unions rather than for any more highminded purpose.

Even if some regulations, such as curriculum requirements, helped to achieve better educational outcomes, libertarians would generally reject them as infringements on the legitimate sphere of parental discretion. As with gun-control laws and sobriety checkpoints, the libertarian point of view holds that people should be left alone unless they demonstrably pose a risk of harm to others; this viewpoint rules out laws based on “prior restraint,” which violates the rights of all individuals in order to prevent the possible commission of future harms by some. Therefore, if some parents intentionally maintain their children in a state of gross ignorance, then they should be prosecuted for abuse. Otherwise, families should be left alone to pursue the educational choices best suited to them.

Home- and private-school regulations are each worth just over a third of the subcategory. The remainder is divided equally among availability of tax credits or deductions for attending private school, number of years of compulsory schooling, and mandatory kindergarten. Among private-school regulations, the lowest weight goes to a variable counting whether a state has a law explicitly authorizing homeschooling. The lack of a law does not necessarily mean that homeschooling is prohibited. In Idaho, there is no homeschooling law, and the practice is therefore permitted and effectively unregulated. However, we do think that having a law is a net benefit because it should provide some legal protection to parents.20 The most egregious homeschooling regulations, in our view, are teacher qualifications—which rule out homeschooling for some parents—and standardized testing requirements, which can be expensive and time-consuming. Next is an index of curriculum control, which is usually broad rather than detailed, followed by indices of notification and recordkeeping requirements, which can be annoying but are usually not onerous. Private-school regulation can actually go even further. In some states, the local school board or other government agency must approve all new private schools. That variable is weighted highly, as is licensure of private-school teachers, while curriculum control and school registration are worth half as much.

Table 4 gives the summary scores and ranking of the states on personal freedom. Note that the range of scores on personal freedom is smaller than that on economic freedom. We take this difference to reflect the fact that liberal and conservative states both like to protect some personal freedoms and threaten others, whereas on economic issues, liberal states simply tend to have bigger governments.


14. Iain Gately, Drink: A Cultural History of Alcohol (New York: Gotham Books, 2008).

15. Saurabh Bhargava and Vikram Pathania, “Driving Under the (Cellular) Influence: The Link Between Cell Phone Use and Vehicle Crashes” (working paper, AEI Center for Regulatory and Market Studies, Washington, DC, 2007), id=1129978.

16. Larry Copeland, “Texting Bans May Add Risk to Roads,” USA Today, September 28, 2010, wireless/2010-09-28-1Atextingbans28_ST_N.htm, accessed December 6, 2010

17. We do not consider laws on the possession of cocaine, heroin, or other drugs. Because every state criminalizes possession of these drugs, there is no variance to work with

18. See in particular District of Columbia v. Heller, 554 U.S. 570 (2008).

19. Two additional states allow permit-free open carry unless local ordinances prohibit it.

20. In this case, we mirror the concerns of the Bill of Rights supporters who argued against the Federalists that a written legal protection of individual rights was necessary even though they were natural rights retained by individuals.

' '