Interest in the measurement of economic freedom originates from two separate lines of inquiry. The first, firmly embedded in the tradition of normative economics, focuses on freedom measurement for enlarging the narrow evaluative bases of welfare economics. The second, rooted in the libertarian tradition, tries to construct objective empirical indices of the extent of economic freedom enjoyed by countries on a world-wide basis. Despite their different origins, methodology and aims, both traditions share a common denominator that informs their theoretical underpinnings and the measures that they construct, viz., individual choice. More precisely, economic freedom measures constructed within each line of inquiry reflect different relationships between ‘making choices’ and ‘having freedom’.
For the purposes of this review, I start with a brief illustration of the role of choice in freedom measurement. I then distinguish three perspectives on choice that are at the basis of economic freedom metrics:
1. choice as picking up alternatives;
2. choice as deliberation;
3. choice as ‘undistorted’ selection.
I analyze the freedom metrics connected with each perspective and discuss their merits.
1. The Role of Choice in the Measurement of Freedom
Whichever line of inquiry one takes side with, the measurement of economic freedom is rooted in some relationship between economic freedom and choice. In an influential paper which originated a lively debate on the appropriate criteria for measuring freedom, Pattanaik and Xu (1990) write that, irrespective of which option the agent considers as the best, availability of opportunities reflects a certain degree of freedom for the decision maker which is impaired if the extent of options contracts, even if the most preferred alternative remains available. (p. 385).
Pattanaik and Xu work within the normative economics tradition. They aim at enlarging the set of intrinsically relevant criteria that should be used in the assessment of states of affairs and — as the passage highlights — they consider freedom as ‘availability of opportunities’ a suitable candidate for the post. In their interpretation, access to opportunities gives to a decision maker the possibility of selecting among alternatives and, in so doing, of enjoying a certain degree of freedom. Their measure of economic freedom is based then on the idea that, as the extent of accessible opportunities increases, so does the agent’s degree of liberty, unveiling therefore a direct relationship between having economic freedom and the making choices in the sense of selecting among alternatives.
Despite the different interpretation of economic freedom, a relationship with making choices can also be traced for the line of inquiry which works in the thrust of the libertarian tradition. Scholars who contributed to this line of analysis regard economic freedom as being unencumbered by unnecessary government interference in the pursuit of one’s own economic activity. Necessary interference is that which guarantees protection of property rights and the smooth functioning of competitive markets. As Buchanan (1975) says, [u]nder regimes where individual rights to do things are well defined and recognized, the free market offers maximal scope for private, personal eccentricity, for individual freedom in its most elementary meaning. (p. 18).
But note that unnecessary government interference reduces economic freedom since it alters both the assignment of property rights and the possibility of trading at prices set by competitive markets, prices which reflect the agent’s marginal valuation of resources. But, if one believes what economics has taught us, changing the value of resources at the margin affects individual choice, leading to losses of economic freedom and unveiling, once again, a relationship between enjoying economic freedom and making choices.
2. Three Perspectives on Choice
Although choice can be considered as a common denominator in the analysis and measurement of economic freedom, yet the way in which their relationship has been framed within the two lines of inquiry differs substantially. By and large, three interpretations find support in the literature. ‘Choice as picking up alternatives’ and ‘choice as deliberation’ are rooted in the normative economics tradition, whereas ‘choice as undistorted selection’ is pursued within the libertarian perspective.
1. Following Miller (1991), choice as picking up alternatives coincides with ‘possibility to act’: an agent enjoys a certain degree of economic freedom if she has alternative opportunities or courses of action to choose from, and the larger the set of opportunities that she may access, the greater the extent of her economic freedom. Possibility to act has generated two kinds of measures. The first one, introduced by Pattanaik and Xu (1990), relies on pure cardinality: their ‘Simple Cardinality-based Ordering’ (SCO) compares pairs of opportunity sets on the sole basis of the number of elements that they contain. The second measure, proposed by Sen (1988, 1991, 1993), suggests that the assessment of the degree of economic freedom must depend on the preference relation of the decision maker over the available options. An opportunity set A offers greater economic freedom than B if it contains more options and at least weakly preferred to those in B.
Measures based on such a perspective between economic freedom and choice have become the catalyst of a number of criticisms. Some have objected to SCO’s endless possibility of distinguishing opportunities, which forecloses the ground to any general classification of ‘types’ of opportunity (Sugden, 1998; Bavetta and Guala, 2001). Others have targeted the role of preferences in Sen and his underlying notion of economic freedom as ‘doing what one wants’ (cf., among the others, Hayek, 1960).
But, a further criticism applies which enjoys a stronger grip. It states that measures of economic freedom which rely either on a pure quantity assessment or on an exogeneously given preference relation do not capture an interesting interpretation of choice since they lack reference to the deliberative process. The reason is that, in order to deliberate, the decision maker must be involved in a selection effort which can neither be captured by SCO (where preferences do not enter in the assessment of economic freedom), nor by Sen’s measure (where preference are exogenously given). Neither of the two measures therefore makes use of information about how the decision maker selects among the available alternatives and about whether she will be able to exercise her positive economic freedom in the decision process.
2. While a compelling challenge to the aforementioned metrics, this latter criticism is also suggestive of the way towards alternative measures. Following its thrust, some recent papers (cf. Pattanaik and Xu, 1998; Bavetta and Peragine, 2000) introduce quantitative assessments of economic freedom that capture how free is the decision maker’s deliberative process. In the pursuit of such a goal, these measures endorse a different relationship between choice and economic freedom (choosing as deliberating) in which the focus is shifted from the post to the pre-deliberation stage of a choice, i.e., that stage of a decision process at which the preferences of the decision maker are not yet shaped. At that stage, the chooser is in the position for developing a will of her own (her own individuality) and the act of choosing, as John Stuart Mill’s On Liberty suggests, becomes the instrument for shaping it. Having options to choose from fosters individuality because certain fundamental qualities of an agent which render him autonomous, such as "perception, judgement, discriminative feeling, mental activity, and even moral preference" (Mill, 1859, p. 122) can be exercised and developed by making choices, i.e., in the deliberation process. The relationship between having economic freedom and making choices takes here a different (and more interesting) turn as choosing becomes, at the same time, expression of and training for the development of a person’s autonomy. Two distinct ideas emerge: (1) choosing is an activity that is valuable in itself (autonomy as ‘doing the work’ of choosing, so to speak, Nozick, 1974); (2) choosing is valuable because it is functional to the development of an autonomous identity (exercising as an instrumental good). In both cases, choosing as deliberating fosters economic freedom because it develops individual autonomy.
When choice is tied to deliberation the measurement of economic freedom has to take into account the development process of individuality. Technically, this is captured by the idea of potential preference, i.e., all the preference relations that an agent may uphold over a set of options. Potential preferences, in turn, identify the relevant options, namely those options that can be chosen on the basis of a potential preference that has turned actual. An individual enjoys a wider degree of economic freedom the larger her set of relevant options. Of course, for the process of formation of individuality to be meaningful, not all potential preferences should be allowed to become actual because, in this case, we would be back to SCO. Nor we can admit shrinking the set of potential preferences to a singleton since, in this other case, we would be back to Sen (Bavetta and Peragine, 2000). A screening device is therefore necessary to select among the potential preferences. The most appropriate requires that options be not too distant in terms of preference from each other (Bavetta and Peragine, 2000; Bavetta and Guala, 2001).
3. The last perspective on the relationship between economic freedom and choice belongs to the libertarian tradition. It may be best illustrated by means of an example. Consider an unnecessary interference that alters the free market determination of relative prices. Just to fix ideas, let it consist of a $1 specific excise tax levied on gasoline. The tax modifies the agent’s marginal valuation of resources as well as her optimal choice. Assuming a linear demand curve over the relevant range, suppose that, at the equilibrium pre-tax price of $10 an individual consumed 30 liters of gasoline per week whereas, at the equilibrium post-tax price of $11 she consumes 25 liters. The burden of the tax is $25 (which go in the government’s coffers) plus $2.5 of deadweight loss. This would be though just the ‘welfare’ cost of the tax which represents a measure of the distortion imposed on the free market choice but that does not capture the entire effect on economic freedom. To compute the latter we should add the consequences of the tax in terms of the mutually beneficial transactions foregone. In general, therefore, the measure of the loss in economic freedom associated with the unnecessary government intervention is given by the marginal value of the distortion multiplied by the number of transactions both undertaken and foregone.
In principle, all empirical measures of economic freedom so far constructed in the literature share these premises (cf., e.g., Gwartney et al., 2000; O’Driscoll et al., 2001). A number of complications though do arise. For example, although theoretical analysis would prescribe summing the measured distortions in each market, this is not always feasible.
In practice, to overcome these difficulties, empirical measures are contructed starting with a definition of a set of relevant economic categories where interference with individual choices may occur. These include the policy design activity of governments (trade policy, monetary and fiscal policy), as well its regulatory role (regulations on banking and finance, on the labour market, on international real and financial transactions) and its protective role (protection of property rights and control over black market activity). Once the set of variables is identified, each component is assessed. The techniques for the assessment vary with availability of data and the characteristics of the component. Finally, each component is weighed (but in some rankings all variables are equally weighed) and a grading scale is constructed which aggregates in different fashions the score assigned to each component in each country. Aggregation preserves temporal and cross-country consistency, i.e., respectively, if a country improves with time, this is reflected in its rating and the distribution of the country ratings reflects the distribution of the actual values among the countries.
The empirical measures of freedom suffer from two major limitations. The first is that appropriate micro-foundations for each component are unattainable given the complexity of the information that this would require. As a consequence, for some components, the index provides us with too coarse an approximation to be reliable. A second limitation is that, though empirical measures of economic freedom are able to distinguish the free from the unfree countries, yet the indices are hardly sensitive to fine distinctions.
