Information, organization, and freedom: Explaining the great reversal

Abstract

We suggest, in a Coase-Demsetz perspective, that the social demand for individual rights—or freedoms (whether civil, political or economic)—is derived from, because complementary to, the changing size of hierarchical organizations. The general downsizing and decentralization process observed worldwide after 1975 is itself the result of the information revolution and the resulting abundance of information. It follows that social demand for freedoms depends in turn—and inversely—on the cost of information (and thus on market imperfection) as well as on traditional determinants such as the distribution of resources and human capital. This implies that freedoms are adopted, implemented, or “produced” by various political regimes according to an objectively observable and contingent determinant. We believe this approach can shed light on the reason for the waxing and waning of freedoms in modern history.

“What is the best constitution? Tell me first for what people and what time” replied the sage Solon (quote from the 1946 Bayeux speech of General de Gaulle).¹

1 Introduction

What makes the rule of law, economic freedom, and personal liberty change across nations and over time?

A classical liberal is confronted by a major puzzle: If freedom is always and everywhere welfare improving, both as a possible argument in the utility functions (after wealth) and through its positive influence on prosperity and economic well-being, why has its adoption come so late and only in a minority of countries? Late indeed in world history since, as noted by Tullock (1987, p.175), “In current usage ‘democracy’ is normally confined to systems where all adults are permitted to vote. Historically, this situation was almost unknown before the 20th century” and by Wintrobe (1998, p. 6), “most of the world for most of human history has lived under dictatorship.”

Among the small number of nations that adopted a democratic constitution in the modern era (starting with the Enlightenment and the Industrial Revolution of the late 18th century), a general regression of freedom began well before 1914, accelerated during the 1920s and 1930s and recovered only slightly during the forties, fifties and sixties. A low level of freedom thus lasted until the last quarter of the 20th century.² What are the determinants of these general waves of freedom, then of ‘unfreedom’, and then of freedom again?

Figure 1, which is based on data from FreedomHouse.org, provides insights into the contemporary expansion of freedom around the world.

Fig. 1

The Contemporary Freedom Wave.

In the past, determinants of democracy and determinants of economic freedoms have been analyzed separately, but this distinction runs contrary to the trend toward a unification of economic analysis and other fields in the social sciences: In the Public Choice tradition (The Calculus of Consent, Buchanan and Tullock 1962) as well as in the Chicago school of political economy (Stigler, Becker’s “Competition and Democracy” (1958), and then in his 1983 paper), there is only one model of man.³ The “homo economicus” maximizes his utility whether in economic and/or in political pursuits, and this view has gained acceptance, beyond the traditional field of economics, especially in political science. Hence it is rather strange that the two strands of literature, the one that explains democracy and political freedoms, and the one that explains economic freedoms, are quite separate and often refer to different determinants.

Finally, none of these theories has been built on the basis of the economic theory of rights, pioneered by Demsetz (1967), despite the fact that these various forms of freedom are basically individual rights (see below Section 3).

Our solution to these problems is to focus our analysis on general trends of various types of freedom in the modern (post 1750) era, when freedoms first became a significant phenomenon in western societies, whereas other scholars have been concentrating on specific factors explaining inter-country differences. Our theory relies on a general definition of freedoms as rights, whether economic or political (since they are rather highly correlated⁴). Thus the problem is best understood as one pertaining to the economics of rights.

We start with a basic definition of freedoms as rights, and proceed with the use of the economic theory of rights first developed as a theory of property rights. We go on investigating the conditions leading to the decentralization of production. Coase (1937) provided the basic framework by modelling the choice between production through market exchanges and that by (hierarchical) firms, or in other terms, the choice between production through the decentralized price system and production by the internal command system. Accordingly, we argue that political and economic freedoms are derived from, because complementary to, the downsizing of many hierarchical organizations, “economic” (corporations) as well as “political” (public administrations and States).

Our approach has the major advantage of not basing the evolution of freedoms on the weak foundation of unstable individual preferences that are supposed to evolve as a result of persuasion, imitation, ideology, and the battle of normative ideas. As in economics generally, we explain individual and collective choices and their evolution by the conditions of the environment, notably prices and costs, rather than by changes in preferences.

We present a model explaining not only the simultaneous evolution of various kinds of freedoms but also the new diffusion of freedoms at the end of the twentieth century, contrasting this diffusion with the relative absence of freedoms during the period spanning 1875 to 1975. The fact that, during recent times, this movement has been both broadly based and simultaneous throughout the world confirms our belief that there may be an underlying common factor at work, since the probability of having several independent factors all moving in the same direction during the same time period across most of the countries of the world is remote.

The remainder of the paper is organized as follows. Section 2 introduces and evaluates the various explanations for democracy, civil liberties and economic freedom. Section 3 discusses our analysis of freedom as rights, based on the role of information and organization, and its relation to country-specific effects of the analyses surveyed in section 2. Section 4 summarizes our findings.

2 The search for determinants of freedoms: history, geography, climate. A bird’s eye view of the literature

As noted above, the existing explanations fall in two categories, one for explaining the extent of democracy, and the other analyzing implicitly the extent of economic freedoms as restricted by the scope of the State and of its intervention in the economy.

Regarding political freedoms, authors focus on permanent liberty differentials between countries and consequently rely mostly on structural factors, which may be constant over time or follow an upward (or downward) trend. It follows that they cannot easily explain the decline in democracy and freedom up to 1975 and the global freedom boom which began at about that time.⁵ Neither can they explain why both civil liberties and economic liberties have followed a parallel, self-reverting, sinusoidal evolution. Lastly, while liberties generally, political as well as economic, are undoubtedly rights, existing explanations do not rely on or refer to the general theory of rights pioneered by Demsetz. The underlying problem is thus one of arbitrarily fragmented theorization (missing factors in the explanation respectively of political and economic freedoms), missing common factors capable of explaining the trend reversal, and accordingly, as we suggest, a missing (unified) theoretical framework.

2.1 Political freedoms

Explanations for the differential diffusion of political freedoms among countries fall into two main categories: those based on national income levels, and distributional, resource-based explanations based on the allocation of wealth and on the coercive power of government.

2.1.1 The income-freedom hypothesis

The simultaneous experiences of the Enlightenment and of the Industrial Revolution led to the supposition of a link between increases in income and freedoms. The most famous theory is Lipset’s⁶ income-democracy relationship—the “modernization” hypothesis—according to which the level of income determines, in the main, the level of democracy and freedom,⁷ with the last two considered equivalent in his view.

In a recent empirical test of the hypothesis, Barro (1999) adds population size, human capital, education and income inequality, a dummy variable for cultural, linguistic and religious traits, and the country’s role as an oil-producing State or as a former colony to the list of explanatory variables. These additions, purported to reflect the Lipset concept of “modernization,” account for a significant portion of the changes in both political rights (PR) and civil liberties (CL) indices.

But the Lipset hypothesis has been seriously challenged on several grounds.

The first difficulty is conceptual. The hypothesis does not explain why democracy would be a superior good rather than an inferior one, and, more fundamentally, why it would enter as a positive argument into the utility function at all. Furthermore, Lipset does not explain the mechanism by which higher income leads to more freedom. Instead he states (1959, p. 75) that “…only in a wealthy society in which relatively few citizens lived in real poverty could a situation exist in which the mass of the population could intelligently participate in politics and could develop the self-restraint necessary to avoid succumbing to the appeals of irresponsible demagogues.” Such an assertion obviously relies on a complex relationship between income and individual and social rational behavior (“modernization”), which is at best unproven.⁸

The second challenge for the Lipset hypothesis is empirical. From 1945 to 1975, countries in the Third World and Central Europe, which later democratized, showed no sign of a revolutionary turnaround in economic growth generally sufficient to reverse the preceding downward trend of democracy. On the contrary, growth typically slowed down steadily in communist countries, as well as in many newly independent former European colonies in the Third World. Thus, while the hypothesis could still explain cross-country differences in income and freedom, it cannot account for the overall boom.

The third challenge comes from econometric testing. Recent work by Acemoglu et al. (hereafter AJRY, 2008, p. 836) shows that “although income and democracy are positively correlated, there is no evidence of a causal effect,” and that the correlation between income and democracy disappears once you introduce country-fixed effects—country specific information—into the model. From this we must conclude that there are indeed country-specific explanatory variables, or country levels of the same variables, other than income, which better account for the common rise in freedom across countries.

2.1.2 Distributional hypotheses

Internal political equilibrium and the competition for influence

One of the oldest political explanations of how rights become established is based on the dispersion of ownership of scarce resources and the ability to use these resources to wrest influence from the government or ruler.⁹ This resource-based approach to political power serves to explain not only the distribution of rights, but also the structure of taxation and the pattern of political redistribution.

A country will adopt an egalitarian constitution if the owners of scarce resources are more or less equally endowed, whether they are small farmers with an equal distribution of land plots (Jefferson) or individual gold diggers in the western gold rush (Umbeck 1981). In the latter case, it is mainly due to the general availability of the “the great equalizer” in the conquest of the West, the Colt revolver.

Such a model can be extended to groups. Becker (1983) finds a competitive equilibrium will result regarding the distribution of taxes and transfers among special interest groups, based on the supply and demand for influence. The same approach can be applied to the distribution of rights. For example, the extent of suffrage in a particular country could depend on the diffusion of resources—countries that have an uneven distribution of resources may also have an uneven distribution of political rights.

Vanhanen (2003) shows, with a large sample of countries, that a comprehensive measure of democracy is dependent on the degree of concentration of resources (broadly conceived and measured) within each country.

While the explanation of rights being tied to resource distribution may account for long run and country specific patterns, it does not fit the fact of the last thirty-year wave that we want to explain. We know that the distribution of resources does not vary much over time and could not have varied, either significantly or in the same required way, in all countries over the period concerned. We do know that the dispersion of wages and incomes, while tending to narrow over the twentieth century, widened again during the last few decades. So the recent increase in wage or income dispersion should, according to the theory, lead to a reduction in democracy and a restriction of freedoms not, as we have witnessed, a “freedom boom” (Rosa and de Vanssay 2010).

Geography and the long run, path-dependent version of the distributional hypothesis

Another argument relying on the historical distribution of resources sheds further doubt on the income-freedom hypothesis. Countries move towards democracy and greater freedom at their own pace. When showing that the link between income and democracy disappears once country-fixed effects are introduced into regressions, AJRY conclude that the income variable actually expresses the influence of historical, cultural and institutional country-specific, factors on democracy—that is, path dependence.¹⁰

Country-fixed effects, they argue, are the result of earlier political equilibria, which determined any later—and possibly much later—change or path of change in economic, civil, and political freedom. There are not likely to be institutions supporting economic growth in countries where power has been concentrated in the hands of a small elite who used it to maintain its rents and position, and will keep doing so even at a high aggregate economic opportunity cost, no matter how worthwhile more economic growth might be.¹¹

This would be the case of countries colonized by Europeans where the local disease environment (malaria and yellow fever) determined a high mortality for the colonizers who created “extractive” institutions instead of colonies of settlement. Their number remained low and they tried to maximize short run profits (e.g. in the Congo and other African possessions).

In contrast, “where Europeans faced low mortality risk because the climate was drier and cooler, they set up ‘settler’ institutions designed to create a political and economic environment that resembled Western Europe.” (Haber and Menaldo 2010). These initial colonial institutions and the size of European colonizers’ population relative to local populations, and thus the degree of wealth inequality, conditioned the distribution of rights and the later path of political and economic development. Where small European elites dominated vast populations of African slaves or Native Americans for instance, they easily crafted political institutions that gave them power beyond their numbers and captured the State, even after independence.

Engerman and Sokoloff (1997) focus on the effect of climate and geography on the profitability of various cultures rather than on mortality: where conditions were conducive to large scale plantation agriculture (sugar, cotton, and other tropical crops) a small minority of European landowners could mobilize large amounts of capital and employ large numbers of local people and/or African slaves. They rigged the political institutions for their own advantage and that oriented the later political and economic path of development toward authoritarian and concentrated political and economic power, thus severely limiting freedoms. They also argue that the natural environment of Canada and the United States had none of the “factorial advantages” of Latin America: it was too cool and dry to grow sugar, there was no significant mineral wealth, and the indigenous population was sparse. “The resulting societies were therefore composed of immigrant small farmers who were evenly matched in terms of educational attainment and political sophistication.” (Haber and Menaldo, op. cit.)

There are problems, however, with this path dependence argument. First, there are several examples of autocratic governments implementing very successful economic development policies, while not allowing freedoms that would likely undermine the power of its established elite (Giavazzi and Tabellini 2005). Second, its validity is geographically limited: how do we account for countries which were not colonized or where democracy was a colonial legacy (Przeworski et al. 2000, p.83), but nevertheless are poor and/or autocratic today? Moreover Engerman and Sokoloff (1997) focus solely on the Americas, while Acemoglu et al. (2001) limit their analysis to Western European colonies. Places that were so inhospitable that Europeans powers were not interested (the Arabian Peninsula) or were part of non-European empires (Turkistan, a part of Russia) or could resist colonization (Japan), lie outside of this framework. That is also the case, of course, for European countries themselves, whether colonizers or not, which are not included in these studies. In order to overcome these limitations Haber and Menaldo (2010) suggest a more general explanation. In their view, the climate, and specifically an adequate level of rainfall (550 to 1300 mm of precipitation per year); uniquely found in temperate zones, determines the development of settled and diversified agriculture conducive to rather small properties instead of the large plantation complexes in the tropics (above 1300 mm of precipitation per year). The surplus they are able to produce gives incentives for the growth of markets and of States. States replace tribalism and create impersonal laws, thus rights. Moreover, the growth and resulting competition of larger States determines financing needs that lead to the granting of rights of managerial control to taxpayers in order to obtain a higher level of consent and, consequently, more tax revenues. In this framework, moderate rainfall and the correlated development of States, are the main determinants of democratic rights and freedoms.

The same limitation, however, applies to this analysis as to that of Acemoglu et al., and of Engerman and Sokoloff: It explains the differences of democracy across nations, but as a permanent factor, and, as such, it cannot account for major temporal variations in the spread of democracy during the modern era. A second limitation comes from the focus on political freedoms (democracy) and a neglect of economic freedoms. This second aspect has been less developed than the study of the determinants of democracy.

2.2 Economic freedoms

We have so far discussed positive theories of aggregate political and civil liberties, or democracy. Surprisingly, similar positive theories are conspicuously missing in the case of aggregate economic liberties, or exceedingly poor when formulated. But they are implicit, in a negative way, in theories of the growth of the State. Classical liberals see the State, through its regulations, taxes, legislations and downright nationalization, as the main agent of destruction of economic freedoms.

2.2.1 The State as a source of reductions of economic freedoms

We suggest that this emphasis is a result of the 20th century development of the State as a substantial producer and supplier of transfers. The growth of the State began in the 1930s and 1940s, and then accelerated spectacularly during the 1950s. Some see the State as an enemy of freedoms because it competes with private investors for ownership of private goods production (including various insurance products typically supplied by the “welfare state”), and because of the development of extensive regulations of private enterprises which amounts to another form of taxation (Posner 1971).¹²

Many authors are bent on showing that these economic freedoms, which, according to most, are conducive to more growth and wealth, are repressed because of a faulty understanding of their positive effects by short sighted or uninformed governments, who do not understand, or underestimate, long term effects of freedom-reducing regulations such as State ownership and taxes. This implies either that most governments have been consistently wrong for more than 70 years of the twentieth century (the ‘inefficiency of politics’ thesis) and have suddenly benefited from a clear understanding of economics basics after 1975 (the ‘miracle of ideas’ thesis), or that the efficiency gains from economic freedom have been consistently dominated during this period by other factors leading to the persistent growth of the State; a more likely hypothesis.

Indeed, even if economic liberties always favor growth, Public Choice considerations can explain that they could be rationally repealed, given the process of collective decision making, either because the interests of the ruler do not coincide with those of other productive agents, or because the interests of various groups of agents do not coincide with each other, nor with the maximization of growth and aggregate income level.

Thus, political factors could explain the amount and level of economic freedoms, positively or negatively (de Vanssay et al. 2005). Some authors have suggested that democracy in itself is inimical to economic freedom in that it leads to increased State intervention, regulation, and redistribution. But this is difficult to reconcile with the positive correlation observed between political and economic liberties (Rosa and de Vanssay 2010).

Since the development of the State above and beyond the size that allows the production of the basic goods of security, law and order (undoubtedly the case in modern countries) implies, as many (classical) liberals would have it, a reduction in the scope for individual economic activities and freedom,¹³ a “negative” theory of economic freedom could be derived from theories of the growth of the State.

But these theories are still unsatisfactory (Holsey and Borcherding 1997,¹⁴ and Tullock 2002) and the recent evolution has been characterized by a relatively stable share of public expenditure in GDP while the economic freedom index (EFI) was booming, making it likely that another variable is required to explain the trend in economic freedoms. This might be due to the composition of the EFI, which includes not only the ratio g/y (the share of public expenditure in GDP), but also to other variables such as regulations, legal structure of business, international trade taxes and tariffs and other import barriers, as well as policies regarding monetary stability, inflation, and government deficits. If the strong rise of EFI is to be explained, while the ratio g/y was non-decreasing (Tanzi and Schuknecht 2000), it would imply that the regulatory and policy elements of the index were all the more strongly liberalized.

As far as we consider the EFI as a reasonable, encompassing, measure of economic freedom, we are justified in not retaining theories that try to explain the scope of the State as an implicit, and negative, measure of economic freedom, because the scope of the State is only a part of the EFI index.

Basically we are left without an articulated theory of aggregate economic freedom as we are in need of a theory of political and civil liberties.

2.2.2 Mobility, openness, and the decline of State control

The progressive opening of national economies since World War II and the recent decline in State power usually linked to globalization has increasingly allowed citizens to escape their respective governments’ coercion and their taxing powers. This has led some to formulate a ‘decline of State power’ hypothesis—the idea that more open economies weaken a State’s control of its citizenry and territory and, by extension, limit the State’s ability to finance itself, leading the ruler to give more guarantees (rights) to the citizens, tying his/her own hands in the process.¹⁵ ‘Mobility makes freedoms.’¹⁶

However, the alleged loss of the power of States is not obvious: As shown by Tanzi and Schuknecht (2000), even though individual mobility did increase, as did macroeconomic openness, the government’s share of GDP has not decreased significantly, and certainly not before 1975. Nevertheless, international openness as a structural fact—smaller countries must be more open than larger ones—can validly appear as an explanatory variable of cross country differential freedom, possibly also with regard to its temporal evolution.

Mobility of goods and people also contributed directly to the dissemination and general increase of information until the era of the Internet, through the easier and accelerated movements of ‘cultural goods’ (books, newspapers, audio and video recordings), and of human capital that is a required complement to the use of information. Material mobility was a multiplier of information. But in the Internet era, mobility of information tends to become a substitute for the mobility of goods and people.

These two contradictory effects, the positive and negative correlation between mobility and information, make it difficult to assess separately the autonomous influence of each factor on individual rights.

Moreover, while the progress of goods’ and people’s mobility since the first Industrial Revolution has been relatively steady until the recent acceleration of the generalized air travel era, it has not been as spectacular and unrelenting as the dramatic and exceptional fall of information costs exemplified by Moore’s law. The fall of the cost of information had also been outstanding before and during the first Industrial Revolution, at a time when the progress of mobility was modest.

Lastly, information exerts a significant influence on the aggregate demand for rights through its impact on the organizational structure of production, and thus on the numbers of efficient individual deciders and the increased productive role of individuals, whereas mobility does not.

We thus argue that the impact of information on freedoms, that we consider in the following section to be the main and common driver of the overall modern waves of individual rights, has been deep and pervasive, and self-reverting through time (as explained below). In contrast, the impact of mobility on freedoms has been weaker, trend-like only, and at times undistinguishable from that of information for which it was a complementary input, and thus has been made in part redundant, in the past few decades, by the direct mobility of information resulting from the ICT revolution.

2.3 The missing factors

What is left of the plausible determinants of freedoms at the end of our survey? The main casualty is the income effect as an explanatory factor, which is at the same time unconvincing theoretically, contradictory in its assumed consequences with regard to political and economic freedoms evolution, and econometrically questionable. We provide an explanation of this result in the next section.

Contrary to Barro (1999, p. S168),¹⁷ we do not believe that trend-like variables, which were purported to explain democracy and freedom, besides income (mostly education), in periods preceding the global boom when freedoms were declining or at least non-increasing, are at the same time able to explain the great reversal of freedom boom that started around 1975. Indeed, there is no adequate balance in the equations between variables with a positive influence and variables with a negative influence on freedoms, and there is no radical change in the values of either that could generate a reversal in the dependent variable within the framework of these same equations.

Given all the difficulties of existing models, we believe the missing factors question reflects a missing theory problem. What is required is a direct theory of the demand for freedoms that should build upon the existing theory of rights. Such a theory of demand for rights—and thus for freedom—that is grounded in microeconomic theory, already exists. The theory of property rights, as outlined by Demsetz (1967), reminds us that rights will be adopted and developed when the costs of defining and enforcing them are lower than the expected profits from their implementation.

In the next section we develop, using Demsetz’s framework and a Coasian theory of the choice of organization, a model of organizational change and its consequent requirements of individual freedoms.

3 Information, organization, and the value of rights

Why is organization important for our problem? Because the overall organization of production, in all societies, reflects a mix, in variable proportions, of two basic (and polar) inter-individual principles of interaction and cooperation: exchange through markets and the price system, or submission through hierarchical rule (for instance in the labor contract, but obviously also in the harsher institutions of serfdom or slavery). Where hierarchical organizations grow, markets contract, and many more individuals submit to the decisions of a few. They do not decide by themselves but obey orders from others. The numbers of independent individuals thus evolves inversely with the development of hierarchies, private as well as public. Large-scale corporate or state bureaucracies accordingly reduce the need for individual freedoms. In contrast, smaller hierarchical organizations contribute to an increase in the number of decision-makers, and these need freedoms in order to decide and act. The rights in production also impact political rights as noted by Milton Friedman: Economic freedoms are a condition for political freedoms. Moreover, also in the political field, smaller, more competitive States (political markets that are “contestable” in the sense of Baumol) grant more political freedoms to the taxpayers they control, in order to dissuade them from emigrating.

Why then is information relevant to an explanation of organization-linked demands for rights? It is because the availability of information determines the more or less centralized mode of the organization of production. Hayek, in The Use of Knowledge in Society (1945), famously emphasizes the fundamentally decentralized distribution of information when he indicates that:

…the “data” from which the economic calculus starts are never for the whole society “given” to a single mind which could work out the implications, and can never be so given.

The peculiar character of the problem of a rational economic order is determined precisely by the fact that the knowledge of the circumstances of which we must make use never exists in concentrated or integrated form, but solely as the dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess. (p. 519)

While it appears that he implicitly relies on a given and fixed distribution of information among all individuals, and assumes it to be broadly distributed, it is not exactly so.

First, he distinguishes (p. 521) between different kinds of knowledge: “unorganized knowledge” (“the knowledge of the particular circumstances of time and place”) and “scientific knowledge”:

…those more likely to be at the disposal of particular individuals and those which we should with greater confidence expect to find in the possession of an authority made up of suitably chosen experts. (emphasis ours, JJR-XdV).

Second, he notes that:

…the “man on the spot” cannot decide solely on the basis of his limited but intimate knowledge of the facts of his immediate surroundings. There still remains the problem of communicating to him such further information as he needs to fit his decisions into the whole pattern of changes of the larger economic system. (ibid., p. 524–525) (emphasis ours, JJR-XdV).

He thus makes two points: first, information in general is unevenly distributed among individuals, and second, there is a need for the acquisition of information by agents with these diverse levels of information.

It is clear that, in history, knowledge (‘unorganized’ and ‘scientific’) has been unevenly distributed. Business leaders, for example, have usually had more information than employees or workers.

But further, the cost of acquisition of information has varied over time (and across countries) and has, consequently, tended to change among individuals. When computers were scarce and expensive, only a few big players (firms, governments) were able to buy or rent them, and thus access the capacity of storage, processing and transmission of information that individuals or smaller firms could not obtain. Today, with the low cost of the PC, nearly everyone can own one (or more), and any individual gains access to all (or almost all) “scientific” or local information available on the Internet.

Hayek’s considerations about the importance of information distribution and acquisition opened the way to a calculus of optimal acquisition of information by all agents in the economy,¹⁸ and of the social creation of individual rights required to exercise the possibility of better choice-making allowed by this information.

So, while Hayek was more concerned with the impossibility (or inferiority) of central planning, due to the wide dispersion of ‘unorganised’ information, he did recognize the complex issues of information abundance, distribution, and demand by authorities and ‘men on the spot.’

From that starting point, and in the light of the recent ‘information revolution’, our emphasis is on the change in the quantity of information in the surrounding economy. We thus extend the Hayekian (1945) line of analysis by considering the dynamics of the cost of information acquisition, its impact on the organization of production, and consequently on the distribution of individual rights, which is necessary for obtaining efficiency in production.

We start from these ‘informational considerations’ to analyze the demand for information based on its cost. We continue with the impact of this cost on the distribution of information among individuals, which also determines the more or less centralized modes of production. In this analytical framework, a change in the cost of information in time and space leads to a change in the mode of production, towards more or less centralization. We find that these factors have substantially varied over time and have influenced the optimal structure of production (‘economic’ and ‘political’), and thus determined the diffusion of freedoms among the population as explained by Rosa (2000, 2006).

A bird’s eye view of the argument is as follows:

Individual freedoms are rights, and individual rights are socially valued because necessary for individual decision-making. The more efficient decentralized decisions are, relative to centralized ones, the more individual rights are required. When information is scarce relative to production, hierarchies develop since they are an information-saving device, while on the contrary when information is abundant relative to production, markets develop, because the transaction costs are mainly information costs, and these are reduced where there is more information.

The relative amount of information—or information/output ratio—has been variable in modern times: high in the Enlightenment/Industrial Revolution context, but then lower during the second industrial revolution (1875–1974), because of the massive increase in productive capacities and the speed of production flows described by Chandler (The Visible Hand), increasing the denominator more than the numerator. And it went up again with the recent Information and Communication Revolution that has occurred since the mid-1970s.

It is the change in the availability (or cost) of information relative to production (and cost) of other goods and services (measured by the gdp), and thus the information/gdp ratio, that explains the succession of waves: the first freedom wave of 1750–1850 (or 1875 according to some authors), followed by the reverse wave of centralization, imperialism, and totalitarianism, of 1875–1975 (or 1960 if we consider the end of the old western empires and the beginning of the liberalization of world trade, but 1975 if we consider the nadir of democracies in the world (Barro 1999)), and then the present wave of decentralization and democracy (1975–), in a sinusoidal sequence of organization and freedoms.

The successive steps of the argument follow.

3.1 The nature of freedom(s)

An economic theory of freedom must be based on a definition of freedom as an economic object, in order to determine under what conditions this object will be more or less abundant in a particular country and specific period. An economic explanation must indicate how the quantity of freedom would vary according to the supply and demand of individual rational economic actors.

In our analysis, individual freedoms are rights to act or not to act. Some examples of these freedoms are rights to acquire, use or destroy physical or intellectual assets, rights to use or not use various words or means of expression, rights to take self-employed work or work for an employer, rights to leave or remain in a certain country, rights to vote or not vote for political candidates and government.

These rights are ‘officially’ recognized (de jure, by legitimate authorities or the Ruler), or alternatively, where there is no legitimate Ruler, they are de facto rights, such as land claims during the Gold Rush (when, as noted by Umbeck, ‘Might Makes Rights’).

They are more or less defined and defended from country to country. Indeed, as Barzel (1989, p. 2) notes, it is clear that:

The rights people have over assets (including themselves and other people) are not constant; they are a function of their own direct efforts at protection, of other people’s capture attempts, and of government protection. Squatters’ rights to the land they occupy are less secure than those of legal owners not because they lack deeds but because less police protection is expected for such holdings. As defined here, property rights are not absolute and can be changed by individuals’ actions; such a definition, then, is useful in the analysis of resource allocation. Economists’ past failure to exploit the property rights notion in the analysis of behavior probably stems from their tendency to consider rights as absolute.

And furthermore: “The distinction sometimes made between property rights and human rights is spurious. Human rights are simply part of people’s property rights.” (ibid., 1989, p. 2).

If economic and political freedoms are property rights, then there is only one category of phenomena to be explained, property rights. This makes it easier to develop a coherent theory that fits with Demsetz’s economic analysis of the evolution of property rights.

3.2 Rights are socially valued when productive

According to Demsetz, property rights will be produced, defined and enforced according to their intrinsic characteristics and the social environment in which they are created.¹⁹ Rights will be readily defined and defended when their value is high and/or their definition and enforcement cost is low. Individuals will be granted freedoms when their actions and decisions are valuable in production processes, and denied rights and freedoms when their capacities for decisions are not required or even counter-productive (for instance in the military). We interpret Demsetz’s finding as follows: Rights define the domain of ‘appropriate’ individual choices, i.e. the domain in which individual choices contribute positively to social wealth production.

We thus find that the relative effectiveness of decentralization to centralization in the organization of production provides the answer. When the decentralization of production is more effective, individual decision-making is necessary, and the cost of coercion in terms of wasted resources for society increases appreciably. On the contrary, during times when centralized production is economically more effective, governments, whether democratic or not, and corporations, do not have a lot to lose in terms of economic growth when using authoritative constraints on individual employees.

Thus, human rights and freedoms are produced and, as such, are variable. Personal freedoms will vary across countries according to the advantages they bring—their value—and their costs of definition and implementation.

3.3 Economic and social value of freedom

The exchange value of an asset is a function of the gross income which it can generate and of the costs of measuring and policing its exchange. (Barzel 1989, p. 8)

These revenues and costs also define the exchange value of individual rights as assets, property rights, potential rights, or options on human activity. The net social value is not maximized by the maximization of rights, but by their optimization, taking into account their production cost.

Economists concerned with property rights often consider any restriction on those rights, called “attenuation of rights”, to be undesirable. A person’s ability to realize the potential value of her or his property depends on the extent of her or his property rights, which consist of the ability to use (and to exclude), to alienate, and to derive income from the property. The ability, or power, to exclude prevents the property from becoming common property, and the ability to alienate and to derive income permits the realization of gains from exchange. Since restrictions in general reduce freedom of action, restrictions on a person’s (property) (our brackets, JJR-XdV) rights reduce the value (of the property) (idem) to its owner, making such restrictions appear to be harmful. (Barzel 1989, p. 85).

But when individuals do not have to decide, individual freedoms are not productive and, on the contrary, hamper the centralized production process. Freedoms may have residual consumption value in themselves, but have then lost their major determinant.

3.4 Individual decisions are especially productive when well informed

Note that the theory of pure and perfect competition (or ‘perfect decentralization’ as Demsetz put it) assumes that information is complete and costless. Decentralized decision then is more efficient, and individual rights are required. When information is scarce relative to production capacity, the requirements of pure and perfect competition are not met. Complete decentralization is not optimal and some centralization results. A complete set of exclusively individual rights is not optimal either.

Indeed, hierarchical organization is linked to imperfection in markets. As shown by Coase (1937), in an economy with positive transaction costs, productive hierarchies for private businesses and bureaucracies emerge because they reduce these transaction costs in the production process. The essential difference between pure market production and production within a hierarchy is the degree of decentralization in the decision making process.²⁰ In a strict hierarchy (a polar case, of course), all decisions are made by the leader, whom all other participants obey. In contrast, in decentralized production or pure markets all participants make their own decisions. In the first case, people who need freedom to decide are relatively few, while in the second, general freedom is essential to the production process.

3.5 The information/output ratio is variable across nations and through time

In the modern era, the ratio of available information to output significantly increased during the Enlightenment/Industrial-Revolution era, and remained high until the second industrial revolution (1875). Thereafter it was lowered by the extraordinary increase in productive capacity and accelerated production flows (Chandler 1977) that largely outpaced the development of information technologies. For instance, for England, the curve of the real income per person followed an exponential trajectory only after 1875 (Clark 2007, Figure 10.2,). Then again, since the beginning of the IT revolution in the 1970s, the availability of information increased rapidly, leading to a general decentralization of production, both economic and political.²¹

These alternate phases of the information/output ratio explain the succession of waves.

The first freedom wave occurred because of both high information level (the rise of science and techniques, as reflected for instance in the French “Encyclopédie” and also British and German ones later) and production level (slow at first, but gathering speed over time, helped by the steam engine and other innovations), leading to decentralization. Headrick (2000) documents the first information revolution (which he defines as covering the period 1700–1850) and the “new information systems” (p. 9). He shows how, during that era, information became better organised (the scientific classification), transformed (the birth of modern statistics), displayed (mapped and graphed), stored (dictionaries and encyclopaedias) and communicated (postal systems). At the same time, income per capita grew slowly as noted by Mokyr (2002, p. 30): “It has become a consensus view that economic growth as normally defined (a rise in national income per capita) was very slow during the Industrial Revolution, and that living standards barely budged upward until the mid-1840s… Yet it is also recognized that there are considerable time lags between the adoption of major technological breakthroughs (or so-called general-purpose technologies) and their macroeconomic effects.”

Then, during the second industrial revolution, 1875–1975, we have a reverse wave of centralization (managerial and political), imperialism, and totalitarianism, resulting from productive capacities outpacing the progress of information technologies. Indeed, during that period, there was no similar acceleration in information availability. It is nevertheless impossible to quantify these diverse phenomena with a single series. We can use only partial illustrations, not a general measure of the phenomenon.²² The diffusion of information by telegraph was modest, person-to-person only, costly (transatlantic cables for instance). The diffusion of information by the telephone was very limited at first after the innovation in 1901. As for radio, it became more widely used only in the 1930s (e.g. by Roosevelt). Centralization of production was manifest in the rising importance of trusts and giant enterprises, as well as of growing States and empires worldwide.²³

And finally, we arrive at the present era (1975–) in which dramatic improvements in information and communication technologies, captured by our Fig. 2 (below), outpace the progress of production.²⁴

Fig. 2

The “Information Revolution”. Source: Internet Systems Consortium, Inc. (ISC), http://www.isc.org/solutions/survey/history and ITU World Telecommunication/ICT Indicator Database.

Lyman and Varian (2003, Summary of Findings) estimate that “Print, film, magnetic, and optical storage media produced about 5 exabytes of new information in 2002… How big is five exabytes? If digitized with full formatting, the seventeen million books in the Library of Congress contain about 136 terabytes of information; five exabytes of information is equivalent in size to the information contained in 37,000 new libraries the size of the Library of Congress book collections.” Another way to look at this ‘information explosion’ is to note that, in the US, there were 14,948 public libraries in 1991, up from 6,123 in 1935 (Peters 1992, p. 23).

Better-informed decision makers improve on average the quality of decision. All of them become more efficient. This includes new entrepreneurs, who were previously employed as middle managers when information was costly and whose efficiency was below a threshold of profitability as entrepreneurs, and whose efficiency now surpasses that threshold due to the new abundance of cheap information.

Thus more information tends to increase the number of decision makers, and therefore the number of rights holders, or freedoms holders. Personal freedoms are a strict complement to the decentralization of decisions. It follows that the social value of freedoms increases and decreases with the degree of production centralization, which in turn depends on the information abundance or scarcity. Freedoms are a factor of production complementary to the decentralized mode of production.

When the social value (in terms of additional production) of freedom increases, it tends to exceed the social costs of freedom (anarchy, delinquencies, even terrorism). Democratic freedom, in particular, is said by some²⁵ to entail a high cost in terms of resources invested in complex political coordination problems, and also in terms of increased redistributions that determine social losses (rent-seeking).

Political factors regarding redistribution will also play a role. In particular, the existence of an authority having a monopoly on the use of force is essential in defining and implementing freedoms. Governments do define and legitimatize freedoms in society. Furthermore, governments are in a position to use their monopoly on organized violence in order to defend these freedoms.

Let us assume that, in fact, governments do produce personal freedoms to answer individuals’ demands, allowing the latter then to use these rights to increase their production. These governments, even if they are of the ‘Leviathan’ type, will find it beneficial to concede freedoms to individuals when the optimal production mode is decentralized, that is when information is particularly abundant and where its more intensive use is more productive (because a decentralized production process is more information-intensive than a centralized one). The government’s interest here is in the tax revenue it will extract from the additional production resulting from additional freedom and more decentralized productions processes.

We will again find higher supply and demand for personal freedoms in periods and places where information is abundant. In sum, the development of a free society must follow the development of an information society.

We thus have a common, universal factor, information, which can explain waves of increase or decrease in freedom and thus of democracy in all countries whatever their past institutional history. But it is clear that countries that inherit a democratic organization will go farther towards new freedoms in riding the information wave than those nations that inherited an autocratic structure.

There are, however, other factors specific to individuals and to countries that can explain observed differences in levels of freedom and differences in the types of freedom available.

3.6 Differential demands for freedom and country-specific effects

For individuals, the demand for freedoms will depend, like demands for all other goods or services, on income and relative price or relative cost.

Not all individuals however value the ability to decide for themselves equally, depending on their human capital and occupations. Specialists, professionals, will value the decision-making power more because they can make efficient use of the expensive information needed to decide. They will demand more freedom than low human capital individuals who are usually in a position of subordination in the productive hierarchies (Rosen 1982). Their jobs are designed to require few choices, and their superiors in the hierarchy supply them with the information needed for the choices that remain.

It follows that the distribution of human capital in a given population must affect the demand for freedom, and thus the average freedom level as well. Since the level and distribution of human capital will vary within the population and between societies, the existence of a common factor (the relative abundance of information) in our approach does not imply that each wave of technological progress, while generally available to all in principle, will affect all countries simultaneously, or even with the same intensity. Indeed, the specific (political and military) history and geography of each country play a significant differentiating role. For instance, the informational, technological and institutional changes from the first freedom wave of 1750–1875 had a greater impact in Europe than in Japan. In that country, during the Tokugawa era (1603–1868), efforts were made to restrict the acquisition of foreign knowledge and censorship was prevalent. Similarly, in China and Iran today, the government attempts to censor and control citizens’ access to the Internet. Therefore, our explanation for the determinants of the general waves of freedom, then of ‘unfreedom,’ and then of freedom again, does not deny the existence of heterogeneity, or country-specific effects. These country-specific effects should supplement our theory which concentrates on broad common factors and general trends.

Obviously, country-specific factors of heterogeneity such as the tax extraction capacity of a government, the dimension of the country, the relative importance of its extractive industry (oil and raw materials, see Ross 2001), its economic and cultural homogeneity, the language used, its commercial openness, and the distribution of human capital, contribute to explaining the level of freedoms.

But these structural factors are not likely to fluctuate over time simultaneously in very different countries, with a sufficient order of magnitude, to explain in the main the general ‘waves’ of freedom and unfreedom observed since the 18th century.

4 Conclusion

We have presented a theoretical model of the demand and supply of rights and freedom, thus providing an answer to Barro’s demand for an explanation of freedom more solidly grounded in theory. Our model does not call upon the income-freedom effect whose theoretical weakness and empirical contradictions have been underlined in the literature, but instead focuses on the abundance of information and its effects on the decentralization of production, and thus of decisions.

For a given level of income, freedom is greater when the optimal organization of production is decentralized. The main determinant of economic decentralization—and we specifically suggest the abundance and low cost of information—contributes to explaining some of the cross-country differences in freedom levels. But, more importantly, it accounts for the alternate waves of freedom and unfreedom, the variations in the number and sizes of nations, and the changing intensity of their territorial competition, so characteristic of the modern era.

Finally, our analysis has important consequences for political activism: ours is not a Panglossian view, because past institutions and the various pressure groups that control them, can slow down or even prevent adaptation to the optimal organizational structures. Thus individual efforts and a process of ‘tâtonnement’ towards an equilibrium solution are necessary, such that, at each point in time, organizations are not strictly optimal and activism plays a role in the search for optimization.