The More Democracy, the Better? On Whether Democracy Makes Societies Open

Abstract

It is a common view that Popper’s defense of the open society has been a defense of Western, liberal democracies. This seems to imply that by fostering democratic institutions we are ipso facto fostering open societies. I criticize this view by arguing that in-built incentives in democratic mechanisms move us away from (or hamper) the open society. Democracy promotes voters’ ignorance, indulges voters’ irrationality, and allows voters to externalize costs. This is contrary to well-informed, rational decisions and personal responsibility that lie at the fundamentals of the open society. I suggest that it has been free-market capitalism, or free-market societies, which has moved us closer to the ideal of the open society and which best realizes open society’s values.

1. Introduction

When Karl Popper’s The Open Society and Its Enemies1 saw the light in 1945, it went almost unnoticed by most political philosophers of the time. Even worse, some received his work with scorn (see, for instance, Eric Voegelin’s and Leo Strauss’ criticisms, Emberley and Cooper 1993). Notwithstanding this, it is also true that his book, and the very concept of the ‘open society’, has had a tremendous influence outside of academia, mainly in the general public (for instance, George Soros, Helmut Schmidt, Edward Boyle, among others). The terror of fascism and Nazism was still alive, and Popper made a rhetorically convincing effort (although, not a scholarly, accurate one) to trace back the roots of totalitarianism to its intellectual basis, back to Plato, then G.W. Hegel and finally Karl Marx. The confrontation between the West and the Soviet Union became more agitated in the years following The Open Society’s publication, after a short peaceful period after the war, offering a kind of intellectual guide for Western values’ defenders. Even though the bleak nightmares of the communist regime were not completely known until the 1960s, it is clear that Popper’s description and criticisms of closed societies very well applied to Soviet-style communism as well. For these reasons, and others, The Open Society was viewed, and cherished, as a systematic defense of Western societies and their representative democracies. Even today, almost eighty years after the publication, Popper’s The Open Society can fairly be considered an overt defense of Western-style democratic governments. To put an example, after the COVID-19 pandemic, some scholars relied on Popper’s work to defend liberal democracies and the open society against what they contended as an intolerable advancement of closed, anti-liberal, and anti-democratic tendencies (Esfeld 2021, 2022).

One of the drawbacks of The Open Society is that it leaves crucial concepts severely underdeveloped, notoriously those of “open” and “closed” societies. We are probably in the face of those cases in which we recognize them when we see them, but nonetheless, it is not easy to draw clear concepts from Popper’s work. After Popper’s work, open societies have been identified with societies that seek to strengthen personal responsibility, individual decisions, market mechanisms and a pluralist society; closed societies, on the other hand, go in the opposite direction. Even though these are not Popper’s criteria (they will be developed below), they do seem to share the common spirit of an open society. In concrete terms, the Swiss socio-political organization is much closer to the ideal of an open society than Venezuela’s dictatorship. But there are other cases in which the line is blurred. As an example, Welfare democracies as those of the Scandinavian countries are open societies in some respects (e.g., civil rights), but closed in others (high tax-burden). Even more ambiguous is the articulation between the open society and democracy. Popper himself defended the open society on the grounds of Western liberal democracies, “the best of all political worlds of whose existence we have any historical knowledge” (Popper 1999, p. 90). It hence seems that democracies (in their representative and liberal form) are the best political-social structure to foster the development of open societies. The features and values we find in the latter find their institutional translation in the former. The Economist, for instance, claimed that The Open Society “remains the best intellectual defense of liberal democracy” and many scholars have taken liberal democracies as equivalents to open societies (see Pralong 1999; Ignatieff and Roch 2018; Landemore 2020). And they have good reasons to do it so: Popper himself seemed to have blessed this close-knit association.

In this paper, I am not interested in an exegesis of Popper’s work. Nor am I interested in providing a more thorough definition of the ‘open’ society that exceeds The Open Society itself. My claim here is rather more general and (a bit) controversial: Popper’s characterization of the open society is good enough, but I challenge the idea that democracy is its best institutional instantiation.2 Even more radically, I believe that real democracies (at least as they have worked in the real world) undermine some of the valuable features of open societies. To show this point, I draw on recognized results in public-choice theory when applied to collective democratic decision processes. So, the aim of this paper is to draw Popperians’ attention toward the tension that exists between democracy and the open society when some results of the public-choice literature are taken seriously into account.

This tension, I hold, leads us to revise the presumed straightforward relation between open societies and democracy. While I do think that there are open societies (or, better, societies that are more open than others), I cast doubt on the idea that this is so because they are democracies. Existing relatively open societies, I claim, are open societies not because they are democratic, but despite being democratic. The element that I find most relevant to assess the openness of a society is whether it enacts mechanisms of social cooperation based on the respect of private property and free markets (which I call “free-market capitalism”). Popper himself calls for (state) ‘protectionism’ against the potential dangers of free-market capitalism. I think he is not right on this: While he overemphasizes the role of democratic institutions for a society to be open, he neglects the positive incentives and features that free-market capitalism (or free markets more generally) brings to open societies.

My argument has two parts. First, I argue that real democracy contains negative incentives that undermine the open society’s main values. In a democracy, political incentives make citizens ignorant, irrational, and prone to externalizing their personal responsibility. This opposes the basic values of the open society as they rely on well-informed judgment, rational decision, and personal, individual responsibility. Second, I argue that free markets in general (or free-market capitalism, in particular), rather than democracy, enhance the positive features and values of the open society. In addition, free-markets, and not mere democracies, are more suitable to solve the socio-economic problems from which open societies suffer (according to Popper, at least). The defense of open societies is a noble cause. But this must be better seen as a defense of free-market capitalism, rather than Western democracy.

2. The Open Society and Western Democracies

I have said previously that Popper leaves the concepts of ‘open’ and ‘close societies’ underdeveloped. This is perhaps unsurprising since Popper, epistemic and methodologically, opposes what he calls “methodological essentialism”, favoring “methodological nominalism” instead (Popper 2020, pp. 29–30). In a nutshell, methodological essentialism aims to describe the inner nature of things and phenomena by fixed, changeless definitions. The goal of scientific inquiry is thus to reach such “essences”. Methodological nominalism, on the contrary, aims to describe things and phenomena behaviors in different circumstances, searching for some regularities and explanations of such regularities. In this context, it would be unfair to demand of Popper a definition that latches onto the “nature” of the open society, or that unveils its essence. He does not do that but offers instead scattered descriptions here and there, highlighting different features that all together deliver a big picture of what an open society would be like. In what follows, I will be more systematic than Popper to offer a fair description of open societies, without providing an essentialist definition.

One of the central features of open societies is the place for rational reflection. In one of the first descriptions of the open society, Popper says that while a closed society is a “tribal” society that submits itself to magical forces, traditions, and taboos; an open society “sets free the critical powers of man” (Popper 2020, p. xliii). This epistemic theme runs throughout the book. Further on, Popper insists on the magical and supernatural beliefs that underpin closed societies, where taboos in different facets of societal life give it some rigidity that restrains change (Popper 2020, p. 164). The point here is that closed societies arrest any form of rational evaluation of such beliefs and superstitions. Tribal societies, for instance, are unable to channel individuals’ critical examination of the principles of a society’s tradition, which hinders change. A closed society thus requires its members to blindly follow its rules, so to speak. What is novel in open societies is that they do generate a sphere in which critical examination of taboos, magical superstitions, and traditions can be rationally assessed. Then, the functioning of the open society, and its sustainment, somehow presupposes that individuals will be free to rationally reflect on society’s values and beliefs. This can be seen in the central place that scientific reasoning would have within open societies: where individuals may exert critical evaluation of others’ beliefs, hold their beliefs conjecturally, and be free to try new ideas, the scientific method will flourish (see Lane 2020).

Another crucial feature of the open society and that somewhat follows from rational reflection is that individuals are charged with personal responsibility for their freely made decisions. Popper says that in open societies “individuals are confronted with personal decisions” (Popper 2020, p. 165), which are based upon rational reflection and evaluation of one’s acts and their consequences. In reference to the “strain of civilization”, to which I will come back shortly, Popper says that closed societies go well with the “organism” analogy, in which members “are tied together by semi-biological ties—kinships, living together, sharing common efforts, common dangers, common joys, and common distress” (Popper 2020, p. 165; also pp. 166–67). In some sense, members of closed societies cannot be fully responsible for their actions because what they do is ruled by these “semi-biological ties”, by the necessities of the whole. But, in open societies, these ties are (almost completely) broken and individuals must face the consequences of their actions, on the basis of their rational personal responsibility. What does it exactly mean “to be responsible” for one’s actions? Popper does not say much, but an economics-inspired answer might help.

Individuals engage in social, cultural, and economic relations with their peers. They might exchange products, ideas, services, or whatever they want. Most of these exchanges have costs and benefits, which are not necessarily monetary. However, the basic fact is that individuals engage in cooperative relations to interchange this with that. A call for individual responsibility (or for individuals to face the consequences of their personal decisions) can be viewed as a call for the internalization of both costs and benefits that these interactions carry over. An investment to start up a new business, for instance, is a risky enterprise that entails a series of these interchanges. They may come with great benefits, which an individual internalizes as gains if the business goes well; but if the business does not go well and eventually ends up bankrupt, personal responsibility demands that costs be also internalized.3 Open societies must allow for mechanisms to enact cost-benefit internalization. Otherwise, personal responsibility, or facing the consequences of one’s actions, cannot really take place. Think of organic, closed, tribal societies as Popper depicts them. Since personal decisions must be assessed in terms of the whole and are based on a non-rational basis (e.g., “this is what the gods want”, “this is the best for the Race or the Class”), there are not really cost-benefit internalizing mechanisms, but pure externalization. Bad personal decisions are paid off by society; good personal decisions are cashed in by society. No place for responsibility for one’s doings.

It can be argued that so much stress on personal responsibility leads to an individualist view of society. There are intrinsic reasons for doing that –personal agency can be valuable in itself, mirroring one’s identity (Hurka 1987). However, the most common approach to personal responsibility is in terms of accountability. For instance, some egalitarians have criticized an only-personal-responsibility view on the basis of equality (Robinson 2014). Is it justified to hold individuals accountable for bad decisions even if it leads to high levels of inequality? Some egalitarians, responsibility-sensitive egalitarians, have nonetheless argued that it must be possible to have personal responsibility and equality (see Mason 2006; Robinson 2014). The reason is clear: modern philosophy has made personal responsibility the bedrock of morality, so it is not easy to grasp how persons can be morally accountable for their actions without a substantial concept of personal responsibility. I share this view, but I would go a step further. Criticisms of personal responsibility frequently come from egalitarian considerations, but I am skeptical that egalitarianism is compatible with Popperian open societies. After all, as Nozick has argued, liberty upsets patterns (Nozick 1974, pp. 160–64), so egalitarianism usually leads to interventions in people’s lives and decisions that can distort open societies’ values. Blurring the idea of personal responsibility too much could end up with an incentive structure in which too many costs are socialized (externalized). This ultimately leads to increased misery for the majority. So, in the end, an only-personal-responsibility view leads to social positive consequences, even though the starting point is only personal responsibility, looking too individualistic.

One of the worrisome corollaries of the central place of personal responsibility in open societies is the emergence of a certain degree of abstractness in social relationships. Popper says that the loss of the organic features (and of semi-biological ties among members) of closed societies can produce losing “the character of a concrete or real group of men, or of a system of such real groups” (Popper 2020, p. 166). This “less concrete, less real” society is what Popper calls an “abstract society”, in which depersonalization, abstract (or virtual) relationships and a lack of genuine personal relations with our peers increasingly take center stage. Even though Popper clarifies that complete abstract societies would never be possible and that it is probably an exaggeration, he does remark that open, modern societies exhibit at least a tendency in that direction, partially realizing their features. This interestingly introduces a disruptive element in open societies, a cost to be paid for the gained freedom. Popper talks about “the strain of civilization” to refer to this phenomenon, which follows from the breakdown of closed societies (and all its consequences) as well as from the leading role that individual responsibility takes on open societies. He says:

This strain, this uneasiness, is a consequence of the breakdown of the closed society. It is still felt even in our day, especially in times of social change. It is the strain created by the effort which life in an open and partially abstract society continually demands from us—by the endeavor to be rational, to forgo at least some of our emotional social needs, to look after ourselves, and to accept responsibilities. We must, I believe, bear this strain as the price to be paid for every increase in knowledge, in reasonableness, in co-operation and in mutual help, and consequently in our chances of survival, and in the size of the population. It is the price we have to pay for being human.

(Popper 2020, p. 168)

So, according to Popper, open societies enjoy more freedom, offer a wider place to develop our capacities, make up our minds about many things, and act rationally. But, on the other hand, it seemingly leaves us in greater insecurity, more isolated, and more needy of emotional and social relations.4 But, this ‘abstractness’, and the lack of tight social cohesion or support, is one of the features that open societies seem to exhibit.

Finally, and more importantly, open societies are democratic societies and the model to follow is Western representative-liberal democracies. As was mentioned in the introduction, many scholars have seen in The Open Society an overt passionate defense of democratic governments in its crusade against totalitarianism. The point is that Western democracies, according to Popper, provide the best answer to the question that Popper believes political philosophy should gravitate around. While political philosophy traditionally centered on questions like “Who should rule?”, Popper thinks it must be replaced by “How can we organize political institutions so that bad or incompetent rulers can be prevented from doing so much damage?” (Popper 2020, p. 115). That is, democracy is the best institutional answer to the design of political power that must be essentially “checked”.

This approach discards one of the naïve theories of what democracy is: “the Rule of the People”; or what high school has lectured us: “the government of the people, by the people, and for the people”. Popper affirms that democracy is an adequate answer to the political question that matters, but that it is not true that “the People” rule in a democracy. Nowhere have “the People” ruled, but governments have. Nor does Popper believe that democracy has any intrinsic value. In political philosophy, supporters of democracy generally align in two camps. Those who defend democracy on instrumentalist grounds (democracy is good because it leads us to realize some valuable aim, see Arneson 2003), and those who do it on proceduralist grounds (democracy is good because it has intrinsic value, see Griffin 2003)5. Popper clearly aligns in the first camp as democracy mainly plays an instrumental role in realizing, and enhancing, the values of the open society. Therefore, democracy is an instrumental answer to a question that mainly concerns institutional design: democracy offers the means to organize political institutions that cause less harm to people.

Besides these points, Popper offers a few hints about how he understands democracy more specifically. It is true that democratic institutions, and theories of democracies, wildly vary. While some have seen deliberative democracy as an ideal to pursue and to realize true democracy (Cohen 2003, 2009), others have argued for republican, representative forms (Locke 1988). Some democracies are semi-direct (as in Switzerland), others are purely representative (as in Argentina). It is not my aim here to discuss alternative theories, so I will not dwell on this too much. But there seem to be at least three features that are somehow ubiquitous. First, democracy, in general, applies majoritarian rule to various decision-making processes. This rule can be limited in several ways (e.g., by constitutional rights), but within the constraints, it is the mechanism that is in general followed. Second, democracy possesses the mechanisms to allow citizenship to participate in (relevant) decision-making processes. Of course, this also operates within limits. In representative, democratic presidentialism most citizens are allowed to elect the president, but not the Central Bank’s interest rate for the next semester (not even the Central Bank’s director). Also, not everyone is allowed to participate (kids are, in general, out of the central political decisions). But wide participation seems to be an important feature. Finally, democracy does not in principle put any barrier to running for governmental offices. Any citizen, again at least in principle, is allowed to run for any office. This contrasts sharply with other governmental forms, such as monarchy or oligarchies.

I think that, to a greater or lesser degree, most Western democracies instantiate these features (although some adhere to a parliamentarian, while others to presidential) and probably Popper himself had many of these features in mind. So, democracy seems to be the best institutional mechanism to enact the values of open societies. It does so by providing the institutional design for citizens to rely on majority rule as a decision-making process for important issues. It allows (almost) everyone to participate in these decision-making processes. And it allows (almost) everyone to run for offices that will be filled by such decision-making processes. Citizens thus have the right to an equal share in political power. Most Western democracies satisfy all these features, although imposing several constraints on what majorities can choose, aiming to protect individual rights and minorities.

Beyond these aspects, others are important for open societies. First, Popper points out that democratic governments are in agreement with the scientific method.6 Democracies enact trial-and-error mechanisms to select leaders, they are not based on absolute truths or on knowing the essence of the Common Good, the Just, and so on. Democratic procedures, he says, are based on conjectures. Popper is not very explicit about it, but his conception of democracy (and its mechanisms) seems to go hand-in-hand with his epistemological considerations of science and with his critical rationalism (Lane 2020; Royer 2023). Political and social knowledge should be like scientific knowledge in the sense that progress can only be made by a trial-and-error institutionalized process. In some respects, democracy is the only system that best matches the scientific method when it comes to conceiving institutional and social design.

This aspect is also closely related to another integral part of Popper’s political thinking. Totalitarian regimes (par excellence models of closed societies) are characterized by what Popper calls “Utopianism” or “Utopian social engineering”, that is, the implementation of deep social transformations to realize the Utopia of an ideal state on earth (totalitarianism being one of the most radical evolutions, see Bernholz 2017 for an analysis of the evolution of totalitarian regimes through history). That is why Utopianism, totalitarianism, and central planification go together. In contrast to this view, Popper proposes a “more democratic way” to intervene in society, through a trial-and-error method that he calls “piecemeal social engineering”, a mild form of state interventionism. For Popper, the main aim of his piecemeal social engineering is, for one thing, to design institutions that foster and protect open societies and their values and, for another, to alleviate suffering and unhappiness in the worse off. It is not the case that politicians can rely on a blueprint of society for complete planification. But they should have at their disposal mechanisms “to make adjustments in the light of discussion and experience” (Popper 2020, p. 340). These adjustments must be local, and small, aiming at reforming institutions slowly and addressing concrete social problems. Democracy and piecemeal social engineering are compatible as democracy guarantees that the social planners (i.e., politicians) can receive feedback from society about the piecemeal adjustments they execute, emulating a social experiment.

Popper repeats many times that the aim of social engineering, under democratic institutions, is to address social problems that emerge from competition, failed cooperation, and the division of labor. Small interventions, therefore, are justified on utilitarian grounds in order to reduce overall unhappiness or suffering. What democracy guarantees in this quasi-scientific process of institutional and social design is that citizens are able to remove bad leaders or planners without bloodshed. That is, democracies permit not only citizens to give feedback to leaders in a transparent way (e.g., in the ballot box), but also a change in government peacefully. Governments of all kinds have been full of bad, corrupt, and even cruel leaders. Changes in government in the majority of systems have involved wars, violent revolutions, and citizens’ blood shedding. But democracy is different not because democratic leaders will be more virtuous or wiser (or less corrupt or cruel), but because democracy possesses the mechanisms to get rid of no-longer-wanted leaders by casting a vote in a ballot box. It is this peaceful mechanism of retrospective punishment that makes democracy, according to Popper, so useful for open societies.

To sum up, open societies exhibit most of the values of classical liberalism, with emphasis on individuals (rather than collectives) and freedom (rather than social security or tight cohesion). I have pointed to four aspects of open societies that Popper himself in particular remarks on rational reflection, personal responsibility, the abstractness of social relations, and democratic values. Setting aside the abstractness (see fn. 3), Western democracies seem to be the best institutional support for open societies since they enhance and foster open societies’ values. Thus, a defense of open societies is a defense of Western democracy, and many scholars have taken it in this way. In what follows I argue against this association by showing that there is a tension between the most salient values of open societies and democracy. In particular, I argue that Western democracies (at least as we have them and as we could realistically hope to have them) do not foster open societies’ values in a special way. Even worse, many aspects of their institutional scaffolding undermine some of the cherished values of open societies as Popper describes them. This would undermine the link between open societies and democracy.

3. Democracy vs. The Open Society

My main thesis is that Western democracies, in the end, undermine open societies’ features and values. Before getting down to the arguments, let me explain the limits and scopes of this thesis. As I mentioned before in passing, in practice, most Western democracies are (relatively) open societies. The United States during the Cold War was much more open than the Soviet Union or Cuba. Sweden today (a social-welfare democracy) is more open than Iran (a conservative theocracy), but probably less than Chile (a representative liberal democracy)7, which is more open than the rest of the South American countries (some of them are on the verge of socialist dictatorships). It then seems to be empirically true that there is a continuum between “ideal open societies” and “ideal closed societies”, with reality lying in the middle. Societies (governments and countries) occupy some place in this continuum, but there are neither pure nor permanent forms. For instance, governments in the West have been moving towards welfare states (moving away, strictly, from the ideal of liberal open societies as welfare states socialize, to some degree, personal responsibilities). Nonetheless, they still remain to a good extent open societies (putting into brackets the 2020–2021 COVID-19 period, and some policies on climate change in the European Union) when compared diachronically and synchronically with other societies.

The question I ponder is what moves societies to one extreme (or to the other) in this continuous spectrum. My view is that it is not democracy that moves us toward more open societies, but, in some respects, it can move us away. My point is that relatively open societies today remain relatively open despite many of the vices that democratic forms of government bring with them. The key here is the sort of political incentives that citizens have in democracies to realize the values that Popper appreciates in open societies. So, I am not saying that no democracy is an open society, but that democracy’s incentives and political-social design do not help us move towards openness (or, milder, it is not the main factor that may move us toward openness). The problems that democracy possesses are frequently overlooked not only by Popper but also by most of the Popperians. But, at the same time, if it is not democracy, there must be another non-democratic institution that truly makes open societies open. In this section, I focus on the first part of this claim. In the next section, I will focus on the latter.

I focus here on a cluster of four arguments, all related but conceptually different: first, democracies’ political incentives fed ignorance in voters. Second, they indulge in irrationality when it comes to political decisions. The main consequence of ignorance and irrationality is the externalization of costs, which hampers personal responsibility. Third, empirical evidence shows that democracy is not the best political mechanism to retrospectively punish bad leaders. So, even though democracies do avoid bloodshed, this does not mean that they are effective in retrospectively punishing bad outcomes. And fourth, democracy is not the best way to solve the socio-economic problems that Popper is concerned about. I think these arguments put into question the Popperian idea that democracies are good at enacting institutional designs for open societies. In relying on literature on public-choice theory, I would like to offer a more realistic picture of how democracies work.

3.1. Ignorance at the Ballot Box

It is said that the solution to the ills of democracy is more democracy. The phrase is attributed to John Dewey, although it is probably older. Be that as it may, it seems to enclose a curious paradox for any instrumentalist on democracy. If democracy delivers bad results or performs poorly, the solution is to keep going, insisting on more democracy. What then is the external parameter to assess democracy performance? Ideally, democracy should promote vibrant discussions about diverse policies; voters should not only keep themselves informed about alternative proposals, and all their consequences in a varied range of fields but also be able to rationally discuss them with their peers. All this is in harmony, aiming at the common good. Thus, it does sound appealing to demand more democracy when democracy fails or performs not so well. The problem here is that this view conflates what is ideal with what is real. It employs some ideal form of how democracy ought to work to offer solutions to problems that concrete, more-or-less flawed, more-or-less virtuous individuals face. But ideal democracies are like unicorns. They are capable of great things, no doubt. But it would be unwise to demand actual horses to live up to unicorns’ performance. Real democracies, mutatis mutandis, should not be expected to live up to ideal democracies’ performance either.

One way to assess democracy’s performance is to ponder whether it enhances, or diminishes, the most remarkable features and values of open societies, as Popperians depict them. One of these values is the place for rational decisions and reflection. Are democracies the best institutional design to exercise rational decisions and reflections when it comes to politics? It is assumed that it is, but I am not so sure. Gordon Tullock (1959) was one of the first to point out democracy’s problems. According to him, some collective-decision processes as majority voting are not an optimal way to allocate resources since they can give place to vote-trading. In addition, democracies do not frequently set the right incentives for citizens to be well-informed about politics. Even worse, some of their incentives make citizens more irrational when it comes to politics and policy making. Real people react to incentives. If they are not good, then it is highly unlikely to meet the standards, no matter how noble and positive they are.

Under an instrumental approach to democracies, all that matters is to obtain good outcomes. Of course, what is a “good outcome” could vary, but let us take Popper’s criteria: democracies make available interventionist means to improve people’s lives (less unemployment, less misery, more freedom, etc.), to foster science-driven discussions, and rational public policies. So far, so good. However, it is important to stress that good outcomes (under any standard) frequently come out from well-informed and rational decisions. A well-informed and judicious citizenship would probably reach better outcomes than a poorly informed and irrational citizenship. In democracy, being well-informed basically means knowing the candidates who are running for office (and who are already in office), their political platforms, and their consequences in a wide range of fields. Even more basic, it implies being able to recognize which candidate better represents one’s view. So, if individuals in an open society can have enough room to exercise rational decisions and be rationally reflective to shape their political institutions, it is somehow assumed that they should be relatively well-informed about the decisions they are making.8

Under the ideal assumption that citizens are always (or even in general) well-informed, this should work. But in actual democracies, citizens are in general badly informed. More than seventy years of research in public choice theory and political sciences has shown how ignorant (or ill-informed) voters are on average, in almost all Western democracies (see Converse 1990; Friedman 2006; Somin 2013; see also Pew Research Center’s surveys). From not knowing the political orientation of the current president to the consequences of protectionist policies for society’s welfare, most voters are just ignorant about what they are deciding when casting a vote in the ballot box (see Caplan 2007; Brennan 2016; Jones 2020 for analyses). Several studies carried out in the last decades are as clear as they are bleak—voters are just ignorant about politics; they barely know the consequences of what they are voting for, even if they know what they are voting for. This is certainly off to a bad start for exercising rational decisions and reflection, as the open society seems to require. But is this due to flawed citizens? Do we need to educate them in advance? It seems that no. Economists have explained this phenomenon as a part of a far-more-reaching phenomenon that naturally follows from basic microeconomics, rational ignorance (Downs 1957; Somin 2020).

The rationale is the following. The acquisition of information is always costly. If the acquisition of information costs more than the benefits that the information is expected to confer, then it is rational to remain ignorant, that is, to lack such information. What goes by ‘rational’ in this case is the utilitarian meaning of rationality, that is, a cost-benefit analysis. Let us give an example. I invited my friends for a meal. To prepare the meal, I need to buy supplies. Suppose that there is the cheapest shopping combination within a 50 km radius. If I knew such a shopping combination, my costs for the meal would be lowered by, say, 5 Swiss Francs (CHF). I would, therefore, save 5 CHF if I had such knowledge. The thing is that I do not know which is such a cheap shopping combination and need to get the information: surveying prices in different places, for each product, in a limited amount of time. True, I could save 5 CHF, but the costs of obtaining the information to save them could largely exceed the benefits that such knowledge would confer. Then, I rationally decide to remain ignorant about supply prices elsewhere and buy them all in my closest supermarket.

When it comes to democracy, the political decisions we have to make need to weigh the costs and benefits of becoming more informed about politics. The information that we would need to obtain ranges over a panoply of fields, from economics to international relations and the environment. How could we know which is the best candidate? How could we know which is the best candidate for this platform? How could we know which is the best platform for these desired results? The only answer is that we need to invest time and money to obtain such knowledge. We need time to do some research, to become acquainted with diverse fields, to gather complicated evidence for thorny and controversial issues (e.g., should we ban recreational drugs?), etc. In modern societies, all this has a price that could be worth paying or not, but it is clear that the costs of getting more informed are quite high for each individual. The problem, as it has been pointed out, is that the benefits for each individual are almost negligible. One of the problems is that the democratic majoritarian rule rules after all. I can sweat blood for three years to gather all the necessary information to cast a well-informed vote in the ballot box for the upcoming election. But, after all, the candidate that obtains the majority of votes wins, regardless of whether they were ill- or well-informed votes. As the philosopher Jason Brennan puts it: “A vote makes a difference only if there is a tie; otherwise, it usually does not matter how someone votes or whether they vote at all” (Brennan 2016, p. 31). Hence, the individual benefits of my well-informed vote are negligible, but my costs were quite high (at least, time-consuming high). So, why bother?

Rational ignorance is the best option given the democratic incentives. When voting has such a vanishingly small probability of making a difference in an election, then it is rational to remain ignorant about crucial things in politics (see Landsburg 2004). To be clear, this is not a problem about peoples’ virtues, it is not about schooling people in the virtues of responsible, well-informed citizenship. The problem is the incentive structure of democratic institutions. When democracy extends itself to such a widely varied range of topics (when we add more democracy to democracy), it becomes “more expensive” for voters to keep track of decisions that might involve them quite deeply (and it becomes harder for them to have any influence at all upon them); it becomes “more expensive” for them to obtain the information to know what to do when an election is approaching. The basic point is that democracy is driven by majoritarian rule and when majorities are extremely large (millions of individuals), the chances for each individual that his/her vote matters are ridiculously small. Once again, why bother?

Almost everyone today acknowledges rational ignorance in democracies, so it is pointless to neglect it. It can be thought that this is a temporary state of affairs that democratic institutions would correct with time. Democracy will educate their citizens and they will eventually become more committed to democracy and better informed (Popper seems to have endorsed something close to this). However, the evidence speaks against these hopes (see Delli Carpini and Keeter 1996). Another plausible solution to voters’ ignorance is more theoretical, the so-called “Miracle of Aggregation”, which has become a sort of ‘conventional wisdom’ among defenders of democracy (Converse 1990; Page and Shapiro 1993). It allows the harmonization of empirical evidence and solid grounds for democratic institutions. I will not dwell on this too much, but the idea is basically the following. The overwhelming majority of voters are ignorant about politics, which is indeed a problem for good policy making. But it is enough to assume (which is not so implausible) that a minority of well-informed voters exists, and they will ultimately decide the results of democratic decisions. Then, everyone has a share in political power, majoritarian rule still rules, but it is guaranteed that good policy making will succeed by virtue of a minority of well-informed voters that will tilt elections in the right direction. So, suppose that 98% of voters are fully ignorant. This means that they vote randomly and that there are no systematic biases. But let us also suppose that 2% of voters are well-informed. Then, in the face of two choices, the majority of votes will be distributed almost equally, but the 2% is bound to go for the right choice. Therefore, the right choice is bound to win (see Caplan 2007 for criticisms).

This is probably not what Popper had in mind when seeking an institutional translation of the open society. Nor is it probably what most democrats have in mind when thinking of the workings and virtues of democracies. After all, the place for rational reflection and responsibility, scientific thinking, and well-informed policy decisions is quite circumscribed. Be that as it may, it saves the foundations of democratic institutions from pervasive ignorance (voting randomly). And, after all, it would remain true that open societies are democracies, even if the details are not as nice as originally thought. Yet, in the last decades some economists, political scientists and philosophers attacked one of the assumptions of the Miracle of Aggregation: voters are mostly ignorant, but they do not commit systematic mistakes (e.g., they are not biased). This seems to be untrue.

3.2. Irrationality at the Ballot Box

Economist Bryan Caplan has famously argued that voters in most democracies are not only ignorant, but they also commit severe systematic mistakes (Caplan 2007; Huddy et al. 2013; For empirical evidence, Caplan 2007, chp. 2, 3 and 5). He argues that people suffer from many biases (anti-market bias, anti-foreign bias, make-work bias and pessimistic bias) that transform ill-informed votes into misguided or irrational votes. This does not mean per se that people are overall irrational and that democracy (no matter how lofty the idea is) cannot work because we are intrinsically flawed (i.e., we are too irrational). Once again, the problem is about incentives. In daily life, most people are quite able to make rational decisions when it comes to practical things, where costs and benefits are inevitably internalized. As Fréderic Bastiat once claimed, “each man is in practice an excellent economist, producing or exchanging according as he finds it more advantageous to do the one or the other” (Bastiat 1964). Rational choice approaches in modern economics and political science support this view as well. So, how so? How can individuals be overall rational in their daily lives (pondering costs and benefits when, say, buying a house), but so irrational when it comes to casting a ballot into a box?

Caplan refers to this phenomenon as rational irrationality. Once again, the burden is not on individuals, but on the incentives that democracies set out. He argues that individuals value both material prosperity and their worldview, that is, an ideology that causes some degree of pleasure when endorsed. So, when voting and balancing costs and benefits in a democracy, individuals face a two-argument utility function: a political ideology that makes them happier and their material wealth. So, the total costs and benefits of voting are nothing but a tradeoff that prices a material wealth foregone and an ideology we want to believe in (Caplan 2007, p. 17). The point, as Caplan argues, is that in democracies the price to be paid for indulging our ideology is almost zero: when voting, individuals risk almost nothing since their votes are negligible to make a difference; but they obtain a lot of “ideological satisfaction” by voting for their preferred party, their preferred policy, or their friends’. In economic terms, “the price of ideological loyalty is close to zero. So, we should expect people to ‘satiate’ their demand for political delusion, to believe whatever makes them feel best. After all, it is free” (Caplan 2007, p. 18). If we associate the demand for ideological loyalty with irrationality, then the price of irrationality is too low, and its demand can be easily satisfied.

Rational irrationality makes coherent something that Joseph Schumpeter said in the 1940s:

The typical citizen drops down to a lower level of mental performance as soon as he enters the political field. He argues and analyzes in a way which he would readily recognize as infantile within the sphere of his real interests. He becomes a primitive again.

(Schumpeter 1946, p. 278)

Microeconomics is at the rescue, again. This is not surprising when it is acknowledged that democracy lowers the costs of being irrational almost to zero. The incentive structure of democracy sets the costs of irrationality too low. After all, our influence in policy-making, as individuals, is (almost) negligible. When buying a new house, we cannot permit ourselves to be irrational. Any cost will be internalized, and we will pay for it sooner or later. But if we endorse some particular ideology that makes us happier as individuals, we can allow ourselves to be irrational, cast a vote into the ballot box that would satiate us, and externalize the costs of our preferred policy. In a controversial claim, Brennan argues that casting misinformed or irrational votes into a ballot box is like polluting the environment (Brennan 2009): we obtain what we want at a low cost, but we externalize the negative results. Setting aside whether the claim is fair or not, it makes sense to treat democratic institutional design as a “public good problem”. Achieving good public policies is an aim that, in democracy, depends on having a sizable well-informed and rational majority of voters. Yet, for each individual, the costs to be well-informed are too high, and the chances to make a difference are too low. In addition, our ideological preferences are also very valuable, and their costs are too low. So, it is rational for each of us individually not only to be ignorant but also to be irrational. So, in democracy, the incentive structure for each of us is not to be well-informed and rational, but the opposite.

These two last sections show clearly where the tension lies when open societies’ values are faced with how actual democracies work. One of the most hailed values of open societies is the public space that individuals enjoy for them to make rational decisions, face individual responsibilities, and be overall critical of engrained traditions and ideologies. If democracies are the best institutional design for open societies, then they should set out an incentive structure that fosters and enhances these values. Democracies should lay out the institutional scaffolding for individuals to become more rational and to internalize responsibilities (costs and benefits) for their decisions. In a nutshell, democracies should produce good outcomes aiming in this direction, if they support and promote open societies. Yet, empirical evidence and arguments from public choice theory show that democracies perform very poorly when it comes to producing good outcomes in this sense. First, it makes ignorance rational since the costs of becoming well-informed about relevant political details are too high, and the benefits too low (rational ignorance, Section 3.1). Second, it makes irrationality too tempting because it sets a near-zero price for it (rational irrationality, Section 3.2). Then, far from generating well-informed and rational voters, it generates the opposite. If this is so, democracies generate a citizenship that has very low chances to produce good outcomes, in the sense of political institutions that enhance rational decisions and responsibilities. Therefore, democracies do not support but undermine open societies.

A final comment is in order. It could be argued that the blame is not on democracy but on individuals (or any other socio-political institutions). Democracy would support and foster open democracy if people were not ignorant and irrational to begin with. To put it differently, if people were more democratically virtuous, then democracy would succeed. This is as true as it is irrelevant. When tailoring political institutions, it is important to tailor them for actual societies, not for ideal societies. The problem, once again, is not people’s vices or people’s irrationality. People are varied, sometimes virtuous, sometimes not. But different contexts may enact their virtues or their vices in different ways, depending on the incentives they face. Institutions that can only work for ideal individuals will only cause harm because they are bound to set out the wrong incentives for non-ideal people to develop their capacities, diminish their vices and enhance their virtues. This is once again the unicorn metaphor. We cannot treat actual, in-comparison-flawed horses as if they were unicorns. This would eventually kill them. The same goes for political institutional design.

3.3. Consequences of Voters’ Ignorance and Irrationality

If voters are to a good extent ignorant and irrational about politics, many of the democratic mechanisms that its defenders cherish do not work as well as they believe (mainly on an instrumentalist basis). These features of the average electorate have far-reaching consequences that do not render democracy a nightmare, but that make it less fit for an open society.

One of the features that Popper attributes to democracy is its peaceful mechanisms to punish bad outcomes, corrupted politicians, or cruel leaders. By voting regularly, citizens can enact mechanisms to correct the course of governments if they want to do so. So, democracies can correct themselves through open, periodic and transparent elections where citizens can punish retrospectively bad leaders. This is, according to Popper, what makes democracies so valuable: they allow changing course without bloodshed.

This is a general formulation of what scholars have called ‘retrospective voting’ in voting theory and political science (Key 1966; Fiorina 1981; Lanoue 1994). The idea is that citizens when voting reflect on the performance of the party (or the politician) in power during the last immediate period. For instance, if the economy performed well, retrospective voting dictates that voters will likely elect the same party in elections; if the economy performed poorly, then voters will unlikely elect the same party in elections. Even though retrospective voting focused almost exclusively on economic outcomes, it has recently been extended to other variables such as war, corruption, or social welfare, in general. So, when voting, individuals would not only make choices based on what they believe will be the consequences of electing that candidate (prospective voting), nor make choices purely on ideological grounds (party-line voting), but they would also punish (or reward) bad outcomes (good outcomes) (retrospective voting).

Popper’s view that democracy is effective in causing positive changes without bloodshed assumes two things. First, that retrospective voting is relevantly enacted during elections. Even though other elements might also intervene (e.g., some voters will always align with their party, no matter past performance), retrospective voting makes a difference. Second, that retrospective voting actually works, that is, it is successful in identifying the causes and those responsible for bad outcomes. Ideally, retrospective voting seems to have a good incentive that counterbalances politicians’ individual interests. Most politicians want to be re-elected and stay in power as long as they can. But if retrospective voting works, then they have an incentive to take public interest into consideration because they know that if they do not do it, they will be voted out of office in the next elections. It can also be an argument that counterbalances voters’ ignorance. Individuals might be ignorant about all the details of public policies, but they know very well their pockets: an empty pocket signals a bad economy. Retrospective voting is thus enacted due to these signals.

Unfortunately, in the real world, retrospective voting is not as effective as it is believed. In consequence, although democracy does seem to work in avoiding violent eruptions (when compared with other regimes), retrospective voting is too limited. Christopher Achen and Bartels (2016) claim that

While we attach great importance to the realism, empirical power, and normative appeal of the retrospective theory, we believe that its implications for democracy are less unambiguously positive than existing literature tends to suggest.

(Achen and Bartels 2016, p. 92)

One of the reasons they give is that it is not obvious that voters know the relevant facts to assess the incumbent political leaders. In the end, retrospective voting shows that election outcomes are frequently erratic without being clear that it outbalances other intervening elements. Although this is a society-relative fact, some voters, for instance, simply exercise party-line voting, that is, they vote ideologically. Brennan talks about “hooligans” to name those voters with strong and fixed worldviews and who are deeply biased about the pros of alternative positions, and the cons of theirs (see Brennan 2016, p. 5). Most voters are not ideological in this sense but because they care little about politics (Brennan refers to them as “hobbits”). The point is that highly ideological voters will have the individual incentives to intervene more in politics, while less ideological voters will not. Contrary to the Miracle of Aggregation, it is not a well-informed and rational elite who would tilt an election in the right direction, but an ideologically biased sector of the electorate would tilt an election in their direction (which could likely be the wrong direction, because irrationality commands). So, the problem is ultimately about how efficient retrospective voting is in punishing (or rewarding) bad outcomes (good outcomes) when most of the population has little interest in politics, and a powerful part of the electorate has so much influence and is so much biased in favor of its view.

Second, it is not so clear that ignorant and irrational voters are able to identify the right causes and responsibilities of bad outcomes. One of the positive implications of retrospective voting is that it would promote democratic accountability (Achen and Bartels 2016, p. 103). For instance, it can be true that voters know about their current economic situation and they know that it has declined in the last few years. But this does not mean that they know the causes of economic impoverishment. If they do not know the causes of economic impoverishment, it is likely that they fail to identify those responsible for it. For instance, years of price controls and overregulation in international commerce will lead to shortages and repressed inflation. If the incumbent politician decides to abolish price controls and deregulate commerce, the consequences of years of mismanagement will be evident soon. But who is to be blamed for it? Those responsible for years of mismanagement? Or the incumbent politicians? If voters are ignorant about basic economic knowledge, it is expected that they will misidentify those really responsible for, for instance, a rapid generalized increase in prices. If voters decide to punish the incumbent politician, and not his/her antecessors, then retrospective voting fails to identify (and punish) the bad politicians. I do not mean that voters’ ignorance or irrationality are knock-down arguments against retrospective voting. Indeed, it does work. But they cast a shadow of doubt on its efficacy and improving capacity. Voters do look back when voting. They do distribute punishments and rewards. But they could fail to do it correctly.

As was said, voters’ irrationality mainly concerns biases that indulge some ideological preferences. Ideologically motivated decisions do not guarantee good outcomes since they are likely to be misinformed and not completely rational. However, in an open society in which individuals face the consequences of their decisions, this would not be a problem. The costs of bad decisions are internalized. For instance, if I have an ideological bias against immigrant labor, my business will pay the costs of having to employ more expensive workers (I am basically constraining the labor supply). Then, I am paying the costs of being biased (irrational) about immigrants. In a democracy, the situation is different. First, voters can indulge their prejudices and biases without paying the costs because majoritarian rule allows the externalization of costs. Suppose that I live under a perfectly democratic government, but my peers for cultural reasons believe that the government must keep prices at bay by pervasive governmental controls (as in Venezuela or Argentina in the last few years). Despite arguments and evidence pointing to the inefficacy, and bad consequences, of price controls (see Schuettinger and Butler 2014), they insist on voting for governments that apply price controls for purely ideological reasons. Majoritarian rule dictates that if a majority is biased and chooses government accordingly, it is not the majority only who will pay the costs of the decisions, no matter how vehemently the rest oppose such measures. So, this delivers a dangerous combination that hampers retrospective voting: it is very limited to correct such ideological mismanagements since individuals do not have any incentive to pay the costs of their biased decisions, and therefore, to correct them.

3.4. Democracies and Socio-Economic Issues

Setting aside democratic mechanisms and voters’ ignorance and irrationality, let us move on to a different aspect of democracies that Popper highlights. According to him, democracies should allow local, targeted, small interventions –piecemeal social engineering. The overall idea is not only to improve political institutions slowly and rationally but to implement “negative utilitarianism”, that is, less suffering, less poverty, and less misery. So, democracies should ideally make available state resources not for the maximization of happiness (as Utopian engineering does), but to diminish unhappiness and suffering. That is why Popper says: “it is my thesis that human misery is the most urgent problem of a rational public policy” (Popper 2002, p. 361). In The Open Society, he says: “We must construct social institutions, enforced by the power of the state, for the protection of the economically weak from the economically strong” (Popper 2020, p. 125).

Popper reproduces a common opinion among the defenders of state intervention in democracy. First, societies under relatively unrestrained market capitalism produce friction among individuals, even a “class struggle” (Popper 2020, p. 165). They would also produce high inequality since some individuals will be economically strong, while others will be weak. Finally, such open societies would not have inner mechanisms to alleviate these problems. In consequence, state intervention under democratic institutions aims to be a remedy for these inner problems that appear in open societies. Popper, as many other defenders of democracy, accepts that open societies work better under a capitalist, market-oriented organization of economy, but that state intervention (in its piecemeal, democratic form) must jump in to diminish suffering, to correct the failures that such societies generate. There are two things to be noted in this argument. First, it is the existence of market failures. Second, markets do not realize some idea of social justice (see Hillman 2009, chp. 1). This of course presumes that democracies (or state intervention more generally) are effective in solving these issues (see also Hillman 2009, chp. 2 for a criticism to this presumption).

Popper is not very clear about what he exactly means by intervention, nor what “piecemeal social engineering” concretely is. But a series of arguments suffices to cast doubts on the efficacy of democratic state intervention to alleviate misery and drawbacks of open societies. First of all, what is a ‘democratic state intervention’? It means, prima facie, the employment of coercion by democratically elected government officials to alter some outcomes of civil society (e.g., to reallocate resources). Let us take unemployment as a case. It can be argued, as Popper himself suggests (Popper 2020, pp. 149, 383–84), that unemployment might be a shortcoming of open societies, brought about by the division of labor and competition in market-oriented economies. State intervention in this case can mean two things: the creation of public employment or public unemployment insurance.

In the first case, the state can inject extra public resources into the labor market to mitigate unemployment by increasing public spending (e.g., by employing more individuals in public offices, spending more on public infrastructure, etc.). Indeed, this looks like a state intervention in fighting unemployment. Yet, when we look at the economy at large, this is not so. What state intervention is doing, in this particular case, is moving resources from one sector of the economy to another. The public sector does not generate its own resources, so it must take them from the private sector (e.g., through taxation). An increase in public spending means that more private resources are extracted from civil society by coercive means, resources that individuals could have used elsewhere or differently. Then, the generation of employment by the public sector not only amounts to the destruction of private resources somewhere to be used elsewhere by governments9, but it also destroys the capital structure, resources are not going where they are needed most, but where state’s officials think they must go (for a criticism of unemployment policies based on public spending, see Hayek 1989, p. 5). So, it is not clear that the results of democratic state intervention are overall positive. They also undermine one of the criteria for “openness”: individuals cannot be responsible any longer for the outcomes of their socio-economical interactions.

Unemployment insurance, in the second case, does not depend on the state necessarily as there exist many private unemployment insurance companies (see Denderski and Stoltenberg 2023). Also, public unemployment insurance increases the costs of labor by transferring non-voluntarily (and paternalistically) labor production to public unemployment insurance funds.10

This makes clear a point that is often overlooked: any state intervention (whether democratic or not) requires resources that governments do not create on their own, but that must be taken from the private sector. Then, public resources, in general, amount to an increase in taxation, public debt, higher costs of labor, or money issuing. There is no other source of public resources. Adding the word ‘democratic’ or ‘piecemeal’ does not change the nature of the intervention. In one way or another, it is the private sector (i.e., the open society itself) that provides the means to employ individuals where state officials think they must be employed, that produces the wealth that state officials think must be saved in case of an eventual unemployment period, etc. None of these means ponders whether it is an adequate use of resources or not.

Putting aside whether there could be democratic limits to this (in principle, there are not strong incentives within democracy itself to do this at a large scale), the real problem is that democratic institutions neither are able to obtain the necessary knowledge of the most efficient use of private resources (Hayek 1945; Lavoie 1985) nor of what individuals really want to do with the wealth they produce. So, state intervention, no matter how noble the intentions are, can directly mean not only the destruction of capital (misuse of resources) but also a distorting alteration of capital structure. The core of the criticism is what has been called “the knowledge problem”. As it was formulated by Don Lavoie (1985, p. 86):

Comprehensive planning, the classic doctrine of planning advocates, seeks to achieve economic coordination without relying on the contention of separate decision makers with one another; it thereby deprives itself of access to one of the most important sources of knowledge exhibited by these kinds of orders. Just as in biological competition, there is the ‘information bearer’ function of DNA, so in the society of Tradition, this function is further served by such developments as language and culturally acquired techniques and habits. In the society of Market, profit and loss signals are added to this array. In the society of Planning, there is no new information bearer and those of the Market are discarded. It is this lack that gives the knowledge problem argument its force.

(Lavoie 1985, p. 86)

Lavoie’s point goes directly against one of the assumptions of Popper’s democratic state intervention—in replacing market mechanisms with centralized ones, there is no new information bearer that can signal how and where the resources can be allocated more efficiently. Friedrich A. Hayek made the same point before, although in a more general way, in his “The Use of Knowledge in Society” (Hayek 1945) when claiming that

The economic problem of society is thus not merely a problem of how to allocate ‘given’ resources—if ‘given’ is taken to mean given to a single mind which deliberately solves the problem set by these ‘data’. It is rather a problem of how to secure the best use of resources known to any of the members of society, for ends whose relative importance only these individuals know. Or, to put it briefly, it is a problem of the utilization of knowledge not given to anyone in its totality.

(Hayek 1945, p. 520)

All these points flag the same problem: the unrealistic expectations of what democracies (and democratic state interventions, no matter how minimal they are) can achieve. Hayek himself confesses that “it took me too a long time from my first breakthrough, in my essay on ‘Economics and Knowledge’ (1936/48), through the recognition of ‘Competition as a Discovery Procedure’ (1978, pp. 179–90), and my essay on ‘The Pretense of Knowledge’ (1978, pp. 23–34), to state my theory of the dispersal of information, from which follows my conclusions about the superiority of spontaneous formations to central direction” (Hayek 2011, p. 88).11

This could raise the following question: if democracy allows state intervention (whether to alleviate suffering or to “piecemeally” transform socio-political institutions), is it compatible with the principles and values of open societies? Social democratic engineering is to devise public means to eliminate (or alleviate) socio-economic problems, but this in reality amounts to intervening in individuals’ decisions within the open society. That is, if individuals decide to invest in some sector of the economy, rather than in others, state intervention means a coercive intrusion in these decisions as private resources are moved to public spending; if individuals decide to use their wealth in some ways rather than in, say, an unemployment fund, state intervention means a coercive intrusion with one’s own funds. This puts at risk the room for personal decisions and personal responsibility. But it also puts at risk the internalization of costs in the face of bad decisions. A typical case is when state intervention takes the form of payoffs or bailouts to big companies (e.g., banks). Bankruptcy of big companies could lead to higher temporal unemployment, then public bailouts can be seen as a way to avoid unemployment, for instance. However, bankruptcy in any open society where individuals take personal responsibility for their acts and decisions just means the internalization of costs (which flags prudency). State intervention, in these cases, is a way to prevent individuals from taking responsibility for their actions and decisions, externalizing costs and increasing the risk of what is called ‘moral hazard’: incentives to augment risk exposition foreseeing that costs will be at least partially externalized. This points to the fact that some forms of state intervention to avoid (or alleviate) socio-economic problems are contrary to open societies’ values.

4. Open Societies and Free Markets

In the previous section, I have argued that the alleged association between open societies and democracy is, at best, weaker than thought. Democracy and open societies seem to be tense, and strong commitments with one might require dropping commitments to the other. Even more, I have argued that democracies (as conceived by most democracy’s defenders, Popperians among them) have incentives that are contrary to the open society. These incentives increase voters’ ignorance and irrationality, undermining not only personal responsibility but also casting doubts on the prospects of a rational design of political institutions. Democracies are better than most political regimes we have known thus far, but they fall short of providing the right political structure for open societies to completely flourish. Nor are they responsible for their flourishment.

But there are open societies, at least in comparative terms. And it is also true that most open societies are democracies, although it is not true that all democracies are open societies. I think that this association is contingent and that there is another element that has fostered open societies in the last three hundred years. Democracy has then only been a secondary factor. Put differently, open societies are what they are not because they are democratic; they are open and democratic because they have left most individuals’ decisions to take place in systems where social cooperation is implemented through the institution of private property, free markets, competence, and the division of labor. Then, by way of conclusion, I want to suggest that these institutions (which I gather by the name “free-market capitalism”; societies that live under free-market capitalism are to be called “free-market societies”) are what have created, fostered, and strengthened open societies. The more relevant decisions are left to free markets, the fewer problems democracies will have. So, a defense of the open society is not per se a defense of Western democracies, but a defense of free-market capitalism.

Three clarifications are in order. First, I do not claim that free-market societies can solve the problems of democracy I have stressed in previous sections. These are problems of democracy and disappear as soon as democracy is increasingly replaced by free markets. In more precise terms, democracy is a political instrument to coordinate the decisions of many people, in which the rule of majority (among other constraints) rules. Previous sections aimed to show that that instrument suffers from some problems, for instance, voter ignorance: coordination in democracy is achieved through voting, where individuals vote, directly or indirectly, on issues they do not care about, where information is too costly and the chances to make a difference are too low. This contrasts with free markets, where decisions are coordinated differently. So, what is changed is the instrument by which individuals can coordinate their decisions and act. Free-market societies do not solve the problem of, say, voter ignorance. However, the problem disappears because coordination is now achieved through market mechanisms.

Second, free-market societies coordinate the actions of individuals on a large scale by means other than political means (e.g., democratic means). In this sense, free-market societies do not improve democracy but replace it (at least partially). However, free markets do not impede per se the emergence of forms of government that are worse than well-functioning democracies. For instance, some have argued that populism is a form of degraded democracy that undermines it. Populism (see Mueller 2016) in general relies on an abstract idea of “The People” to go against the rule of law, and individual rights, and employ a specific rhetoric to exalt genuine representation (in terms of anti-elitism, anti-pluralism, they-us logic, etc.). It can be argued that by weakening democracy, free-market societies pave the way to degraded forms of democracy, such as populism (some could mention Donald Trump in the United States or Javier Milei in Argentina as examples of this phenomenon, although I disagree). My view is that this is not a free-market societies’ problem, but a problem of political representation under democratic institutions. Populism is transversal to different societies, some of them far away from free-market capitalism (such as Venezuela). It is an interesting question why both relatively open and closed societies can give rise to populism, but I believe that this is because some shared political structure can lead to it. It is true that free-market societies do nothing to improve democratic institutions, but they do nothing to promote populism either. It is expected that if free-market mechanisms take over coordination problems, democracy (and populism as well) will lose ground.

Third, it can be argued that I hold a realistic view of democracy (democracy as actually works) and an idealist view of free-market societies (as they should ideally work). This is a complex discussion I cannot dig into it here but let me offer a more realistic view of free-market societies. I do not claim that free markets work perfectly or do not suffer from problems (as I have said previously learning and discovery are part of free markets as processes, which implies the existence of imperfections and dysfunctionalities). What I aim to show is the incentive structure that free-market societies possess and how individuals behave within such an incentive structure. My view of free markets and agents within free markets is closer to the Austrian School of Economics, in which individuals always act in situations of imperfect knowledge, in which individuals learn and discover new strategies and react to changes in unpredictable ways, in which individuals do not always assess their possibilities successfully, etc. In this sense, my view of free-market societies is quite realistic since it does not presuppose that individuals should exhibit extraordinary features. After all, markets are products of actual human action.12

After these clarifications, I will round off my argument that open societies are better realized in free-market societies than in democracies. In the previous sections, I have shown that democracies suffer from in-built incentives that indulge ignorance and irrationality in the electorate. These hinder rational decisions. This is because democratic societies allow individuals to externalize the costs of their ignorance and irrationality through majoritarian rule. This hinders personal responsibility. What is the solution, then? To change the incentives that indulge ignorance and irrationality. In other words, promoting incentives and mechanisms that internalize costs and benefits, which lead to greater levels of information gathering and rationality. This then means to move from democratic processes to coordinate and make decisions to market processes. As I have suggested before, when individuals face the costs (and the benefits) of their decisions, they have more incentives to obtain more information and think more rationally. The internalization of costs and benefits then promotes personal responsibility and rational decisions. The par excellence place where this happens regularly is in processes of social cooperation where the institutions I have mentioned before are present. So, a relatively open society is not one in which citizens cast votes in a ballot box more frequently, but one in which citizens can make more decisions in free markets, in which private property is respected, and competence is encouraged. In this case, weakening democracy and strengthening markets means promoting open societies.

The underlying argument is very simple, but it is worth making it clear. Open societies aim to promote personal responsibility and rational decisions. This can translate into democratic societies or free-market societies. In democratic societies, individuals have incentives to externalize the costs of their decisions, indulging ignorance, and irrationality. This is contrary to open societies’ values. In free-market societies, individuals always have incentives to acquire more knowledge and think more rationally about their choices since most of the costs (and benefits) of their decisions are internalized. This is in line with open societies’ values. Therefore, free-market societies realize better (in the sense of laying out the right sort of incentives) open society’s values than democratic societies.

Another aspect in which free-market societies outperform democratic societies is in having the means to alleviate the socio-economic problems Popper is most concerned about. As a matter of fact, systems that encourage social cooperation on the basis of free contract, free trade, the division of labor and the respect of private property have promoted social and economic development as never seen in human history. Poverty and misery have been the rule since the beginnings of society, until very recently when free-market societies (or relatively free-market societies) started spreading and flourishing. Take for instance life expectancy between 1800 and 2000. Thanks to the increasing access to medicine, better food, and clean water (which are driven by economic growth) life expectancy at birth grew from 30 years to 67 years (Riley 2001). Or take good-service production in relation to population. Between 1800 and 2020, the population multiplied by a factor of 6, but the amount of goods and services produced and consumed by the average person multiplied by a factor of about 8.5, despite the never-seen increase in population (see Maddison 2001). While starvation and famines were the rule before, they have now almost disappeared. More conceptually, free-market societies can better deal with the practical problem of channeling dispersed information and knowledge that can never be given to a selected group of minds. It is the price system, in well-functioning markets, that can more easily solve problems of economic calculus, allocate resources where they are more needed, and destroy resources when they are less needed. As Hayek famously argued, it is the price system that can coordinate dispersed information and actions of different individuals at different places, serving in a more effective and rapid way the welfare of the whole society.

This leaves clear that it was not democracies responsible for the prosperity of the last centuries; it was not piecemeal social engineering nor any form of state intervention, but it was the private initiative in free markets that took people out of poverty. The same goes for unemployment. The diversification and specialization of activities increased and improved human capital. Individuals get employed because their skills and labor are needed somewhere, not because public policies made it possible. Therefore, it is not empirically true that open societies have brought about increased levels of misery and poverty, but quite the opposite: they radically reduce poverty and misery. Socio-economic problems still exist, but the solution is to foster open societies. This means fostering social cooperation based on private property, free markets, and the division of labor. That is, free-market societies.

Let me finish with replies to two common counterarguments. One counterargument is that free-market societies need democratic institutions to work as intended. The second one is that free markets typically suffer from systematic failures that lead to misallocation, inefficiency, or might hinder justice. Therefore, states (in particular, democratic states) are called for. With respect to the first argument, it can be said that it is true that free markets need stable rules to work as intended, but it does not necessarily mean that they are realized by democratic institutions. Not even by governments. As Peter Leeson (2014) and Edward P. Stringham (2015) have shown, private governance can offer an adequate framework for private enterprise and large-scale individual coordination to take place and flourish. If they are right, then democratic institutions are not necessary for free-market societies. They may be, at best, the best friendly political framework that has been found so far. But, as I said previously, it is a contingent fact that relatively free-market societies also have democratic governments.

As for the second argument, it must be noted that free-market societies can enact collective decisions, provided that they are voluntary. Different forms of voluntary cooperation can emerge which entail some form of collective decisions. The bone of contention is not really about collective decisions, but about whether they are voluntary or not. The call for government intervention in the face of so-called market failures (or collective action problems) is a call for non-voluntary collective decisions that, allegedly, are able to fix the failure. This is a broad topic that has given place to many controversies, but I will try to be brief. The idea of market failures is usually associated with (unrealistic) expectations of ideal competition or ideal functioning markets (see Hayek 1989, p. 4). Any deviation from the ideal is a failure that can and must be corrected. Yet, such an ideal is unattainable and unidentifiable (Pennington 2017), based on an unrealistic world of perfect competition. Failures must then be relativized. On the one hand, markets do fail, but it is not clear a priori and externally what the failure was (that is why Hayek says that markets are processes of discovering and learning). On the other hand, if it is not clear what the failure was, then any external intervention (i.e., government intervention) is bound to also suffer from ‘government failures’ (as Milton Friedmann has said). There seems to be an unfair imbalance here: While market failures are recognized, government failures are almost ignored. In any case, the burden of proof is on the pro-government-intervention side.

5. Conclusions

We owe to Popper one of the best defenses of Western values and open societies. Popper and many of his followers thought that the defense of the open society was ipso facto a defense of Western democracy. When we look at democracy ideally, this can be so. But real democracies, with real individuals with real incentives, fail to live up to the open society’s standards. It is a long-lived mistake in political philosophy to defend political institutions in the light of their performance under ideal conditions. With the same argument, even monarchies could be defended: it is possible to have a noble, well-intentioned, and wise king who promotes welfare, justice, and freedom, but in reality, the probability is too low, and history shows us that this has rarely been the case. So, a realistic, instrumental defense of democracy must take seriously the real incentives that individuals have in order to obtain information and to think rationally about political issues. The last seventy years of empirical research in this area, plus the contributions of microeconomics, have shown us that the performance of democracy is rather poor. It is not democracy that has fostered open societies because it undermines many of the open society’s values.

But if it is not democracy, what makes open societies open and so much better? I have suggested that it is social cooperation under the institutions of private property, free markets, the division of labor, and competence that has strongly driven societies towards openness. In this way, Popper’s defense of the open society is a defense of free-market societies, where individuals’ decisions are mostly made in markets, not in the ballot box. The problems of open societies are not to be solved with more democracy but with less. With more markets, not with less.