Chapter One:

Introduction

Does democracy promote development? Should we view regime-change as a development policy? Or should we view it as “development-neutral,” or even as a negative factor in development? What is the effect of regime type on economic growth, infrastructure, human capital, social equality, and the overall quality of life?[1]

Few questions in the post-Cold War era have engaged politicians and policymakers with such intensity. While democracy has always been an objective of U.S. foreign policy, perhaps at no time in our history has it been cast in such a central role. At the same time, democracy has come to be incorporated into the mission of the World Bank, the United Nations, and other development agencies. All presume that the causal effects of democracy on development are, on balance, positive.

Yet the scholarly evidence for this proposition is thin. Political scientists, economists, and sociologists generally view the effect of regime type on development as inconsistent, at best. It is not clear whether democracies produce better policies and policy results than autocracies. Some argue, with an eye towards the East Asian NICs, that economic growth is most likely to be achieved through a period of strict authoritarian rule, deemed necessary to instill discipline in the labor force, to prioritize long-term savings and investment over current consumption, and to resist the rent-seeking pressures of organized groups.[2] Democracy is often associated with policy sclerosis[3] and with skewed political representation, a situation in which relatively educated and well-organized voters (e.g., public sector unions and urban elites) are able to monopolize state resources and prevent measures to redistribute resources to the rural poor.[4] Democracy may also encourage a clientelist, rent-seeking style of politicking in which side-payments to special interests trump the provision of collective goods.[5] Democracy may even open the floodgates to ethnic conflict and social disorder.[6] In short, there are many reasons—and a considerable amount of anecdotal evidence—to suggest that democracy does not stimulate positive development outcomes.[7]

Evidence for these judgments is drawn from case studies and, increasingly, from cross-national statistical studies. The latter typically focus on the relationship between regime type and economic performance. Here, the median finding is a null result. The net effect of regime type appears to be neutral or inconsistent. Scholars are somewhat more optimistic about the causal effects of democracy on social development; yet, even here there is great ambiguity. Democracies do spend more on social policies, but it is not clear that there is a robust association between spending levels and policy achievements in the developing world (e.g., longer life expectancy, higher literacy, and so forth). The social policy bucket is very leaky.[8]

These arguments are reviewed at greater length in subsequent chapters of the book. At present, it is sufficient to observe that the scholarly view of these matters is ambivalent. While democracies may prevent certain domestic policy disasters, such as widespread famine,[9] their positive accomplishments appear to be quite thin. Democracy has not proven to be associated with development, as that term is commonly understood.

Of course, democracy might still be defended for other reasons, e.g., because it enhances citizen participation, civil liberties, and civil society. While these are important outcomes, it should be pointed out that this sort of argument is unlikely to inspire great enthusiasm among those currently forced to live on less than one dollar a day—roughly one fifth of the world’s population. In summarizing the results of a recent poll in eighteen Latin American countries the authors note that “the preference of citizens for democracy is relatively low; many Latin Americans value development above democracy and would even stop supporting a democratic government if it proved incapable to resolving their economic problems.”[10] Indeed, the demand for “food first” sounds more sensible than the call for “democracy first.” Democracy is emphatically not equivalent to justice; it is, at best, a component of justice.[11] Thus, it would be quite wrong for first-world actors to presume that a democratic organization of politics is preferable for countries in the developing world if another regime-type promises greater material reward.[12]

Is democracy a luxury to be enjoyed only by countries rich enough to afford it?

We think not. Our reasoning hinges on a matter of conceptualization and measurement, as well as a larger theory of regime dynamics. We begin by sketching the overall argument and proceed to a discussion of possible causal mechanisms

The Argument

The academic literature provides scant evidence for the proposition that democracy fosters development, as we have observed. Yet, it is notable that these empirical probes examine focus largely on the immediate or short-term causal effects of regime type. Whether a country is democratic or authoritarian today is expected to influence its developmental trajectory in the next year or decade (as defined by a given study’s research design).[13] The theoretical issue is thus framed along a single dimension. Countries are classified according to their regime-status in the present or recent past.

We believe that this expectation is implausible. A change in regime is unlikely to produce marked changes in the quality of governance in the immediate term. A fortiori, it is unlikely to produce marked changes in the quality of policy outcomes in the years immediately following a regime change. Regime changes are often periods of extreme instability and unpredictability. It would be surprising, therefore, if the performance of countries moving from authoritarian to democratic rule were substantially improved during a period of transition. It is no surprise, then, that empirical tests often fail to reject the hypothesis that regime type is uncorrelated with a particular development outcome.

Our contention is that the effects of political institutions are likely to unfold over time—sometimes a great deal of time—and that these temporal effects are cumulative. Regimes do not begin again, de novo, with each calendar year. Where one is today depends critically upon where one has been before. Countries build their political institutions over long periods of time. Historical work suggests that democracy and authoritarianism construct deep legacies, extending back several decades, perhaps even centuries.[14] It follows that we should concern ourselves with the accumulated effect of these historical legacies, not merely their contemporary status.

Thus, we introduce a second dimension – time – to our judgment of regimes. This means that, for purposes of discussion, there are four theoretically relevant ideal-types, not two: A) old authoritarian regimes, B) young authoritarian regimes, C) young democratic regimes (newly democratized countries), and D) old democratic regimes. There are also a corresponding set of transition possibilities among these four regimes, as illustrated in Figure 1.1. Our supposition is that the quality of governance increases from A to B, from B to C, and from C to D. The worst-governed polity is an old autocracy, the best-governed polity an old democracy.


Figure 1.1:

Regime-types in Time

A=Old autocracy; B=Young autocracy; C=Young democracy; D=Old democracy. Arrows represent possible regime transitions.
However, in our view, both dimensions of this phenomenon are rightly conceptualized as matters of degree. The concepts “Old democracy,” “Young democracy,” et al. are therefore regarded as heuristic tools rather than discrete political entities. Indeed, there are an infinite number of regimes that fall in between these sharp categories and, equally important, there are no theoretical end-points. Hence, Figure 1.1 is constructed as a line graph, rather than a 2x2 matrix.

In short, democracy is a good thing and its goodness is cumulative. The more a country has of it the better off it will be, all other things being equal. The less a country has of it the worse off it will be, ceteris paribus. These are matters of degree, hinging upon the twin components of the underlying concept.

It follows that we wish to consider democracy as an accumulated stock concept rather than as a level concept, as it has traditionally been understood. We conceive of this stock as the accumulation of democratic experience over time, so it is comprised of both dimensions displayed in Figure 1.1: regime-type (a country’s degree of democracy-authoritarianism at a given point in time) and regime history (how long it has had those characteristics). The expectation is that the greater a country’s stock of democracy, the greater its “flow” of good governance.

Causal Mechanisms

In what ways might a country’s democratic stock contribute to its developmental potential? Why might old democracies experience improved economic development, political development, and human development relative to autocracies and young democracies?

Identifying causal mechanisms is a daunting task. The theoretical variable of interest – “democracy” – encompasses a wide range of features, and the policy outcomes associated with “development” are even broader in scope, and perhaps more difficult to reach agreement on. As a result, the causal story we have to tell is necessarily quite complex. Democracy affects development through multiple channels.

In this initial discussion we focus on seven causal mechanisms that we suspect have important implications across a variety of policy areas: 1) transparency, 2) civil society, 3) accountability, 4) learning, 5) equality, 6) consensus, and 7) institutionalization. (Later chapters will concentrate on causal pathways specific to particular policy areas.)

These are not new concepts. Indeed, they are common elements in the usual contrast among regime-types. Democracies are often said to enjoy advantages in transparency, civil society, accountability, learning, equality, consensus, and institutionalization. Our contribution is to point out that these traditional arguments are much more persuasive if applied to established (consolidated) democracies, for the development of these distinctively democratic virtues takes time, sometimes a great deal of time.

Transparency

Democracies, almost by definition, operate with a greater degree of openness than autocracies. Leaders and issues must be paraded before the electorate, institutions enjoy a degree of autonomy from each other (and are often in direct competition with each other), and the most important information-dispensing institution, the press, is comparatively free from political control (freer, that is, than it would be under authoritarian rule).[15] In a democracy, politics is everybody’s business. This means that the mechanisms by which political transparency might be achieved are generally much more prevalent in a democracy than in an autocracy.

The basic point is not contested. What deserves emphasis, from our perspective, is that the mechanisms by which transparency becomes realized are not achieved simply by the installation of multi-party elections and constitutionally guarantees of freedom of the press. Indeed, the very goal of transparency may take time to establish as a valued good. This is especially true for countries emerging from a long period of authoritarian rule, where post-transition elected governments are likely to champion strong government, disparage the goal of openness, and be intolerant of critical views in the public sphere. In any case, it takes time for institutions that produce transparency – principally, government agencies, legislative committees, the media, the judiciary, and NGOs -- to become skilled investigators of public malfeasance and monitors of the public weal. They must first carve out their respective roles, develop constituencies, defend their independence, gain knowledge of complex issues, create knowledge networks across institutions, and develop a reputation for perspicacity and probity such that their word on a subject carries weight.

Civil Society

The creation of a strong civil society – a sphere independent from politics and from the market – is virtually impossible under circumstances of authoritarian rule. Indeed, neither the strong authoritarian state nor the weak authoritarian state can effectively guarantee the flourishing of voluntary associations. In the first instance, the institutions of the state or ruling party tend to pervade all public deliberations. In the second instance, there is no space safe for public deliberations because there is no public order. Thus, it may be argued that civil society is premised on the existence of a democratic regime.[16]

However, voluntary associations representing diverse purposes, interests, and ideals do not spring up immediately following the declaration of multi-party elections. In some cases, there is scarcely a semblance of civil society prior to the instauration of multi-party competition. In other cases, where such organizations are already in place, it takes some time for them to gain traction, i.e., to reach out to new constituencies and to mobilize resources. Even in well-educated and comparatively developed societies such as the former Soviet bloc the development of civil society has lagged behind the democratization process.[17]

Yet, over time, the density and diversity of voluntary associations seems to grow under conditions of democratic rule. Despite the decline of party organization in some longstanding democracies, the general pattern of civil society is one of persistence and increasing diversity. In developing societies, the trend is even more marked.[18] Thus, it seems fair to regard civic associations as a hallmark of established democracies.

Accountability

Where conditions of democracy hold, we expect that mechanisms of accountability will, over time, become more established.[19] In this principal-agent relationship, the principal is the electorate, composed of all citizens (we leave aside the precise definition of the citizenry), and the agent is the class of elected officials (acting through appointed members of the bureaucracy). The class of issues to which such mechanisms apply constitute the domain of the principal-agent relationship. In order for mechanisms of accountability to exist the principal must have some way to monitor the activity of the agent; otherwise, there is no connection between what the agent does, and potential punishment or reward that the principal might bestow.

In order for mechanisms of accountability to be fully operative across a nation several conditions must hold: a) the agent must be identifiable and relatively coherent (otherwise, there is no person or persons to reward or punish); b) the principal must have monitoring capacity; c) the members of the class of principals must be large (a mass electorate rather than a narrow selectorate), and, most importantly; d) mechanisms by which the principal can regularly reward or punish the principal must be available.

It follows that mechanisms of accountability are likely to be weak and/or limited in purview in an authoritarian regime. To be sure, there is likely to be an identifiable agent. However, this agent will be difficult to monitor because authoritarian regimes tend not to operate in an open, transparent fashion. Equally important, the class of persons with an accountability relationship to the agent is generally very small, extending perhaps to the military, an aristocracy, a landowning class, or the agent’s ethnic grouping. The “selectorate” is much smaller than the potential electorate. Of course, the citizens at the base can exact revenge upon an irresponsible leader by overthrowing him or her. But this is a crude, irregular, and generally ineffective mechanism of accountability, and operates (to the extent that it operates at all) only in the most extreme cases of malfeasance. This means that the class of issues to which mechanisms of accountability can be said to apply in a dictatorship is limited to those of concern to the selectorate and/or those which might anger the broader citizenry to the point of open revolt. In sum, accountability is not a concept with much use in a typical authoritarian regime.