Democracy and development

Is democracy inherently a good thing? And do democratic institutions

facilitate economic development? Democracy is a good thing because it
facilitates free human choice and it furthers the good of political participation.
But the answer to the latter question is an empirical one, and there is
debate within the development field about the effects of electoral democracy
on the development process. Some argue, for example, that the experiences of
Korea, Taiwan, or Indonesia show that a strong authoritarian state is better
able to engineer a successful process of economic development than an
electoral democracy such as India (because of its ability to discipline
fractious demand groups). We will argue, first, that democracy is inherently
desirable; second, that the empirical record of authoritarian developing states is
about as mixed as that of democratic states; and finally, that only democratic
institutions give any promise of tilting economic development policies toward
the interests of the poor.

Normative theory of democracy

Democracy is a good thing, both intrinsically and instrumentally.
Intrinsically, it is a necessary component of the ability of individuals to live
freely and autonomously. Instrumentally, it is an institutional guarantee that the
policies and laws created by a government will have a reasonable fit with the
fundamental interests of the people. Thus democracy is a central determinant of
the quality of life, and a central element in the ability of men and women to live
freely and autonomously as human beings. This is no less so in poor and
developing countries than it is in the developed countries.
So, at least, the moral intuitions of a liberal western philosopher would assert.
we need to look more closely at the meaning of democracy and democratic
citizenship.
And we need to consider several important empirical questions: do democratic
institutions facilitate economic development of the right kind? And do
democratic institutions guarantee, or even make probable, the result that
government policy and law will reflect the fundamental interests of the people?

Economic development “of the right kind”

We have seen that economic development “of the right kind” involves
several dimensions:

• growth in the productive capacity of society: growth in productivity of

labor, agriculture, and capital (leading to growth in per capita incomes and
per capita assets)
 2

• development that leads to significant and continuing improvement in the

quality of life for the poor and the near-poor (that is, the majority of the
population in most developing societies)
• development that serves to broaden the distribution of economic assets and
incomes
• development that leads to improvement in conditions of health and safety in
the workplace
• development that leads to improvement in “quality of life” issues for all:
improved access to health care, clean water, education
• development that leads to sustainable environmental change and resource
use
• development that leads to improvement in gender equity over time

Do democratic political institutions have positive effects on the

achievement of some or all of these characteristics of economic development?
Does a transition to stable electoral democracy in a developing society help to
facilitate economic development “of the right kind” in that society?
What is democracy?
We can represent the central characteristics of a democracy from two
points of view: from that of the individual citizen, and from that of the political
institutions through which the values of democracy are realized in a particular
social context. Let us begin, then, at the level of the citizen. There are several
central and defining normative commitments that jointly define the political
theory of a democracy. In the briefest possible way, we can offer a preliminary
definition of democracy in these terms: A democracy is a polity in which
collective decisions (laws, policies, procedures) are the expression, direct or
indirect, of the preferences and choices of the collection of equal citizens of the
polity.
Democracy thus pertains to the self-rule of a politically constituted
social group—a state or provincial authority, or a city or town. Several elements
distinguish a political group from other forms of association: the fact that the
political unit is empowered to coerce its members through the collection of
taxes, restrictions on the use of property, and the imposition of regulations and
laws; and second, that the authority of the unit does not depend upon the
continuing voluntary consent of the individual for the exercise of its authority.
The citizen may sometimes vote with his or her feet (by departing the
jurisdiction); but while resident within the jurisdiction of the political unit, the
citizen can be compelled to act according to the laws, policies, and decrees of
the political authorities. And laws have the invariable characteristic of
restricting freedom; that is, they inevitably work to prevent people from acting
on choices they otherwise would have made.
 3

It is sometimes debated whether there is ever a moral justification

for coercive legislation by the state. 1
I will take it, with Hobbes, Rousseau, Mill, and Rawls,
that the individuals within a society require some central authority in order to
establish a system of law, to prevent violence, and to enact policies in the
common good. Society requires a state. And democratic theory attempts
to provide the most general blueprint possible for the legitimate state.

Tenets of normative democratic theory

The central tenets of normative democratic theory are these:

• All adult members of the collectivity ought to have the status of citizens
(that is, there ought to be no restriction in political rights for different
groups of people within the polity; universal citizenship principle).
• All citizens ought to have the broadest set of political rights and liberties
possible, compatible with the extension of equal rights to all (that is, there
ought to be full equality and the broadest possible liberty for all citizens; the
liberty principle and the equality principle).
• Legislation ought to reflect the principle of the sovereignty of the people.
When and where legislation is required, it ought to result from a process
which involves the meaningful expression of interest and preference by all
citizens (popular sovereignty principle).
• The legislative process ought to weight no individual’s or group’s
preferences more heavily than those of any other individual or group (equal
weight principle).
• Finally, a democratic society is one that is fully subject to the rule of law:
legislation rather than personal authority produces limitations on individual
liberty, and legislation is neutral across persons (legality principle).

What is a citizen? A citizen is a person. A person is a moral

individual, possessing a plan of life, a conception of the good for him- or
herself, a set of needs, a set of rights and liberties, and, finally, a set of
preferences that derive from needs and the conception of the good. The
individual’s preferences represent the embodiment of his or her wishes with
respect to a given set of outcomes or choices.

Democratic institutions
These represent the chief desiderata of a democratic polity. But these
principles do not dictate a specific implementation. Rather, it is necessary for a

1
 4

given polity to design a set of political institutions through which the principles
of liberty, equality, and sovereignty are realized. There is a logical gap between
the principles and the institutional implementation, in the sense that people can
always debate whether the particulars of local institutions adequately realize the
relevant underlying values. It is likely, moreover, that different institutional
arrangements represent different ways of accommodating the underlying values,
and represent different types of tradeoffs among them.
In discussing institutional design it is useful to recall the discussion in
several different ways of aggregating and comparing multiple goods. A
democratic state is a complex system involving multiple features (electoral
institutions, parties, constitutional protections) and producing multiple goods
(individual liberty, effective legislation, secure property rights, popular
sovereignty). Institutions can be designed de novo, or they can be adjusted
through a series of corrections and reforms. And as we consider the process of
adjustment of an institution, it is necessary to consider carefully the “objective
function” by which we intend to guide the adjustment and reform process. Are
we willing to make tradeoffs among the goods produced by the institution—e.g.
give up some popular sovereignty in order to achieve more equality of assets?
Or do we mean to accept only Pareto-improving innovations—that is, those that
improve at least one good without reducing any other good?

An ideal type of democracy

Consider this institutional sketch of a democratic system. The polity
adopts a constitution that defines maximal political rights and liberties, and
defines the status of citizenship. The constitution prohibits the establishment of
laws that limit or constrain the constitutional rights and liberties of citizens, or
that create inequalities in basic rights among different groups of citizens. The
constitution further creates a legislative process through which elected
representatives engage in a majoritarian process of debate and legislation.
Representatives are elected and can be removed by the electorate; and the
legislative process is itself governed by majoritarian voting rules. Legislation
cannot contravene the constitution, and a separate super-majoritarian process for
revision of the constitution is established. This sketch embodies each of the
values indicated above: universal citizenship, maximum liberties, and popular
sovereignty. The sketch corresponds fairly closely to the political theory of the
United States government. Note that this sketch privileges liberty and equality
(by placing the constitution prior to the legislative process). As a result, it
restricts popular sovereignty. Even if a majority preferred legislation that
restricted liberties (for all or for a group), such legislation would be
unconstitutional.
We can imagine other institutional sketches as well. We might imagine
building a polity on the popular sovereignty principle first: all legislation
emerges on the basis of majority vote of all citizens, and all legislation is in
 5

principle possible. Such an approach would privilege popular sovereignty, but

would potentially interfere with the liberty principle or the equality principle
(since it is possible that a majority would prefer to reduce liberties or undermine
equality.
It is possible to provide a taxonomy of possible democratic systems
(figure 1). The variables generating the taxonomy are “constitution/no
constitution”, “representative/direct”, and “divided/unified”. This produces six
variants (since direct government is by definition unified government). For any
species of democratic government, we can always ask the fundamental question,
how well do these institutions work to establish and implement the values of
universal citizenship, maximum liberty, full equality, and popular sovereignty?2
In addition to describing the institutions of constitution, elections,
legislation, and executive action, a political system also has a surrounding
cluster of supporting institutions: mass media, political parties, political fund-
raising, and legislation surrounding the electoral process. Once again, for any
particular configuration of institutions of these sorts, we can ask the question:
how well do these institutions establish and implement the central values of
democracy?
Finally, we need to find a place within our theory for the “instruments
of coercion” within a society—the military and police, and the organs of private
violence. A modern state—whether developed or developing—marshals
capacity for a significant level of coercion. It is possible for political authorities
to make use of this capacity for their own political purposes; likewise, it is
possible for military and police authorities to use coercion and the threat of
coercion to political purpose.
This brief discussion serves to establish the abstract geometry of a
democratic polity: constitutional definition of the status of citizens,
constitutional establishment of basic rights and liberties, establishment of an
electoral process through which representatives are appointed, establishment of
a majoritarian legislative process through which legislation is brought into
being, establishment of an executive power which has the authority and charge
to implement and enforce legislation, and establishment of a judicial branch
charged to interpret the law and to judge law-breakers.

2
 6

polity

constitution no constitution

representative direct representative direct

divided unity divided unity

govt govt govt govt
Figure 1. Categories of democratic government

Institutional variants of democratic regimes

The ideal type of authoritarian government

What is the alternative to democratic government? Authoritarian
government is just as complex as democratic government, in that there are many
different ways of institutionally implementing a system in which the few govern
the many. But let us lay out an “ideal type” of authoritarian government that is
common in the developing world. I will focus on what is sometimes called
“bureaucratic authoritarianism;” important variants include military dictatorship,
party dictatorship, or “strong-man” dictatorship, for an account of bureaucratic
authoritarianism in Latin America.) In the bureaucratic authoritarian state, a
strong man rules the state, making use of a complex bureaucratic
organization to create legislation and policy and an extensive coercive
apparatus (army, police) to enforce government policy. But since both
bureaucracies and police organizations are complex social organizations,
autocrats have less than absolute power. They confront classic “principal-agent”
problems in inducing the various organizations to do their bidding. So there is
some looseness in the lineages of power from the center to the administrative
peripheries of the polity. Second, most societies contain non-political centers of
power with which the autocrat must contend—land owners, businessmen,
 7

financiers. Finally, even the most autocratic regime must give some weight to
the preferences of the masses of the population. Coercion has its limits, and the
autocrat must remain aware of the potential of popular unrest in response to
unpopular policies (increases in staple prices, increases in taxes, reduction in
customary rights).
What are some of the common characteristics of authoritarian regimes?

• Frequent use of force and threat of force against the population

• predatory treatment of the national economy—taxation, access to positions
of wealth, rent-seeking
• bureaucratic interference with the market (especially in financial markets)
• tendency towards capital-intensive growth
• low ability to moderate and negotiate ethnic or nationalist conflicts

Transition to democracy
It is common in recent history to find developing societies in a state of
transition from authoritarian regimes to democratic regimes. Military
dictatorships, bureaucratic oligarchies, and other authoritarian regimes have
found themselves subject to irresistible forces which compel them in the
direction of a degree of progressive democratic reform: extension of political
rights to citizens, establishment of limited electoral processes, extension of the
ability of independent parties to organize themselves, extension of some degree
of freedom of press, and so forth. Here a series of questions demand answer.
First, to what extent is it possible for skillful elites and rulers to orchestrate the
process of democratic liberalization in such a way as to preserve their power and
privilege within the resulting regime? Second, what are the features of
institutions which best serve to bring about effective democratization? Third, is
there a relatively clear distinction between effective democracies and sham
democracies? Finally, what if anything can we say about the progressive
features of hybrid political systems—polities that are intermediate between
authoritarianism and democracy? Are the steps along the road to democracy
unambiguously positive with regard to individual freedom and other democratic
virtues?

Philosophical issues concerning democracy

It is worth noting that there are well-known paradoxes underlying the
theory of democracy. The Arrow paradox establishes that there is no logically
consistent and fully general voting system that maps individual preference
orderings onto a single consistent social preference ordering.
There are also difficult philosophical issues that arise in the endeavor of
explicating the concept of preference. Are preferences entirely arbitrary and
subjective? Or is there a principled relationship between one’s fundamental
values, plan of life, conception of the good, and one’s preferences (or a subset of
 8

them)? Is there a principled basis on which others may criticize one’s scheme of
preferences? And, finally, is there an objective basis for saying that some of a
person’s preferences are more important than others—or that one person’s
preferences are more important than another’s? These problems are critical for
democratic theory, because collections of individual preferences underlie the
principle of popular sovereignty. If I prefer one zoning code over another
because I prefer to have silence in the neighborhood while doing my early
morning exercises, whereas you prefer the second option to the first because it
alone will allow you to earn your living—should your preferences be given
more weight than mine?
Empirical issues
To this point we have focused largely on the normative theory of
democracy. However, it is crucial to recognize that democratic institutions are
institutions—they have real empirical and causal properties, and function
according within the context of forces that give them a real empirical trajectory
that may be at odds with the ideal theory. So at this point in the story it is
appropriate to turn to a realist theory of democratic institutions, and to ask sharp
empirical questions about the actual characteristics and tendencies of democratic
political institutions.
Let us turn now to some of the empirical questions that surround the
issue of democracy within the context of developing societies. How do the
typical institutions of electoral democracy affect the process, character, and rate
of economic development? Do the institutions of electoral democracy have the
effect of inducing more egalitarian economic development? Do such institutions
serve to emphasize the interests of the poor? Can broader political participation
improve the situation of the poor?

Multi-case studies of democracy and development

There has been an extended debate about democracy and development,
and the relations between democratization and economic growth. Do the
institutions of electoral democracy facilitate or impede development? Samuel
Huntington characterizes the debate in terms of “conflict” and “compatibility”
theorists (Huntington and Harvard University. Center for International Affairs.
1968). Some have maintained that democratic regimes are in general less
capable of managing effecting economic development than authoritarian
regimes. The central premise of this reasoning stems from the observation that
development requires change, and that change affects some voters adversely. So
governments dependent on electoral support in the next election will typically
tend to avoid choices that impose hardship on significant numbers of voters.
(Adam Przeworski’s arguments in Democracy and the Market (Przeworski
1991) represent a thoughtful argument to this effect.) Others have argued that
democratic regimes are positively associated with economic development, and
especially with more egalitarian modes of development. Finally, there is a body
 9

of thought which holds that democracy is neither positive nor negative with
respect to economic development. ((Sirowy and Inkeles 1990) provides a
careful review of this issue and the empirical data that pertains to assessment of
the various hypotheses.)
The issue of the dynamic causal relations between democratic political
institutions and the pace and character of economic development can be probed
in several different ways. First, we can approach the problem theoretically or
deductively: given what we know about the character and institutional dynamics
of democratic institutions, and given what we know about the character and
needs of economic development, what causal connections does underlying
theory lead us to expect? Second, we can approach the problem through multi-
case studies in which we operationalize the concepts of democracy and rate and
character of development, and then examine to see whether there are meaningful
statistical associations among the resulting variables. Both approaches have
been pursued in the literature of the political economy of development, with
deeply mixed results.

The theoretical case

Note that political theory leads us to expect causal connections flowing
in both directions. The institutional arrangements of electoral democracies, with
the dynamics created for majoritarian governments by the political calculus of
voting blocs, can be predicted to give rise to the likelihood that some
development choices will be more difficult than others. That is, the institutions
of democracy are likely to impose a characteristic “look” on the process of
economic development. But likewise, features of the economic development
experience, short-run and long-run, may have significant effects on the stability
and character of political institutions. For example, the structural adjustment
crises of Latin America in the 1980s posed serious challenges to the stability of
democratic institutions in a variety of countries. ((Peeler 1998) describes the
experience of Mexico, Venezuela, Peru, and Bolivia from the point of view of
this direction of the causal arrow. Peeler takes the view that there is a generally
positive causal relationship flowing from the presence of democratic institutions
to effective economic development.)
The central theoretical dynamics that come into play include at least
these:

Conflict theory
• Development requires decisive policy choice and effective policy
implementation; authoritarian regimes are more decisive and more effective
in implementing policy.
• Ethnic and sub-national conflicts interfere with economic development, and
are most effectively suppressed by strong authoritarian government.
 10

• Authoritarian governments are more able to effectively defer consumption

in favor of savings. Democratic regimes are under a political imperative to
increase social welfare spending, which reduces the rate of accumulation.
• Democracy undermines investment (Huntington and Dominguez 1975).

Compatibility theory
• Progressive development requires policy choices that lead to a development
pathway that produces a wide distribution of the benefits of growth;
democratic regimes are more effective at producing wide distribution of
benefits (because of the strong tendency of authoritarian regimes to
structure economic activity towards “rent-seeking” activities, enrichment of
the ruling circle, and widespread corruption).
• democratic regimes are less prone to corruption and rent-seeking; they are
less “predatory”.

Assessment: democracy and development

Issues of democracy and development have an empirical manifestation;
since World War II over 100 nations have undergone a variety of processes of
political and economic development, so it should be possible to examine this 50-
year and 100-nation experience for statistical and causal associations among the
variables of interest. Is there a demonstrable correlation between the attributes
of democracy and the attributes of effective economic development? A large
number of empirical studies have been undertaken in the past 30 years to
investigate this question.3 However, the empirical case is suggestive but
inconclusive. The data support some optimism in support of the “compatibility”
theory: that democratic institutions have a net positive effect on economic
development. However, the association is empirically weak, and there are a
number of counter-examples in both directions: authoritarian regimes that have a
good development record, and democratic regimes that have weak development
records. In their major review of available cross-country studies of democracy
and development, Sirowy and Inkeles conclude that (1) there is little support for
a strong positive causal relation between democracy and development, and (2)
there is little empirical basis for choosing between the “conflict” hypothesis and
the null hypothesis (Sirowy and Inkeles 1990). Overall these authors conclude
that there are few robust conclusions that can be supported on the basis of
existing empirical multi-case studies of these factors. Sirowy and Inkeles
believe that methodological flaws in the studies are an important part of the
problem—leading to the possibility that more refined studies may shed greater
light. Przeworski and Limongi arrive at a similar conclusion. They examine 18
cross-country studies, and conclude that these studies do not provide a clear

3
 11

basis for conclusion about the causal properties of democratic institutions with
regard to development (Przeworski and Limongi 1993 : p. 60). Both of these
review essays point to the methodological difficulties that stand in the way of
effective statistical test of these causal hypotheses.4
This suggests, however, that it is reasonable to work on the assumption
that democratic institutions are compatible with effective economic
development.

Democracy and the poor

How does the presence of democratic institutions affect the viability of
progressive economic development strategies? Recall that “progressive”
economic development is defined as development that is designed to result in
wide distribution of the benefits of growth, significant and sustained
improvement in the quality of life of the population, and significant and
sustained improvement in the incomes and assets of the poor and near-poor.
The promise of democracy from the point of view of progressive
economic development follows from a very simple argument. The poor are
numerous. As parties compete for electoral support they have an interest in
adopting policies that favor the interests of the poor. It is in principle possible
for a political party representing the interests of the disadvantaged to acquire
substantial political influence in a third-world democracy, through its electoral
significance. And in countries in which there is such a political party, we should
expect that government policy will be accordingly tilted back in the direction of
the poor. Therefore we should expect a tendency for state policies to
accommodate the economic interests of the poor, and to begin to redress the
anti-poor tilt that is characteristic of authoritarian politics.
These considerations suggest that progressive development strategies
and third-world democratization movements need to flow hand in hand: regimes
whose political base depends on support from the poor and the near-poor will be
the most motivated to pursue a poverty-first program, and the most capable of
implementing such a program; whereas the existence of such a program within a
developing democracy provides a plausible basis for mobilizing further mass
support for the progressive development party.
There is a realistic core to this optimistic argument, but it is over-
simple in this formulation. More extensive democracy can be a central means of
furthering poverty-first economic development. But it is also clear, both
empirically and theoretically, that broad-based electoral democracy does not
unavoidably result in conferring political influence on the poor. There are
constraints on the political capacity of such a party. First, there are numerous

4
 12

channels through which elite interests can subvert the political goals of a party
of the poor. And second, there are structural constraints on the policies that such
a party can advocate, let alone implement, without creating an economic crisis
that worsens the condition of the poor.

Can democracies take hard measures?

Consider one final question-mark on the role of democracy within
development. Electoral democracies are reasonably effective in mobilizing
groups in defense of their economic interests, and the results bear the mark of
this process. It is difficult to implement policies within an electoral democracy
that impose economic hardship on politically effective groups. But development
(and economic reform more generally) unavoidably involves hardship for
various social groups. So the question arises: Do effective political demands
within the context of an electoral democracy paralyze development? The answer
to this question depends a great deal on institutional variables below the current
level of discussion: the political competence of existing parties, the ideology and
commitments of the governing party, the quality and effectiveness of leadership,
the level of confidence the electorate has in a regime’s intentions and
competence, the character and goals of existing sub-party organizations, and the
details of parliamentary institutions.5 The strongest conclusion that can be
drawn on the basis of the recent experience of Poland, for example, is that it is
possible to implement an aggressive program of reform through democratic
means, but that the political pressures build substantially as the reform program
begins to impose hardships on the populace. Moreover, there are instances
elsewhere in Eastern Europe (Hungary, for example), in which governing parties
have not succeeded in putting together strong electoral support for a unified
program of reform; in these cases, gridlock appears to be a very possible
outcome.6

Co-optation of democratic institutions by elites

It is a familiar fact in the democracies of the developing world that
economic elites often manage to retain disproportionate influence within a
democratic electoral system. The reasons for this privileging of elite interests
are not hard to find. Elites have privileged access to the instruments of political
influence—education, literacy, campaign finance. Elites are able to oppose
political strategies through the threat of capital strike. And elites are compact

6
 13

groups, so that their collective action problems are more easily handled than
those of more numerous groups.7 These considerations suggest that elites are
well-positioned to defend their economic interests within an electoral
competition—with the result that they will be able to preserve the benefits of
pre-existing anti-poor biases in economic policies.
Second, to the extent that non-elite groups emerge as politically
significant it is possible, perhaps likely, that the groups that stand to gain the
most political influence through democratization are not the poor, but the near-
poor: urban workers and consumers, better-off farmers, and the like. And the
interests of these groups are not identical with those of the poor. Consider one
example of a process that is almost ubiquitous in the developing world: the
political influence of civil servants, urban workers, and urban consumers. These
groups have an interest in securing food price policies that guarantee lower food
costs; they have an interest in development strategies that enhance urban
amenities (transportation, sanitation); and they have an interest in wage policies
that favor them. Further, these groups are well-positioned to back up their
demands with effective political action: mobilization around political parties,
personal and political relationships with government officials, and the threat of
urban unrest. So it is common to find that LDC policies reflect an urban bias:
food price policies, provision of infrastructure, and wage policies that favor
urban workers and civil servants. These politically-created benefits have the
effect of improving the material welfare of these groups—but at the expense of
the rural poor. The result of these policies is to depress the market-determined
incomes of farmers, to reduce the level and quality of amenities flowing to the
rural sector, and to further exacerbate the wage differentials between rural and
urban sectors. A consequence of this line of analysis, then, is to raise the
possibility that more democracy may in fact reduce the amount of attention the
poor (and particularly the rural poor) receive within the politics of development
policy.8
These arguments are not intended to discredit the significance of
democratic institutions in furthering a poverty-first economic strategy. Indeed,
it is unlikely that such a strategy will emerge except through an effective,
politically competent demand for such a strategy by the rural poor, supported by
an effective and administratively competent party strongly committed to its
interests. But democratization is not the only ingredient of a successful poverty-
first policy, and arguments in preceding paragraphs are designed merely to show
that it is quite possible for democratic electoral mechanisms to lead to outcomes
that neglect the poor or are positively biased against them.

8
 14

This line of thought suggests, then, that effective political action in

support of progressive economic development policies is most likely to come
into place within a context of effective electoral democracy, in the presence of
an administratively competent party of the poor.9