DELEGATIVE DEMOCRACY?

Guillermo O`DonneII

Guillermo ODonnell a Helen Kellogg Professor of International Studies, and Academic Director of the Helen Kellogg Institute for International Studies. The present work was done n the framework of the project East-South System Transformation, directed by Adam Przeworski. Research for this paper was supported by general research funds of the Kellogg institute, as well as by grants from the Ford Foundation and the International Development Research Centre to the Grupo de Estudos Polticos of CEBRAP, Brazil. This support is gratefully acknowledged. The pages that follow are a partial and preliminary product of research I am undertaking (a revised version will become part of a book I am writing). Several themes I can only allude here are discussed in On the State, Democratization and Some Conceptual Problems (A Latin American View with Glances at Some Post-Communist Countries), forthcoming in World Development, September 1993, and pre-published as Kellogg Institute Working Paper # 192 (April 1993).

ABSTRACT
This paper presents a first result of ongoing research on emerging forms of democracy. The author argues that it may be necessary to conceptualize a new type of delegative democracy, as different in some crucial respects from the representative democracy that has been theorized in the existing literature. The emergence and workings of delegative democracy are seen as closely interwoven with the deep social and economic crisis that some newly democratized countries are undergoing. But those relationships remain to be worked out by research in progress.

Preliminaries
Here I depict a new animal, a subtype of existing democracies, which has yet to be theorized. As often happens, the similitudes of this one with other, already recognized animals are many, with some cases shading off between the former and some variety of the latter. Still, I have convinced myself that the differences are significant enough to attempt such a depiction. Second, the drawing of neater boundaries between those types depends on empirical and, also, on more refined analytical work I am in the midst of. Third, if this is really a new animal (and not a member of an already recognized family, or a form too evanescent to merit conceptualization), sorting out its relationships with other factorswhat is cause, or effect, or mere correlationis the way to make the whole exercise of interest. As the reader will see in the pages that follow, in these respects many of my opinions are tentative, and I do not try to disentangle several causal relationships that are entailed in the argument. In papers, meetings, and discussions those of us who have worked on transitions and democratic consolidation have repeatedly said that, since it would obviously be wrong to assume equifinality of those processes, we need a typology of democracies. Some important efforts have been made, 1 focused on the consequences, in terms of types of democracy and policy patterns, of various paths to democratization. But, contrary to what I expected to find, my ongoing work suggests that the more decisive factors for generating various kinds of democracy are not those related to the characteristics of the process of transition from authoritarian rule. On one hand, longer-term historical factors and, on the other, the degree of severity of the socioeconomic crisis newly installed democratic governments may inherit, seem more important. The main points of my argument are: 1) Existing theories and typologies of democracy refer to representative democracy as practiced, with all its variations and subtypes, by developed capitalist countries. 2) Some newly installed democracies (Argentina, Brazil, Peru, Ecuador, and Bolivia, plus the Philippines and Korea, some Central and Eastern European countries, andat bestmany of the countries emerged of the dissolution of the Soviet Union) are democracies, in the sense that they meet Robert Dahls criteria for the definition of polyarchy.2 3) But these democracies are not-nor seem to be moving toward representative democracy; they present a set of characteristics which tempts me to call them delegative democracies (DD). 4) DDs are neither consolidated nor institutionalized democracies, but they may be enduring; in many cases, no imminent threat of an open authoritarian regression, nor advances toward institutionalized representativeness, are in sight. 5) FinaIly, argue that we see the effect of an important interaction: the deep social and economic crisis that most of these countries inherited from their authoritarian predecessors
1

Karl, Terry Lynn and Schmitter, philippe C., Modes of Transition and Types of Democracy in Latin America, Southern & Eastern Europe, International Social Science Journal, 128 (May 1991) pp. 269-284. 2 Dahl, Robert, Polyarchy. Participation and Opposition (New Haven: Yale University Press, 1971) and democracy and Its Critics (New Haven: Yale University Press, 1989).

powerfully multiplies the consequences of certain conceptions and practices that lead in the direction of delegative, not representative democracy. Now I will briefly state some criteria that underlie my preceding argument: 3 A. The installation of a democratically elected government opens the way for a second transition, probably longer and more complex than the transition from authoritarian rule. B. That second transition is supposed to be from a democratically-elected government to a democratic regime or, equivalently, to an institutionalized, consolidated democracy. C. Nothing guarantees that this second transition will be made: new democracies may regress to authoritarian rule, or they may stall in a feeble, uncertain situation. This situation may be enduring, but without opening avenues for the achievement of institutionalized types of democracy. D. The crucial element in determining the outcome of the second transition is success or failure in the building of a set of institutions which become important decisional points in the flow of political power. E. Such an outcome is contingent upon governmental policies and political strategies of various agents which embody the recognition of a paramount shared interest in the task of democratic institution building. Successful contemporary cases have exhibited great care, by a winning coalition of political leaders, in advancing toward the creation and strengthening of democratic political institutions and, to a lesser extent, of interest representation. In turn, these achievements have facilitated reasonable success in dealing with the social and economic problems inherited from the authoritarian predecessorsSpain, most clearly; Portugal, although not immediately after democratic installation; Uruguay; and, according to every indication up to now, Chile. F. In contrast, the cases l mentioned at the beginning of this section have achieved neither institutional progress nor governmental effectiveness in dealing with the respective social and economic crises. Most of these cases fail into the category of delegative democracy. Before dealing with some of the theme implied by the preceding enunciations, I need an excursus on what I mean by institutions and institutionalization.

In have argued these views in Reis, Fbio Wanderley and ODonnell, Guillermo, eds., A Democracia no Brasil. Dilemas e Perspectivas (So Paulo: Editora Vrtice, 1988)

On Institutions Institutions are regularized patterns of interaction that are known practiced and regularly accepted (if not necessarily normatively approved) by given social agents who by virtue of those characteristics, expect to continue interacting under the rules and norms formally or informally embodied in those patterns. Sometimes, but not necessarily, institutions become formal organizations: they materialize in buildings, seals, rituals, and persons and roles that authorize them to speak for the organization. I am concerned here with a subset: democratic institutions. Their definition is elusive, so I will delimit the concept by way of some approximations. To begin with, democratic institutions are political institutions in a broad sense: they have some recognizable, direct relationship with the main themes of politics: the making of decisions that are mandatory within a given territory, the channels of access to those decisions and to the governing roles that enable making them, and the shaping of the interests and identities that claim access to those channels and decisions. The boundaries between what is and is not a political institution are blurred and tend to vary across time and countries. This is an interesting empirical and theoretical question: it pertains to what institutions may or may not be politicized in various types and stages of democratization. We need a second approximation: some political institutions are formal organizations belonging to the constitutional, network of a polyarchy; these include Congress, the Judiciary, and at least more than one political party. Others, such as fair elections, have an intermittent organizational embodiment but are no less indispensable. The question about these, quite obviously, is how they actually work: Are they really important decisional points in the flows of influence, pressure, and policy? If they are not, what are the consequences for the overall political process? Other levels indispensable for me workings of democracy in contemporary societiesthose that pertain to the formation and representation of collective identities and interestsmay or may not be institutionalized, or they may be operative only for a biased part of the potentially relevant sectors. Through pluralist or neocorporatist arrangements, those patterns are highly institutionalized and organizationally embodied in consolidated democracies. Now I describe some characteristics of a functioning institutional setting. 1) Institutions incorporate AND exclude. Institutions establish what agents, on the basis of what resources, claims and procedures, are accepted as valid voices in their decision processes, both in making decisions and in implementing them. These are necessarily selective criteria, which fit (and favor) some agents, may lead others to reshape themselves in order to meet that criteria, and which for various reasons are impossible to meet, or unacceptable, for others. The .scope of en institution is the degree to which it actually incorporates and excludes a set of potentially relevant agents.

2) Institutions shape the probability distribution of outcomes. As Adam Przeworski has noted,4 institutions process only certain actors and resources, and do it under certain rules. This predetermines the range of feasible outcomes, and the likelihood of those within the range. Democratic institutions, for example, preclude the use or threat of force, and the outcomes it would generate. On the other hand, the subset of democratic institutions based on the universality of the vote, as Philippe Schmitter and Wolfgang Streek have argued,5 is not good at processing intensities of preferences. Institutions of interest representation get closer to that processing, although at the expense of the universalism of the voting mechanism and of the citizen principle and, often, of the democraticness of their organizational decision-making. 3) Institutions tend to aggregate, and to stabilize the aggregation of the level of action and organization of agents interacting with them. The rules established by institutions influence strategic decisions by agents regarding the degree of aggregation that is more efficacious (in terms of the likelihood of favorable outcomes) for them. Institutionsrather, the persons in institutional roles that enable them to make decisions that are attributed, and based upon, the institutions authorityhave limited information-processing and attention capabilities. Consequently, those persons prefer to interact with relatively few agents and issues at a time. 6 This tendency toward aggregation is another reason for the exclusionary side of every institution. 4) Institutions induce patterns of representation. For the same reasons noted, institutions favor the transformation of the many potential voices of their constituencies into a few that claim to speak as representatives of the former. Representation involves, on one hand, the acknowledged right to speak for some relevant others and, on the other, the ability to deliver the compliance of those others to what the representative decides. Insofar as this capability is demonstrated and the given rules of the game are respected, institutions and the various interacting representatives develop an interest in their mutual persistence as interacting agents. 5) Institutions stabilize agents/representatives and expectations. Institutional leaders and representatives come to expect from each other behaviors within a relatively narrow range of possibilities, from a set of actors that they expect to meet again in the next round of interactions. Certain agents may not like the narrowing of expected behaviors, but they anticipate that deviations from such expectations are likely to be counterproductive. This is the point when it may be said that an institution (which probably has already become a formal organization) is strong: it is at equilibrium, and it is in none

See, esp., Przeworski, Adam, The State and the Economy under Capitalism (New York: Harwood Academic Publishers, 1990) and Democracy as a Contingent Outcome of Conflicts, in Elster, Jon and Slagstad, Rume, eds., Constitutionalism and Democracy (Cambridge: Cambridge University Press, 1988) pp. 59-80. 5 Streeck, Wolfgang and Schmitter, Philippe C., Community, Market, State and Associations? The Prospective Contribution of Interest Governance to Social Order in Streeck, Wolfgang and Schmitter, Philippe C., eds., Private Interest Government. Beyond Market and State (London: Sage Publications, 1985) pp. 1-29. 6 See esp. March, James and Olsen, James, Rediscovering Institutions. The Organizational Basis of Politics (New York: The Free Press, 1989).

of the agents interest to change it except in incremental and basically consensual ways. 6) Institutions lengthen the time horizons of actors. The stabilization of agents and expectations entails a time dimension: institutionalized interactions are expected to continue into the future, most likely among the same (or in a slowly and rather predictably changing) set of agents. This, together with a high level of aggregation of representation and of control of the constituencies, is the foundation for the competitive cooperation that characterizes consolidated democracies: one-shot prisoners dilemmas can be overcome, bargaining (including logrolling) is facilitated, the compensation of various trade-offs along time becomes feasible, and sequential attention to issues unloads an otherwise unmanageable agenda. The establishment of those practices further strengthens the willingness of all relevant actors to recognize each other as valid interlocutors, and enhances the value that they attach to the institution that shapes their interrelationships. This virtuous circle is completed when all or mot democratic institutions achieve not only reasonable scope-and strength but also, at a more aggregate level, they reach the high density resulting in multiple, and stabilized, mutual relationships that locate those institutions as important decision points in the overall political process. The re gime of a consolidated, institutionalized democracy thus emerges. Perhaps a good way to summarize what I have said is that, in the functioning of contemporary, complex societies, democratic political institutions are a crucial level of mediation and aggregation between, on one side, structural factors and, on the other, not only individuals but also the diverse groupings under which society tends to organize its multiple interests and identities. That intermediate institutionallevel has important impacts on the patterns of organization of individuals, making some of them representable voices in the political process and excluding others. That same institutional level has less immediate and easily detectable, but still important, impacts on structural factors, particularly with respect to the changes that public and private decisions may provoke when formulated with an appropriate time horizon Institutionalization undeniably entails heavy costsnot only exclusion but also the recurring, and all too real, nightmares of bureaucratization and boredom. The alternative, instead, submerges social and political life in the hell of a colossal prisoners dilemma. This is, of course, an ideal typical description. I find it useful for tracing, by way of contrast, the peculiarities of a situation characterized by a dearth of democratic institutions. A noninstitutionalized democracy is characterized by the restricted scope, the weakness, and the low density of whatever political institutions it has. Other, nonformalized but strongly operative practices especially clientelism, patrimonialism, and, indeed, corruptiontake the place of the former, jointly with various patterns of highly disaggregrated and direct access to the policymaking process.

Toward a Characterization of Delegative Democracy Delegative democracies are grounded on one basic premise: he (or eventually she, i.e., Indira Gandhi, Corazn Aquino, and Isabel Pern) who wins a presidential election is enabled to govern the country as he sees fit, and to the extent that existing power relations allow, for the term to which he has been elected. The President is the embodiment of the nation and the main custodian of the national interest, which it is incumbent upon him lo define. What he does in government does not need to bear any resemblance to what he said or promised during the electoral campaignhe has been authorized to govern as he sees tit. Since this paternal figure has to take care of the whole nation, it is almost obvious that his support cannot come from a party; his political basis has to be a movement, the supposedly vibrant overcoming of the factionalism and conflicts that parties bring about. Typically, and consistently, winning presidential candidates in DDs present themselves as above all parties: i.e., both political parties and organized interests. How could it be otherwise for somebody who claims to embody the whole of the nation? In this view other institutions such as Congress and the Judiciaryare nuisances that come attached to the domestic and international advantages of being a democratically elected President. Accountability to those institutions, or to other private or semi-private organizations, appears as an unnecessary impediment to the full authority that the President has been delegated to exercise. Delegative democracy is not alien lo the democratic tradition. Actually, it is more democratic, but less liberal, than representative democracy. DD is strongly majoritarian: democracy is constituting, in clean elections, a majority than empowers somebody to become, for a given number of years, the embodiment and interpreter of the high interests of the nation. Often DDs use devices such as the ballotage: if elections do not directly generate a clear-cut majority, that majority must be created for supporting the myth of legitimate delegation. Furthermore, DD is strongly individualistic, but more of a Hobbesian than a Lockean variety: voters are supposed to choose, irrespective of their identities and aftiliations the individual who is more fit for taking care of the destinies of the country. Elections in DDs are a very emotional and high-stakes process: various candidates compete to be the absolutely zero-sum winner of the delegation to rule the country with no other constraints than those imposed by nakedi.e., noninstitutionalizedpower relations. After the election, voters/delegators are expected to return to the condition of passive, but hopefully cheering, spectators of what the President does. Extreme individualism in the moment of constituting presidential power combines well with the organicism of the Leviathan. The nation and its authentic political expression, the Movement, 7 are postulated as living organisms. The nation has to be healed and saved by uniting its chaotically dispersed fragments (sectorialism, political parties, egoism) into a harmonious
7

ln various of his writings Giorgio Alberti has perceptively insisted on the importance of movimientismo as a dominant (and highly negative) feature of politics in many Latin American countries. Albertis most recent analysis of this theme is Democracy by default, economic crisis, and social anomie in Latin America, Universit di Bologna, Facolt di Scienze Politiche and CESDE (Centro Europeo di Studi sulla Democratizzazione), paper presented to the XVth World Congress of Political Science, Buenos Aires, 1991.

whole. Since that body is in disarray, and since its existing voices only reproduce its fragmentation, delegation includes the rightactually, the obligationof applying to the nation the tough medicines that, even though many of its members cannot recognize it now, will heal it. For this organicistic view it seems obvious that only the head really knows. The President and his personal staff are the alpha and omega of politics. Furthermore as in major surgery, some of the problems of the nation can only be solved by highly technical criteria. Tcnicos, especially in economic policy, must be politically protected by the President against the manifold resistances of society, until the process of convalescence is well advanced. In the meantime, it is obvious that resistancein Congress and parties, or from interest representation associations, or in the streetshas to be ignored. The organicistic discourse rhymes poorly with the dry arguments of the technocrats, and the myth of delegation is perversely consummated: the President isolates himself from most existing political institutions and organized interests, and bears sole responsibility for the successes and failures of his policies. Indeed the conceptions have sketched were strongly present in recent bureaucraticauthoritarian regimes, particularly the more technocratically oriented ones. They are present, too, in delegative democracies. But the different overall political context makes a difference. In DDs, parties and Congress express their criticisms about the policies undertaken. Sometimes the courts, based on legalistic, formalistic reasons, block blatantly unconstitutional measures. Workers and capitalists associations complain loudly. The party (or parties) that elected the President despair about their loss of popular support, and begin to refuse parliamentary support to his policies. This further increases the political isolation of the President, his difficulties in forming a stable coalition in Congress, and his propensity to sidestep, ignore, and/or corrupt that and other institutions.8 I will return to these themes after the more general considerations with which I close the present section. The idea of representation involves an element of delegation: through some procedure, a given collectivity authorizes somebody to speak for it and eventually to commit the collectivity to abide by what the representative has decided in her capacity as representative. Consequently, representation and delegation are not polar opposites. This is why it is difficult to sharply distinguish types of democracy which are organized around what I would call representative delegation from those where the delegative element is strongly predominant. Representation entails the idea of accountability: somehow the representative is held responsible for the ways in which he acts in the name of those for whom he claims to be entitled to speak. In consolidated democracies, accountability operates not only, nor so much, vertically in relation to those who elected the officer (except, retrospectively, at times of elections), but horizontally in relation to a network of relatively autonomous powers (i.e., other institutions) that have the capacity of calling into question and eventually punishing improper ways of discharging the responsibilities of the given
8

I do not believe that these themes can be reduced to theimportant-discussions currently underway about various forms of presidentialism and parliamentarism. Clearly presidentialism has more affinity with DD than parliamentarism. But if in a given country delegative propensities are strong and pervasive, the actual workings of a parliamentary system could be easily subverted and/or lead to impasses even worse than the ones discussed here.

officer. Representation and accountability, in turn, entail what in previous work 9 I have called the republican dimension of democracy: a careful distinction between the spheres of public and private interests of office holders. Notice that what matters for the effectiveness of horizontal accountability is not only the values and beliefs of officers (whether elected or not) but also the fact that they are embedded in a network of institutionalized power relations. Since the punishing capabilities of those, relations may be mobilized, a rational actor will calculate the likely costs when she considers undertaking some kinds of improper behavior. Of course, the actual workings of this system of mutual responsibility leave much to be desired everywhere. Still, what holds in institutionalized democracies in terms of the influence of the rule-like force of certain codes of conduct and of deterrence of improper acts, even if difficult to gauge, entails a very significant difference from situations where little of the above holds. Because policies must go through a series of relatively autonomous powers, decision making in representative democracies is slow and incremental. But, by this same token, it is usually vaccinated against gross mistakes, so that many decisions have a reasonably good chance of being implemented, and responsibility for mistakes tends to be widely shared. Delegative democracy begins with very low institutionalization and, at best, it is indifferent toward strengthening it. DD gives the President the apparent advantage of practically no horizontal accountability. DD has the additional apparent advantage of allowing swift policy-making, but at the expense of a high likelihood of gross mistakes, of hazardous implementation, and of concentrating responsibility for the outcomes on the President Not surprisingly, these Presidents suffer the wildest swings in popularity: today they are acclaimed saviors, tomorrow they are cursed as only fallen gods can be. Whether it is called culture, tradition, or historically-structured learning, the plebiscitary and caudillista tendencies toward delegative democracy are detectable in most Latin American (and, for that matter, many Central/East European, post-Soviet, African, and Asian) countries long before the present social and economic crisis. With the anachronistic exception of the dictatorship of ancient Rome, this type of rule has been theorized as a chapter in the study of authoritarianism under names such as caesarism, bonapartism, caudillismo, and populism. But we need to see this type of rule also as an eventually enduring democratic form, which has some interesting overlaps and differences (which I cannot elaborate here) with those authoritarian forms. But, even if DD belongs to the democratic gender, it would be hard to find something that is more uncongenial to the building and strengthening of democratic political institutions.

Historical Background The great wave of democratization prior to the one we are witnessing occurred
9

See note 3.

immediately after World War II, as an imposition by the allied powers on defeated Germany, ltaly, Japan, and to some extent Austria. The resulting conditions were remarkably different from the ones faced today by Latin America and Eastem Europe: 1) After the destruction provoked by the war, economic expectations of the respective populations were, most likely, extremely moderate. 2) There were massive injections of capital, principally, but not exclusively (i.e., the condonation of Germanys foreign debt), through the Marshall Plan. 3) As a consequence, and due to an expanding world economy, those countries soon achieved rapid rates of economic growth. These were not the only factors, but they greatly helped in the successful consolidation of democracy in those countries. Furthermore, those factors contributed to political stability and to rather stable public policy orientations. In contrast in the transitions of the 1970s and 1980s, as an eloquent reflection of the much less congenial context in which they occurred, winning the first election after the demise of the authoritarian regime guaranteed that the victorious party would be defeated, if not virtually disappear, in the next election witness Spain, Portugal and Greece, as well as Argentina, Bolivia, Brazil, Ecuador, Peru, and Uruguay. But this regularity appears together with important variations in terms of the social and economic performance of the new governments. Most of these countries inherited a difficult situation from the preceding authoritarian regime and were severely affected by the world crisis of the 1970s and early 1980s. In all of them the socioeconomic situation at some point was recognized as extremely critical and requiring strong, determined action by the government. But there is no question that however serious, objectively and subjectively, the economic problems in Southern Europe, they look mild when compared with those inherited by the newly democratized countries in the East and the South (with Chile as a partial, exception). High inflation, economic stagnation, a deep financial crisis of the state, increased inequality, and sharp deterioration of social policies and welfare provisions are aspects of this crisis. But, again, important differences emerge: the Uruguayan economy performed decently. It lowered annual inflation from three to two digits, and GNP, investment and real wages increased slowly (but increased). The Uruguayan government pursued incremental economic policies, most of them negotiated with Congress and various organized interests. Chile is following the same path. Instead, Argentina, Brazil, and Peru adopted a strategy of drastic, surprising economic stabilization policy packages: Austral in Argentina, Cruzado in Brazil, Inti in Peru, and their no less unfortunate successors. These policies did not solve any of the inherited problems; rather, it is difficult to find a single problem that they did not worsen. Bolivia also adopted one of these packages, actually the most orthodox of all. It is hailed as a big success, since inflation was sharply reduced. But after several years GNP and investment are still anemic, and the brutal repression that dealt with workers resistance to the package hardly qualifies as democratic. With the benefit of hindsight (although the skepticism of a few of us is on record from the very beginning) it is easy to see that those packages have been disastrous, although there is no agreement whether they are disastrous per se,

or because the particular ones adopted were flawed for some specific reasons, or because they were alt right but exogenous political factors caused their failure. Whichever the reasons, it is clear that the newly installed Chilean government is not going to follow this path. Post hoc the disastrous experiences of those policies, this is not too surprising. This makes Uruguaya country that inherited a situation that was no better than Argentinas and Brazilsa most interesting case. Why did the Uruguayan government not adopt its own package, especially during the euphoria that followed the first stages of the Austral and the Cruzado? Was it because President Sanguinetti and his collaborators were more intelligent, better economists or better informed than their Argentina, Brazilian, and Peruvian counterparts? It happens that in this case of redemocratization, although far from being the perfect institution that exists nowhere, Congress effectively carne back to work at the very moment of democratic installation. Because of constitutional restrictions and historically embedded practices, the President does not have the power to unilaterally decree things such as the stabiIization packages of the neighboring countries. The President of Uruguay, for the enactment of many of the policies typically contained in those policies, must go through Congress. In other words, the elements of secrecy and surprise that seem so fundamental to those packages are de facto eliminated. Furthermore, going through Congress means having to negotiate those policies not only with parties and legislators, but also with various organized interests. Consequently, against the apparent preferences of top members of the executive, the economic policies of the Uruguayan government were condemned to be incremental, sometimes inconsistent, and limited to quite modest goalssuch as achieving the decent performance we have seen, not the heroic goals with which the stabilization packages of other countries were heralded. Looking at Uruguay one learns about the difference between having or not having a network of institutionalized powers that texture the policy-making process. Or, in other words, between representativa and delegative democracy. Crisis Now I will focus on the purest Latin American cases of delegative democracy Argentina, Brazil, and Peru. I do not need to detail the depth of the crisis these countries inherited from the preceding authoritarian regimes, or the even worse condition they are in today. A deep social and economic crisis is the ideal terrain for unleashing the delegative propensities that may be present in a given country. Such a crisis generates a strong sense of urgency. Problems and demands accumulate for new democratic governments which are inexperienced and must operate through a weak and disarticulated (if not disloyal) bureaucracy. Presidents get elected promising that theystrong, courageous, above parties and interests, machoswill save the country. Theirs is a government of saviors (salvadores de la patria). This, in turn, leads to a magical style of policy making: the delegative mandate to rule supposedly emanated from the majority, strong political will, and adequate technical knowledge should suffice to fulfill the saviors missionthe package follow as a corollary. The resulting style (and conception) of policy-making can only ignore parties (including the ones that supported the election of the President),

Congress, the Judiciary, and practically all interest-representation organizations. Soon afterwards, whenif there is any chance for the economic policy packages to succeed, it lies in the supporting actions of those and other agentsthe self-induced solitude of that style of policy-making, plus the institutional weaknesses it has accentuated, has enormous difficulties in bringing the relevant parts into the alternate magical solution: the socioeconomic pact. Then, characteristically, the Executive complains about the selfishness of politicians and interest representatives, further undermining the institutions it began by ignoring and entering into even more harmful sequences of packages and futile attempts at pact-making. The return with a vengeance (objectively and because of the deteriorated authority of the President) of the crisis has consequences that need study that I have not yet completed. Suffice it to say that the longer and the deeper the crisis, and the less the confidence that the government will be able lo solve it, the more rational it becomes for everyone to act: 1) at highly disaggregated levels, especially in relation to state agencies that may help to solve or alleviate the consequences of the crisis for a given group or sector; further disarticulation and weakeningand corruptionof the state apparatus follow; 2) with extremely short time-horizons; and 3) with assumptions that everyone else will do the same. This colossal prisoners dilemma is the exact opposite to the conditions leading to the emergence and strengthening of democratic institutions, or to reasonably effectively dealing with the salient problems of the time. Once the initial hopes are dispelled and the first packages have failed, cynicism about politics, politicians, and government becomes the pervading mood. lf these governments can hope to keep some support from the population, they must, at least, control inflation and implement some social policies that would show that, even though they cannot rapidly solve most of the underlying problems, they do care about the fate of the poor and (politically more important) of the recently impoverished middle-class segments of the population. This, minimal as it is, is a very tall order. First, those two goals are extremely difficult to harmonize. Such incompatibility is not a logically necessity; it springs from the fact that, to be overcome, it needs an efficaciouslean but strong state that neither the historical heritage of these countries, the profound fiscal crisis that is part and engine of the overall economic crisis, nor the current wave of fervent antistatism allow. Second, orthodox adjustment programs are, at least in the short run, hardly consistent with improving the welfare of a large proportion of the population. One consequence is that governments and social actors begin to act under ever shorter time horizons, which makes them even more unlikely to solve the underlying crisis. Governments like lo have continued popular support, and politicians want to be reelected. Only if the predicaments entailed by what I said above were solvable within the few years of a presidential constitutional mandate, would winning an election be a triumph instead of a terrible curse. How does one get elected, and how does one govern after elected in this type of situation? Quite obviously and most destructively in terms of the building of the public trust that helps a

democracy to consolidateby doing exactly the contrary in one and another period. President Bushs Read my lips shows that even institutionalized democracies are not immune to this type of trick; but the consequences are more devastating when there are few and weak political institutions, not to say a much deeper social and economic crisis. Presidents have been elected in Argentina, Bolivia, Brazil, Ecuador and Peru promising expansionist neoKeynesian policies and many other good things to come with themonly to immediately, or shortly after beginning their mandate, the opposite. How necessary various degrees or harshness (or blandness) of adjustment policies are I cannot discuss here. But it is easy to see that the impact of that shift is not exactly in the direction of building public trust, particularly if the immediateand most perceivableimpact of those policies hinders the already low levels of welfare of most of the population. In addition, the marginalization of parties and Congress from the most important decisions facing the country has three serious consequences: i) it further deepens the very defects that are imputed to those institutions; ii) when, finaIly and fatally, the Executive needs Legislative support, it is bound to find a Congress that is not only resentful, but that does not feel politically responsible for policies that began by bypassing it; and iii) this situation, jointly with the criticisms of the Executive about the slowness and irresponsibility of a Congress that stalls in the required support, are an important factor in the sharp decline of prestige of all parties and politicians, as opinion polis in these countries abundantly show. If we also take in consideration that, for reasons I cannot elaborate here (but closely connected to what I said above), the Executive does nothing to strengthen the Judiciary, the resulting dearth of reasonably effective and autonomous institutions places immense responsibilities on the President. We should remember that often he is elected promising that he would save the country without high costs to anyone, and that as soon as he is elected he gambles his government to the success of policies that entail almost the opposite of those promises. This soon results in policy making under despair: the shift from wide popularity to generalized execration can be as fast as it is steep. The result is a curious mixture of Presidential omnipotence and impotence. Omnipotence which begins with the spectacular enactment (by Executive order, decreto, not by law) of the first policy packages and continues with the flurry of decisions aimed al complementing and, unavoidably, correcting the numerous unwanted consequences of the former. This further accentuates the anti-institutionaIizing bias of these processes and ratifies traditions of high personalization and concentration of power in the Executive. But the other side of the coin is extreme weakness, if not sheer impotence, in making those decisions effective regulations of societal life. As noted above, consolidated democracies are slow at making decisions. But once those decisions are made, they are likely lo be implemented. In the cases that concern us here, in contrast, we witness to a decisional frenzy: an appalling number of rapidly made decisions (decretismo). But, because they are unilaterally issued while they hinder important and politically mobilized interests, those decisions are unlikely to be implemented. In the midst of a severe crisis

and of increasing popular impatience, this style of policy-making leads to new flurries of decisions which, because of the experience many sectors have had in resisting the previous ones, are even less likely to be implemented, Furthermore, because of the way those decisions are made, almost every political social and economic agent can reasonably disclaim responsibility for those policies. As it was delegated to him, the President did what he deemed best. When failures accumulate too visibly and repeatedly, the country is stuck with a widely criticized President who, abandoning the initial dynamism, merely tries to survive in office until the end of his mandate. The resulting period of passivity and of utter disaggregation of public policy does nothing to improve the situation of the country. It is strikingand suggestive of the quite remarkable capacity of endurance of these curious democraciesthat, with very few exceptions, not even in those cases has a successful coup detat taken place. In contemporary Latin American democratizations, only Uruguay and Chile are managing to escape the infernal circles I have described. But Uruguay and Chile, as soon as they redemocratized, brought back their past working institutions and congenial practices which the other new Latin American democracies, as well as most of the new democracies in other regions, lack. This is the quandary: effective institutions and congenial practices cannot be implanted by decree. As consolidated democracies show, the emergence, strengthening and legitimation of those practices and institutions take time, during which a complex process of positive learning is involved. On the other hand, the tremendous economic and social crisis of most newly democratized countries would require that those institutions were already in place to deal with reasonable effectiveness with the numerous, urgent, complex, and interrelated problems. But that very crisis hinders even further the already difficult task of institutionalization. I am depicting what I feel is a terrible drama. It is the drama of countries lacking a democratic tradition thatas all new democracies didhave to cope with many negative legacies of their authoritarian past, but which, in addition, are faced with an extraordinarily deep social and economic crisis. I have mentioned but not analyzed the social dimension of this crisis. This theme is too broad and complex to be discussed here. It will suffice to mention the enormous inequalities existing in Latin America, which pose not only obvious problems of elementary social equity, but also of the political organization and representation of broad and, in some cases, majoritarian segments of a population that has won the right to vote. In particularly unequal countries, such as Brazil and Peru, this has led to wild fluctuations from one election to the other, which have further hindered the emergence of a reasonably stable and representative party system. Those inequalities have deepened since the early 1970s, with the further aggravation that broad segments of the middle class have dropped into poverty. In Latin America we have persistent and deep inequalities. In Eastern Europe we are witnessing a rapid process of inequalization, which may be even more politically explosive than the Latin American pattern. Finally, it is clear that, whatever economic view one subscribes to, taming

inflation and resuming growth is contingent on the stabilization of favorable expectations. Such achievements, in turn, are indispensable for national and international capitalists making the investments which would allow recovering reasonable rates of economic growth. But, given the combination of factors I have described, the prevailing style of policy-making has a self-defeating quality: abrupt and often surprising decisions, in the midst of social disarray and in front of strong resistances, may diminish inflation. But, since many uncertainties spring from those resistances, those policies have a long way to go before convincing relevant agents that they can count on a stable time horizon for their decisions. The remaining hope for governments is to continue doing more of me same but, such insistence is likely to further increase social disarray and resistances. This, in turn, paves the way for another presidential candidate who, by promising a complete overhaul of existing policies, imposes a resounding defeat on the party of the current president...only to immediately reenter this circle. An optimistic view of these cycles would argue that they have a rather predictable quality, upon which some longer-term perspectives could be built; but this view begs the question of how long the bulk of the population would be willing to play such a game. Another optimistic possibility would be that a predominant segment of the political leadership learns the self-defeating quality of those cycles and agrees to change the terms in which they compete electorally and govern. This sounds to me as practically the only chance, but the obstacles to such a roundabout but ultimately happy outcome are many.