Bolivia - The Long Way to Stabilization

1. Introduction

In the middle of the 1980s, one of the poorest countries in Latin America received considerable attention from scholars and politicians around the world. Bolivia had succeeded in ending a hyperinflation that ran at 24.000% at its peak in only a few weeks and had successfully stabilized its economy subsequently. The rightist government, which had been elected in 1985, had introduced the most orthodox reform package that had been implemented in Latin America until then. This reform package not only comprised macroeconomic stabilization measures but also introduced far-reaching liberalization measures.

Even though this achievement is impressive, another aspect of the stabilization merits attention. The Bolivian economy had been on a downward slope for some time and had started to deteriorate badly in 1982. Despite the apparent and urgent need for stabilization, reform was delayed for three whole years. This paper addresses the question why in Bolivia's hyperinflation crisis of the 1980s stabilization was delayed for so long.

This question arises because it seems irrational at first sight that reform was not undertaken despite a hyperinflationary situation. Moreover, while a lot of literature exists on the nature and consequences of the 1985 reform package in Bolivia, there is only little writing assessing the reasons for the delay from a theoretical perspective. Several models exist in the literature which offer theoretical explanations of delay in reforms. In order to test these models and to find a model that plausibly explains the process leading up to stabilization in Bolivia, these models should be applied to the Bolivian case.

I will begin with an overview over existing models on delayed reform. By identifying the key parameters and variables and roughly comparing them with the Bolivian experience, I will come to the conclusion that only a few offer a good explanation of the delay which occurred in Bolivia. Especially relevant are the models by Haggard and Kaufmann (1995), whose explanation focuses on the institutional weaknesses of a young democracy, Labán and Sturzenegger (1994), who concentrate on uncertainty about individual benefits as a major obstacle to reform, and Alesina and Drazen (1991), who model delay as a war of attrition resulting from a distributional struggle. Since distributional conflict is one of the key features of the process leading up to stabilization in Bolivia, I will then examine in detail whether the Alesina-Drazen model captures the key characteristics of the Bolivian case. I will conclude that it does offer a good explanation because all three predictions that it makes about the features of the stabilization process can be discerned in the Bolivian experience: Agreement on the need for reform and simultaneous disagreement on the distribution of the costs of reform, failed attempts at reform preceding stabilization and stabilization coinciding with political consolidation Section 2 of this paper will give with a brief chronology of the events leading up to stabilization. Section 3 will offer an overview about the different types of models that explain delays in reform. After determining which models offer reasonable explanations for the Bolivian experience, Section 4 will examine in detail whether the economic crisis and reform in Bolivia can be explained by the parameters set by the Alesina-Drazen model. Section 5 concludes.

2. Hyperinflation, Crisis and Reform in Bolivia

2.1 Pre-crisis Bolivia

2.1.1 The Political Situation

Modern Bolivian history begins in the year 1952 with the Bolivian Revolution, which brought the MNR (National Revolutionary Movement) to power. This party represented the interests of several of the major political forces in Bolivia: The miners, the peasants, the middle class and, initially, the armed forces (García, 1991, 455-6) resulting in the MNR agenda being mainly populist, with a strong focus on the public sector

The MNR stayed in power for twelve years, until it was deposed by a military coup in 1964. What followed was a period of political instability with frequent changes between military and civilian regimes. The alteration in power, exacerbated by intense conflicts over income distribution, led to frequent changes in populist and anti-populist policies. Leftist leaders generally tried to fulfil their distributional aims by raising public sector wages and employment. They failed, however, to finance these expenditures by raising tax revenues and therefore tended to rely on inflationary finance and foreign borrowing. Rightist leaders, on the other hand, often used government subsidies to channel money to favored segments of society, which they paid for through a reduction of public wages and increases in foreign borrowing. It is striking that both leftist and rightist leaders continuously relied on a large public sector and financed their policies largely through foreign borrowing. It is therefore not surprising, that the accumulation of debt posed a major problem to any government looking to reform the Bolivian economy.

The period of political instability culminated in the years between 1979 and 1982, in which seven military and three civilian heads of state presided over the country in only three years. In 1980, elections took place which produced a plurality (but no majority)¹ of votes for the UDP (Unión Democrática y Popular) and their candidate, former president Hernán Siles Suazo. However, he could not take office since General Luis Garcia Meza successfully plotted a coup and installed one of the most corrupt and brutal regimes in Bolivian history, which lasted until 1982.

2.1.2 The economic situation

Bolivia is one of the poorest countries in Latin America. Its economy has traditionally focused mainly on two industries: mining and petroleum². Due to the ruggedness of the Bolivian territory, transport is costly and only the production of high-value added commodities is profitable for export. The majority of the manufacturing industry produces only for local consumption. However, almost half of the Bolivian population earns their living in the agricultural sector, although oftentimes this is through subsistence agriculture only (Statistisches Bundesamt, 1991, 42).

After the Revolution of 1952, the MNR nationalized the mining and petroleum companies into the public enterprises COMIBOL (Corporacíon Minera de Bolivia), the state mining corporation, and YPFB (Yacimientos Petroliferos Fiscales de Bolivia), the state petroleum company, and "an economic system which can broadly be called state capitalism" (Morales and Sachs, 1990, 177) was established. Despite the frequent changes in government and despite the simultaneous emergence of a middle class of small private entrepreneurs, the predominant role of the state in the economy was upheld.

The period of political chaos was accompanied by the onset of an economic crisis. In 1982, when the UDP government took office, Bolivia faced negative growth of GDP and a growing public sector deficit (in 1982 93,53 Bil. Bolivian pesos, or 14.7% of GDP). Export revenues were falling, while at the same time external debt was mounting (and stood at 803 Bil. US$ in 1982) (Dunkerley, 1990, Appendix II and Statistisches Bundesamt, 1986, 61). The deteriorating economic situation was finally the main reason for the retreat of the military and the installation of the civilian regime under President Siles Suazo.

2.2 The UDP government and Hyperinflation

When the military withdrew from politics in 1982, the results of the 1980 elections were used to reconvene Congress. Hernán Siles Suazo's UDP (Unión Democrática y Popular) coalition had received a plurality of votes in the 1980 elections, but the rightist MNR and ADN (Acción Democrática Nacionalista) shared a majority of seats in Congress. According to the constitution, the Congress selects the president in cases, like the 1980 elections, where no candidate receives an outright majority of votes³. In 1982, "the standoff among competing party and social forces was resolved through a behind-the-scenes bargain struck among the military, unions, business groups, and parties: Siles would assume the presidency but the opposition parties would control the legislature." (Haggard and Kaufman, 1995, 184).

Siles' governing coalition UDP, consisted of three leftist parties: The MNRI (Movimiento Nacionalista Revolucionario de la Izquierda), the MIR (Movimiento de la Izqierda Revolucionaria), and the PCB (Partido Comunista de Bolivia), while the MNR and ADN formed the opposition. The governing UDP was a rather fragile coalition: Its composition changed several times and frequent cabinet changes were the order of the day. It was the first democratically elected government in 18 years and the Bolivian Left in particular pinned high expectations on to the UDP administration.

However, from the beginning the UDP coalition faced serious problems. First of all, as explained above, the UDP lacked a majority in parliament. Second, the UDP, who owed its success mainly to the support of the COB (Central Obrero Boliviano), the Bolivian association of trade unions, was vulnerable to demands from the workers. Third, the UDP coalition was fragile and lacked decisive leadership.

The most pressing problem, however, was the deteriorating state of the economy. Several external factors aggravated the crisis, such as a rise in the world interest rates and a cut off from foreign loans, as well as a fall in world prices for Bolivia's main exports. Most notably, tin prices fell by 24.3% from 1980 to 1982 (and by 71.2% between 1980-86), resulting in lost tax revenues (Dunkerley, 1990, Appendix V). Bolivia's government thus faced a situation in which falling foreign capital inflow, higher debt service and falling tax revenues⁴ diminished the government's traditional sources of capital. At the same time, the Left pressured the government for an increase in public sector spending.

For these reasons, the Siles administration turned to seignorage financing, the printing of money. Because of this, inflation soared and reached its peak at 24000% (August 1984- August 1985). Capital flight reached unknown dimensions. Several attempts at stabilization failed, both because of resistance from the Left as well as obstruction by the Right in Parliament. Morales and Sachs therefore conclude, that " the hyperinflation under Siles was not so much a result of an explosion of new spending as the inability to restrain spending _..._. The coalition members were never able to agree on policies to restrain spending, while the government's right wing opponents in the Bolivian Congress rejected all proposals to broaden or even stabilize the tax base" (Morales and Sachs, 1988, 21)

In the beginning of its rule, the UDP coalition satisfied the COB demands for wage increases. When this became impossible and the government resisted the trade union demands, the COB turned bitterly against the UDP. Thus, the last attempts at stabilization failed mostly because of the fierce trade union resistance.

In late 1984, when it became apparent that the UDP government would not be able to solve the economic crisis, Siles Suazo called early elections for July 1985.

2.3 The stabilization

In August 1985 a new government, a coalition of the centre-right MNR and right-wing ADN, was inaugurated under former president Dr. Paz Estenssoro. Within three weeks, the new government presented and implemented its "New Economic Policy" (NEP), designed to stabilize the economy. The most important components of the orthodox NEP were the stabilization of the exchange rate via unification of official and informal exchange rate, the liberalization of trade and capital accounts and the reduction of government deficits, mainly through a reduction in public sector wages and employment and the introduction of a value- added tax. Further characteristics of the NEP were the suppression of the labor movement and the successful attempt to draw capital inflows (such as IMF loans) to Bolivia. (Pastor, 1992, 88). The NEP thus not only aimed for macroeconomic stabilization but also intended to abandon the system of state capitalism in favor of a more liberal market economy.

Begin NEP

The new government's stabilization program successfully ended hyperinflation and brought stabilization to Bolivia. Within a few months, the monthly inflation rate had decreased significantly, as can be seen in the diagram below.

Abbildung in dieser Leseprobe nicht enthalten

Source: James Dunkerley, "Political Transition and economic stabilization: Bolivia"

Jeffrey Sachs, one of the key advisors to the Paz Estenssoro government, and Juan Antonio Morales state that "the key to price stability proved to be a stable unified exchange rate, protected by tight fiscal and monetary policies" (Morales and Sachs, 1988, 32). When the government slightly relaxed its fiscal policy in late 1985, inflation again soared, but was stopped quickly through a resumption of fiscal discipline.

3. Theoretical Background

3.1 Delayed stabilization: Competing models

Bolivia is no exception in the large number of states that protract stabilization measures even though the economic difficulties indicate utmost urgency for reform. How can it be that these states delay reforms against all economic reasoning? This puzzle has engaged several scholars' attention, who have suggested various explanations for the delay in stabilizations. In this section I will give a short overview over some of these models⁵. By applying these models to the Bolivian hyperinflation and stabilization of the 1980s we will see that some models appear to give a good explanation of the Bolivian case while others bear little relevance with respect to this specific case.

3.1.1 Nonpolitical explanations

One line of argumentation is that reforms are delayed out of ignorance. This means that either the decision makers do not know which kind of policy change could yield a positive result, or that they lack the technical expertise and/or supplies necessary to undertake these reforms (Drazen, 2000, 407-8). However, since in Bolivia the Siles government did undertake several attempts at reform, ignorance seems to be no viable explanation for the delay in that case. The same holds for the explanation that reforms are postponed out of the irrational inclination not to tackle the problem.

Another line of argumentation is the model of optimal waiting (e.g. Orphanides (1992)). This means that in some cases, at least in the decision makers' eyes, the problem will solve itself with time, so that no action is necessary. A modification of this model holds that reform will be put off because the external circumstances currently impede such measures but that they will be enacted as soon as the external conditions are more favorable. Again, this model does not seem to provide a valid interpretation of the Bolivian case. The economic crisis in Bolivia neither could be expected to be a temporary phenomenon which would solve itself without any action taken, nor had the external circumstances changed to more favorable conditions when reform was finally implemented.

3.1.2 Vested Interests

This type of explanation (e.g. Olson (1982); Tornell (1998)) takes as a starting point that it is a minority of the population who chooses policies. If it happens to be the case that this minority either profits from the current situation or would be hurt by socially beneficial policies, they will not enact these reform measures. This rationale holds some significance for the Bolivian case, since some individuals, especially from the upper class, made considerable fortunes from hyperinflation and the upper class in general would have been hurt by the higher income and wealth taxes which seemed likely to come in the wake of economic reform. It was therefore not in their interest to change any of the existing policies. This may explain the right-wing opposition in Congress against the stabilization packages introduced by the UDP government. However, it does not explain the fierce trade union resistance against reform.

3.1.3 Uncertainty about individual benefits

The models in this category address the puzzle that groups who would benefit from reform often do not support these reforms ex-ante.

Fernandez and Rodrik (1991) have derived a model in which a reform program is proposed that will benefit a majority of voters. However, it is uncertain who will be among the winners and losers of the reform. The losers cannot be compensated ex post. Since every individual fears to be among the losers, she will vote against reform. Thus a majority ex ante opposes a reform that will benefit a majority ex post. Similarly, a reform that hurts a majority ex post may be adopted by a majority ex ante. Because of its negative effects however, this kind of reforms will subsequently be reversed and thus, in both cases, the status quo will prevail. The drawback of this model is that, like in the Bolivian case, it does not explain why reforms are eventually adopted. It is therefore not sufficient as an explanation of the Bolivian experience.

A dynamic model in this category, formulated by Labán and Sturzenegger (1994), seeks to explain the delay in fiscal adjustment in a highly inflationary environment. Their model assumes an economy with two opposing groups, rich and poor. A stabilization program is proposed, but since there is uncertainty about the net benefits of stabilization, reform is opposed by the poor. Inflation thus continues and is exacerbated by the fact that the rich adapt to the inflationary circumstances and dollarize their assets. As the status quo continues to deteriorate, the costs of reform rise. However, the net benefit of reform relative to the status quo increases over time, so that the poor finally agree to stabilization. The striking result is that "the poor reject an agreement in the first period even though they know that they will suffer an extra period of inflation and then stabilize on even worse terms" (Drazen, 2000, 420).

This model seems to capture many of the features of the Bolivian stabilization process. As assumed in the model, Bolivian society can roughly be classified into two groups, rich and poor. When the economic situation started to deteriorate and the first stabilization packages were introduced, the workers strongly resisted these measures. Therefore, inflation continued and accelerated as capital flight increased. Only when the economic crisis had reached overwhelming dimensions did a majority vote for a change in government and did the workers end their protests. As predicted in the model, they could have achieved more favorable terms of stabilization had they accepted stabilization in the very beginning of the crisis.

3.1.4 Communication failures

This type of model is based on two basic assumptions: First, there is a distinction between the proposer (e.g. the government) and the chooser (e.g. the electorate) of reform. Second, there is asymmetric information between the proposer and the chooser. The delay in reform arises because the proposer does not communicate his reasoning credibly to the policy chooser who therefore does not support reform.

One model in this category is concerned with cheap talk (e.g. Crawford and Sobel (1982)). Cheap talk refers to costless statements made by the policy proposer. Because talk is cheap for the proposer, the policy choosers do not trust those statements and may not believe in the proposer's good intentions. Generally, the more the interests of the proposer and the chooser are correlated, the higher the probability that the proposer actually conveys information in his cheap talk. In these casese the chooser will be more likely to believe his statements.

According to this model, there should not have been a major lack of credibility of the UDP government, because its interests were closely correlated with those of the COB. The model then does not explain why the COB opposed the government's plans at reform despite its belief in the government's good intentions. However, the model might be helpful in explaining the growing trade union opposition to government policy as COB and UDP drifted further apart ideologically with time.

Another model in this category has been devised by Cukierman and Tommasi (1998). In their model, the policy-maker has superior information which allows him to devise a socially optimal policy. However, if he fails to communicate this information to the electorate, the policy-chooser, this policy will not be implemented. Interesting enough, this model yields the result that right-wing policies are more likely to be successfully implemented by a left-wing government and vice versa. However, this was precisely not the case in Bolivia and thus this model does not explain the delay of stabilization in Bolivia. The left-wing government had unsuccessfully tried several times to implement a more right-wing, orthodox stabilization package. Stabilization was only realized when the right-wing government of Paz Estenssoro took office. It could be argued that the failure of the Siles administration lay in its failure to communicate its policy intentions to the electorate. However, this leaves the question open, why the right-wing government was more successful even though they introduced their reform package by decree and without much explanation.

3.1.5 Distributional struggle

Key to this type of models is the notion that reform is a public good whose benefits all members of society will enjoy with the result that no one is willing to pay for it. The resulting struggle about the distribution of the cost of reform can be described as a war of attrition, for which Alesina and Drazen (1991) have presented a formal model. Reformis costly and thus all interest groups tries to wait until any of the other groups gives in and agrees to bear the major share of the burden of stabilization. The decision to obstruct reform despite worsening economic conditions springs from the belief that one's group is stronger than the others and thus able to hold out longer. The reasoning behind that decision is that the cost of waiting is smaller than the cost of reform that the group would have to bear if it agreed to stabilization. Until one group concedes political deadlock is the main feature of the political process. Stabilization occurs only when one group's cost of waiting any longer is equal to the costs of reform which it will have to bear if it concedes. This model seems to explain much of the Bolivian experience, which was characterized by the intense distributional struggle. Two opposing interest groups, the workers and the upper and middle class refused to agree on a reform program. No reforms were undertaken until the economic crisis had reached dimensions which made it unprofitable for the workers to wait any longer. When they gave in, stabilization occurred - at a high cost to the lower class.

3.1.6 Economic Crises

Here the basic idea is that it takes an economic crisis to induce reform (e.g. Nelson (1990),

Williamson (1994)). This notion is reflected in many of the models mentioned above, such as the Labán-Sturzenegger or Alesina-Drazen model. In both models it is the deteriorating economic situation that leads to an agreement between the opposed groups and ultimately to stabilization. The severe economic crisis in Bolivia was certainly one of the reasons that stabilization finally occurred. However, it is not sufficient for us to note that stabilization was preceded by crisis - we need to look at why crisis made stabilization possible. Moreover, the argument that crisis is the trigger of reform is somewhat tautological: What would one be stabilizing if not instability?

3.1.7 Recent Democratization

A very different approach to the question why reforms are delayed is the hypothesis that economic reform is generally more difficult in a young democracy, "in which power is deconcentrated, institutions new, and politicians fearful of public anger and limited in their temporal horizons" (Bunce, 2000, 719). In the Bolivian case it could be argued that reform was not possible until the democratic regime had been firmly established and a new government had taken office through democratic means. Only after this test and subsequent consolidation of the democratic regime, sweeping changes like the stabilization package of 1985 became possible.

One line of argumentation in this respect is that "the key political as well as economic challenge is the routinization of politics and the reduction of the high level of uncertainty associated with the transition" (Haggard and Kaufman, 1992, 340). The idea is that economic reform will not be feasible before the expectations from democracy have stabilized and there has been a downward adjustment of the expectations regarding the distributional capabilities of the new regime. This scenario explains some features of the Bolivian case. When the Siles government took office after a long period of authoritarian rule, expectations were extremely high. Only when this government collapsed and Siles called for new elections, did the

Bolivians realize that a democratic government could not automatically bring prosperity at no price. After their expectations had become more realistic, economic reform was undertaken. Another line of argumentation focuses on constitutional design. In a study comparing four crisis-induced democracies that faced severe problems with economic policy, Haggard and Kaufman (1995, 197) stress the importance of weak institutional characteristics of the party system, which "aggravated partisan conflict and underlying social tensions". Democratization had been accompanied with fragmentation and polarization of the political spectrum. They argue that economic reform became possible only when the incentives, political capabilities and institutional powers of the executive had been altered in order to strengthen the executive and to weaken the influence of the social forces opposed to reform. Bolivia being one of the cases examined in this study, this explanation provides a lot of insight into why reform was delayed for so long in this country. The political debate was highly polarized between the workers on the one side and the private sector on the other, while especially the Left showed a high degree of fragmentation. The failure of the Siles government and the severity of the economic crisis then allowed the Paz Estenssoro government to concentrate power in the executive and push through with stabilization.

3.2 Explaining the Bolivian experience

As we have seen there are many competing models as to why stabilization and economic reform are delayed. While some of these models offer plausible explanations for the proceedings in Bolivia during the first half of the 1980s, others appear to bear little relevance for this specific case.

Three models deserve closer examination since they seem to explain the Bolivian experience very well. These models are the Labán-Sturzenegger model, which attributes the delay to uncertainty about individual benefits, the Alesina-Drazen model, which characterizes delay as a war of attrition and the Haggard-Kaufman model, which ascribes the delay to deficiencies in constitutional design. Since Haggard and Kaufman have already studied the Bolivian case closely, I will not repeat their line of argumentation here. Rather, I will try to explain the Bolivian experience with a model that has not yet been applied to this case.

While the Labán-Sturzenegger model offers a good explanation as to why the poor initially resist reform, it does not focus on the motives of why the rich might also strongly resist any attempt at reform. Since this is one of the characteristics of the Bolivian case and because mention of the intensity of the distributional struggle is predominant in the literature, I will focus on the Alesina-Drazen model which gives an explanation for the obstructive attitude of all interest groups in society.

3.3 Alesina-Drazen's War of Attrition

The Alesina-Drazen model works with the following assumptions: They presume a situation of economic deterioration in which reform is necessary and, more importantly, these reforms will be associated with certain stabilization costs to society. They further assume that society consists of different socioeconomic groups with conflicting distributional objectives. Since reform is a public good and every interest group will enjoy the benefits of stabilization independent of the share of costs of stabilization they bear, all of these groups try to incur the lowest possible costs from reform. The third assumption is that, if the groups enter a fight about the cost allocation, this fight will be associated with fighting costs. Fighting costs are the sum of the costs that arise from further deterioration of the economy and the costs associated with organizing the fight. However, in return the winners of this fight can expect to incur much lower costs of the stabilization process than the loser.

Alesina-Drazen argue that as long as for a group the costs of fighting are smaller than the costs of reform, this group believes that it will be able to force the other group to concede and thus achieve a favorable allocation of reform costs. If all groups are in such a situation, a "war of attrition" is launched which is characterized by the obstructionist attitude of all interest groups involved. This stalemate over the allocation of reform then causes delay. Meanwhile, however, the economy continues to deteriorate. This deterioration in turn raises the fighting costs for the interest groups involved since they suffer additional economic hardship. The war of attrition continues until one of the groups perceives its costs of fighting any instant longer as equal to the burden it will have to bear when it gives in and allows the other groups to decide on the distribution of the costs. At that moment the group concedes and stabilization can occur through reform.

Alesina-Drazen identify three features of stabilization processes which can be characterized as wars of attrition. First, even though there is a general agreement about the need for stabilization, there is disagreement on the distributional question, namely how the costs associated with stabilization are to be allocated. Second, stabilization is frequently preceded by unsuccessful reform attempts. As a third feature of a war of attrition situation, stabilization usually occurs simultaneously with a political consolidation, with one group growing politically dominant. This group then usually succeeds in shifting the larger share of the burden on to the politically weaker groups.

4. Bolivia: An example for a war of attrition?

4.1 Socioeconomic groups

Bolivia has long been an inhomogenous country, divided among others by region, language and ethnicity. The groups that played a predominant role during the economic crisis of the 1980s were socioeconomic groups. The conflict between these groups, which I will call Labour and Capital, dominated the economic crisis of the early 1980s.

By Labour I understand the lower classes, mainly workers. In the first half of the 1980s, their interests were represented by the powerful trade union federation COB (to which about 95% of the Bolivian workforce belonged (Keesings, Volume 29, February 1983, BOLIVIA)). The COB was the association of all Bolivian trade unions. The largest and most influential among them was the Federacíon Sindical de Trabajadores Mineros de Bolivia (FSTMB), the miners' trade union. The labour movement had been crucial in forcing the military's retreat to the barracks in 1982, having put enormous pressure on the military regimes through numerous strikes and demonstrations. Many Bolivians therefore regarded the COB as the truly democratic political actor in Bolivia. The center-left UDP coalition capitalized on the strength and influence of the trade union movement and owed the COB support much of its success in the 1980 elections. However, the UDP reliance on the COB produced a dependence of the UDP government on the trade union and made it very vulnerable to COB demands. Those were particularly strong because the Labour movement viewed the leftist government "as their part of the policy cycle and therefore sought to make up for the previous decline" in real wages (Pastor, 1992, 80)⁶. Among the most frequent demands on the trade union agenda were wage increases and job security. Especially the FSTMB called for co-management that would allow workers to share control of their companies. The fight against inflation did not have priority on the COB agenda and only became an issue when the repercussions of inflation started to seriously affect the workers' real income.

The other important socioeconomic group in the Bolivian crisis of the 1980s, Capital, consists mainly of the upper and middle class. This group had begun to develop a class-feeling and political weight during the Banzer era in the 1970s (Malloy and Gamarra, 1988, 162). It mainly consisted of small and medium entrepreneurs who had privately invested in the Bolivian economy and was politically represented by the political parties on the right. Those were the center-right MNR, headed by Paz Estenssoro and the Accíon Democrática Nacionalista (ADN) of former dictator Gen. Hugo Banzer. At the same time, the dominant interest group of the private sector, the confederation of Bolivian Private Entrepreneurs (CEPB) had continuosly increased its influence and effectively voiced the concerns of the Bolivian private sector. The private sector favored a more open and market-centered economy. Their immediate concern, however, was the protection of their profits and they therefore resisted any form of tax increase that would decrease their revenues.

Even though the private sector also suffered from the economic crisis, they did far less so than the workers. Inflation, for example, is a highly regressive tax. Moreover, it is worth to mention that some rich individuals made fabulous fortunes by way of privileged access to cheap (that is converted by the official exchange rate) dollars from the Banco Central or by dollar savings which they then sold off on the informal markets.

Both Labour and Capital were powerful veto players. The trade union movement drew its strength from the UDP dependence on its electorate, while Capital's influence came mostly from the majority which `their' parties enjoyed in parliament.

4.2 The features of a war of attrition?

Alesina-Drazen name three features which must be fulfilled in order to characterize a situation as a war of attrition. Below I will examine whether these features can be discerned in the Bolivian economic crisis and stabilization of the 1980s.

4.2.1 Yes to stabilization - but how?

The first feature of a war of attrition is that "there is agreement on the need for a fiscal change but a political stalemate over how the burden of higher taxes or expenditure cuts should be allocated. In the political debate over stabilization, this distributional question is central" (Alesina and Drazen, 1991, 79).

When the UDP government took office in late 1982, there existed wide consensus in the Bolivian society that some action had to be taken against the economic crisis. However, there was equally widespread dissent on how stabilization was to be achieved.

Labour, represented by the COB, held three major convictions: First, they demanded redistributionist policies favoring the popular sectors, such as wage increases, growing social expenditures and higher public employment. Second, they pressed for "structural and legal changes that would increase the autonomous corporate power of the COB within the state" (Malloy and Gamarra, 1988, 160). Third, they called for more state intervention in the economy with the long-term goal of a socialist state. Under no circumstances were they willing to agree to policies where the lion's share of social costs would have to be borne by the lower classes. In particular, they resisted any kind of IMF-style stabilization plan.

The private sector, on the other hand, pushed for an orthodox stabilization through a strict austerity program. In their opinion, stabilization should furthermore be accompanied by liberalization and a more market-oriented policy, in short a "a roll back of the state's role in the economy in favor of the private sector" (Malloy and Gamarra, 1988, 153). At the same time, they sharply resisted any attempt to consolidate the budget through income and wealth taxes.

Obviously, the positions of Labour and Capital were irreconcilable. They manifested themselves in many debates and fights about minor and major policy initiatives of the government, such as the numerous wage increases (which Labour hailed and Capital opposed) or the elimination of government subsidies on fuel and certain basic foodstuffs in November 1983, against which Labour staged a general strike but which found the support of the private sector.

4.2.2 Failed attempts at reform precede successful stabilization

The second characteristic of a war of attrition condition is that "successful stabilizations are usually preceded by several failed attempts. Often a previous program appears to be similar to the successful one" (Alesina and Drazen, 1991, 82)

President Siles politics have been called "politics of economic packages" (Malloy and Gamarra, 1988, 164). The UDP government was acutely aware of the worsening economic crisis and therefore tried to reverse the negative trend through their economic policy. However, as we have seen above, their actions were constrained by the enormous influence of the trade union movement and the lack of a majority in Parliament. This explains the "compromise" character of most of the UDP reform packages.

Abbildung in dieser Leseprobe nicht enthalten

Sources: Keesings; Dunkerly

As can be seen in the table, especially the economic package of April 1984 is characterized by its orthodoxy and resembles in parts the successful package introduced by the Paz Estenssoro government upon its accession to power in 1985. It included among others a large devaluation of the peso and promised a tax reform. However, as usual, fierce Labour protests forced the government to water down the reforms.

4.2.3 Political consolidation

According to Alesina and Drazen, the final feature of a war of attrition situation is political consolidation: "When stabilization occurs, it coincides with political consolidation. Often, one side becomes dominant." (Alesina and Drazen, 1991, 81) They go on: "The burden of stabilization is sometimes quite unequal, with the politically weaker groups bearing a larger burden. Often this means the lower classes with the burden of a successful stabilization being regressive".

In Bolivia, political consolidation occurred in July 1985, when the elections produced a clear majority (55%) for the rightist parties⁷. The overwhelming support for the Center-Right can be explained by the dismal economic situation and a political Left which had proven incapable of solving the crisis. Ibáñez Rojo contends that the failure of the Left had taken Bolivia to a "Hobbsian scenario of all against all, in which any protagonist capable of offering a minimal collective good (economic stabilization) was bound to enjoy considerable democratic legitimacy" (Ibáñez Rojo, 2000, 176).

The Center-Right (as a representative of Capital) used the power they had received through the election result and launched their extensive economic reform program only three weeks after taking office. This program was much more orthodox and far-reaching than any of the packages introduced by the Siles government had been. As was to be expected, the COB passionately opposed these reforms and declared an indefinite general strike until the reforms package be withdrawn. This time, however, the government reacted rigorously, declared a state of siege and arrested 1500 hunger striking trade unionists. The COB finally called the strike off. Labour's back was broken I argue that this is the moment, when Labour conceded in the war of attrition. Alesina and Drazen contend that the interest group which concedes in the war of attrition will have to pay the larger share of the costs of stabilization. Was this the case in Bolivia as well?

It is clear, that both groups benefited from the end to hyperinflation. Labor probably benefited from it to a higher degree, because inflation had hit the workers harder than the middle class. However, the workers nonetheless paid a high price for stabilization.

While the unification of the exchange rate had no direct negative impact on the lower classes, the tight fiscal and monetary policies certainly did. One of the basic objectives of the stabilization plan was the consolidation of the budget. This was to be achieved by a reduction of state expenditures and simultaneously by an increase in state revenues.

The reduction in state expenditures was mainly achieved through a reduced public sector wage bill. The government froze public sector salaries, spending and investment. This meant a fall in real income for all public sector employees, mostly workers, in the state enterprises. Additionally, the government reduced public sector employment significantly. Employment of Bolivia's largest enterprise, the state-owned mining company COMIBOL, fell by 77%, so that approximately 23000 workers lost their jobs⁸ (World Bank, 1992, 4). Unemployment, which had stood at 15,4% in 1985, rose to 20% in 1986 and then stagnated on this high level (Statistisches Bundesamt, 1991, 41). These policies clearly affected Labor significantly stronger than the middle and upper classes.

On the other hand, the Paz Estenssoro government sought to raise new revenues. The first move in this direction was an increase of the price of gasoline from 4 cents per liter in August 1985 to 28 cents per liter in September 1985 (Sachs, 1986, 20). Furthermore a new tax system was introduced which was mainly based on the introduction of a regressive value-added tax. Again, the lower classes were hit much harder by these measures than Capital.

Additionally, the new government liberalized the labor market and generally repressed the Labor movement by way of the "antiworker and antipopular agenda of the new government" (Pastor, 1992, 89).

However, the liberalization of trade and capital accounts, which included the elimination of import prohibitions and the adoption of a low uniform tariff, hit the private sector hard. Even though the CEPB had been demanding a more market-oriented economy in the past, many entrepreneurs felt threatened by the competition that they suddenly faced. They benefited from these policies only indirectly, in that they projected a neo-liberal image of the new Bolivian government and hence helped to draw foreign capital to Bolivia.

Consequently, it is not true, that the private sector benefited from all stabilization measures while Labor had to bear all the stabilization costs. It is true, however, that Labor bore by far the largest part of these costs and was hit much harder by the reform program than the middle and upper classes. The Bolivian case therefore fulfills the proposition brought forward by Alesina and Drazen, that the politically weaker group will have to bear the larger burden of stabilization, when it finally occurs after political consolidation has taken place.

4.3 The timing of reform in Bolivia

Alesina and Drazen argue that stabilization occurs at the moment in which one group concedes in the war of attrition and agrees to bear the major share of the burden. This happens when for that group the costs of remaining of the fight are just equal to the costs of conceding (the share of stabilization costs which they will have to bear).

In Bolivia, the end of the war of attrition, which finally led to successful stabilization, came on October 3, 1985, when the COB called off its general strike after the fruitless attempt to force the new government to abandon its orthodox reform package. At that point in time, the costs to Labor of remaining in the fight were mounting, while the probability of a successful outcome was quickly declining.

The costs of the war of attrition for Labor were threefold: First, hyperinflation continuously caused an erosion of the workers' real income, with inflation running faster and at a higher rate than any wage increases that Labor could achieve. As we have explored above, the economic crisis generally hit the workers harder than the middle and upper classes. Second, the Labor movement incurred the costs of mobilizing the masses for the fights in the war of attrition. The strikes, demonstrations and other actions required not only a lot of logistics but was always accompanied by the insecurity about what the government would do, and if they would react with repression. Third, the deteriorating situation caused infighting and fragmentation of the Labor movement, especially the COB⁹. Thus the fighting costs are not only the costs associated with the deteriorating state of the economy but also the costs of the fight itself. As we have seen, an important variable in this respect, which Alesina and Drazen do not explicitly mention, is political repression.

Naturally, the expected gains from a success in the war of attrition were substantial: Labor sought a rather socialist reform, which shifted the major burden of stabilization costs on to Capital. Considering the reform plans presented by the Siles government and especially the orthodox reform package introduced by Paz Estenssoro, Labor also had a lot to lose.

However, expected gains have two components: the gains themselves are multiplied with the probability that they can be achieved. The probability, though, that further fighting against reforms would result in a reform plan suited to Labor's interests, had started to decrease with Bolivia's mounting economic difficulties and had received a major setback in the 1985 elections, when a majority of voters had opted for the center-right parties. When the COB staged a general strike in response to Paz Estenssoro's New Economic Policy, he reacted sharply and enacted a state of siege - his determination to resist any pressure from the Left further diminished Labor's chances of success. When the new government arrested a great number of protesting workers and sent the prominent activists into internal exile, the Labor movement realized that the probability of success had almost vanished. At that point, the expected gains were equal to the costs of remaining in the fight, and it was clear to the leaders in the COB that the costs would soon surpass the expected gains. This is why, at that moment in October 1985, Labor conceded in the war of attrition and consequently had to bear the major share of the orthodox stabilization program.

5. Conclusion

This paper has given an overview over existing models and theories that have been devised in order to explain the phenomenon that economic reform is oftentimes delayed despite an urgent need for immediate action. Bolivia is an extraordinary case in this category because reform was delayed for such a long time that a full-grown hyperinflation emerged. In order to explain this puzzle, I have focused on a model presented by Alesina and Drazen (1991) which characterize delay as a war of attrition between several socioeconomic groups. I come to the conclusion that this model offers a good explanation of the Bolivian experience.

However, it should be noted that some incidents are not explained by the model: First, the fact that force was applied to force the Labour movement to concede and second, the puzzle that far-reaching liberalization measures were enacted even though they hurt large parts of the upper and middle class. For the first phenomenon, the forceful repression of the Labour movement, I suggest that this be explicitely incorporated into the model as additional fighting costs. The second phenomenon proves harder to incorporate. It could be argued that liberalization was good for the exporting industry and therefore did benefit those entrepreneurs. However, since in Bolivia the majority of exports are comprised of mining and petroleum products, which were produced by state-owned firms, there were only few private entrepreneurs who profited from liberalization measures. The majority of Bolivian industry is import competing industry. Another reason for liberalization could have been the intention to signal the international arena Bolivia's willingness to comply with orthodox policy prescriptions and thus draw foreign loans to the country. This question certainly calls for further research.

Having shown that the Alesina-Drazen model gives us useful insights into the process leading up to stabilization in Bolivia, the next step should be to devise policy instruments in order to prevent wars of attrition to occur in the future. The purpose of understanding the mechanisms of delay should be used to counteract obstructionist tendencies in future economic crises.

References

Alesina, Alberto and Allan Drazen (1991) `Why are stabilizations delayed?', American Economic Review, Vol.81 (5): 1170-88

Crawford, V. and J. Sobel (1982) `Strategic Information Transmission', Econometrica, Vol. 50, 1431-51

Cukiermann, A. and M. Tommasi (1998), `When Does It Take Nixon to Go to China', American Economic Review, Vol. 88, 180-97

Drazen, Allan (2000) Political Economy in Macroeconomics, Princeton: Princeton University Press

Dunkerley, James (1990) Political transition and economic stabilisation: Bolivia 1982-1989, London: Institute of Latin American Studies University of London

Fernandez, R. and Dani Rodrik (1991), `Resistance to Reform: Status Quo Bias in the

Presence of Individual Specific Uncertainty", American Economic Review, Vol 81., 1146-55

Gamarra, Eduardo and James M. Malloy (1988) Revolution and Reaction. New Brunswick, Transaction Inc.

García, José Z. `Peru and Bolivia', in Jan Knippers Black (ed.) Latin America: It's Problems and Its Promise, pp. 448-470, Boulder: Westview Press

Haggard, Stephan and Robert R. Kaufman (1992) The Politics of Economic Adjustment,

Princeton: Princeton University Press

Haggard, Stephan and Robert R. Kaufman (1995) The Political Economy of Democratic Transitions, Princeton: Princeton University Press

Keesings Record of World Events, http://www.keesings.com

Labán, R. and F. Sturzenegger (1994) `Distributional Conflict, Financial Adaptation and Delayed Stabilization', Economics and Politics, Vol. 6, 257-76

Morales, Juan Antonio and Jeffrey D. Sachs (1988) `Bolivia's Economic Crisis', Working Paper No. 2620, Boston: National Bureau of Economic Research

Morales, Juan Antonio Morales and Jeffrey D. Sachs (1990) `Bolivia's Economic Crisis', in Jeffrey D. Sachs (ed.) Developing Country Debt and Economic Performance, Vol. 2 Country Studies Argentina, Bolivia, Brazil, Mexico, pp. 159-268, Chicago: University of Chicago Press

Nelson, J., ed. (1990) Economic Crisis and Policy Choice: The Politics Of Adjustment in the Third World', Princeton: Princeton University Press

Olson, Mancur (1982) The Rise and Decline of Nations, New Haven: Yale University Press

Orphanides, A. (1992) `The Timing of Stabilizations', Finance and Economics Discussion Series No. 194, Washington DC: Federal Reserve Board

Pastor, Manuel Jr. (1992) Inflation, Stabilization and Debt: Macroeconomic Experiments in Peru and Bolivia, Boulder: Westview Press

Rojo, Enrique Ibáñez (2000) `The UDP Government and the Crisis of the Bolivian Left (1982-1985)', Journal of Latin American Studies, Vol. 32 (1): 175-205

Sachs, Jeffrey (1986) `The Bolivian Hyperinflation and Stabilization', Working Paper No. 2073, Boston: National Bureau of Economic Research

Schacter, Mark, Margaret Grosh and Steen Jorgensen (1992) Bolivia's Answer to Poverty,

Economic Crisis and Adjustment: The Emergency Social Fund. Washington D.C.: The World Bank

Statistisches Bundesamt (1991 Länderbericht Bolivien 1991, Wiesbaden: Statistisches Bundesamt

Statistisches Bundesamt (1986) Länderbericht Bolivien 1986, Wiesbaden: Statistisches Bundesamt

Tornell, A. (1998) `Reform from Within', Working Paper No. 6497, Boston: National Bureau of Economic Research

Williamson, J., ed. The Political Economy of Economic Reform, Washington DC: Institute for International Economics

[...]

---

^[1] In those elections, the UDP polled 35,9%, the MNR 35,8% and the ADN 15,3% (Malloy/Gamarra, 1988, 131)

^[2] In the last 30 years another branch has reached importance: The illegal coca industry (Coca production in 1984 was estimated to be 152 000 tons (Statistisches Bundesamt, 1991, 45)). This poses a problem to the Bolivian state in so far, as coca revenues do not raise revenues in taxes for the Bolivian state and as the USA put enormous pressure on the Bolivian government to fight against the coca trade.

^[3] Article 90 of the 1967 Constitution

^[4] The bulk of tax revenues came from the state enterprises COMIBOL and YPFB

^[5] For the economic models I will largely follow the categorization presented by Allan Drazen in chapter ,,Inaction, Delay, and Crisis" of his book ,,Political Economy of Macroeconomics"

^[6] For a detailed description of the relationship between UDP and COB see Enrique Ibáñez Rojo, ,,The UDP Government and the Crisis of the Bolivian Left (1982-1985)"

^[7] The right-wing ADN recieved 28.6% of the vote, the center-right MNR 26.4%

^[8] The fall in employment is a result of the government's policy as well as of the collapse of tin prices in those years.

^[9] For a detailed description see Enrique Ibáñez Rojo (2000) ,,The UDP Government and the Crisis of the Bolivian Left (1982-1985)"