SciELO - Scientific Electronic Library Online

 
vol.37 número112LA ECONOMIA POLITICA DE LAS REFORMAS ECONOMICAS EN LA ARGENTINA índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Cuadernos de economía

versión On-line ISSN 0717-6821

Cuad. econ. v.37 n.112 Santiago dic. 2000

http://dx.doi.org/10.4067/S0717-68212000011200005 

THE POLITICAL ECONOMY OF THE CHILEAN TRANSITION
TO DEMOCRACY*

CLAUDIO SAPELLI**

ABSTRACT

Chile's economic performance since the mid 1990s has been outstanding. Several reasons account for this success. The strong initial position of the economy was a crucial asset. Broad consensus on economic matters helped a lot. This paper argues that the institutional setup was key to this success.

Of course, there are many reasons why events unfolded as they did (the people, the ideas at the time, the external conditions, the internal economic situation, the institutional setup, etc.). We study in the paper how changes in the institutional setup may have constrained behavior during the transition, and also discuss how it may help (or hinder) the preservation of such economic performance in the future. For this we use the models from political economics and analyze how the institutional constraint contributed to the transition.

The paper concludes that the institutional setup was important. Policy making involves the interplay of what might be termed big and small institutions. Running through much of the literature on economic reform is the recognition that it is not enough to get certain policies right; it is also important to get institutions right. The technocratic approach to this problem is the creation of independent agencies: independent central banks, autonomous regulatory agencies, independent judiciaries, and so forth. The Chilean process is remarkable in its concern to appropriately design both big and small institutions so to provide an adequate framework for decision making by the private sector.

Key words: Political economics, credibility, time inconsistency, institutions.

JEL Classification: H0, P0, O5.

1. INTRODUCTION

The Chilean transition to democracy has been praised and criticized broadly, both within and outside Chile. Politicians, leaders of social organizations, and intellectuals alike have provided strong, and sometimes passionate, arguments on the shortcomings or political wisdom of the transition process.

Critics of the Chilean transition are, on one hand, critical of what they consider the perpetuation of military authority in critical areas, the inability to solve the cases of human rights violations and punish those responsible, and the lukewarm approach with which the Concertación government has approached political reforms. On the other side of the ideological spectrum, they question the government's commitment to pursue further structural reforms in the economic realm and its emphasis in what they judge unimportant political reforms.

The defenders of the Chilean transition underline its ability to abridge the deep sense of mistrust and polarization that had characterized Chilean society since the sixties. Additionally, they highlight the government's ability to maintain economic stability and implement, in that context, an ambitious program of investment in the social sectors.

Whatever view one decides to follow, all would agree in that Chile constitutes an elaborate and complex transition from authoritarian to democratic rule. The elements that allowed it to successfully come to terms combine a broad set of variables including institutional, cultural, and even personal factors. These were able to keep in balance economic, political and social expectations which, at least initially, seemed to many actors as impossible to combine into a positive sum game.

In that sense, the main task of the transition was not just to move from a military government to an elected one. It had to be done in a manner that would provide adequate signals to all relevant actors. To govern the country effectively required balancing two conflicting expectations: that the government would change everything, altering substantially the framework for economic and political development, or that the government would not change enough, continuing with the path drawn by the military government.

To achieve this the Aylwin government had to produce quick results: acting in a way that incorporated a strong sense of political inclusiveness and participation; and at the same time demonstrate effectiveness in managing the economy. This had to take place within a clearly defined and rigid institutional setting, provided by the 1980 Constitution and other legal bodies implemented by the outgoing military government. Hence, the room for maneuver was tight and the time was short. How did the new government face these challenges? This paper centers on the formulation of economic policy, on questions such as, "What factors influenced the continuation of economic policy?" and "How were populist pressures controlled?" In response to these questions, the paper underlines the key role of institutions.

2. THE POLITICAL ECONOMICS OF THE TRANSITION TO DEMOCRACY

We study here how changes in the institutional setup may have affected behavior in the transition, and also how it may help to preserve a good economic performance in the future. Our main hypothesis is that constraints that took the form of new institutional constraints contributed to the transition.

The organization of this section is as follows: we first summarize the relevant literature which is on both sides of the boundary between economics and political science. We give particular attention to the normative aspects of the literature in the areas of political economics, constitutional economics and public choice. Then we look at the institutional changes that have occurred in Chile and concentrate on a few that may have contributed importantly to the way the transition took place regarding economic policy.

In the realm of economic policy, an institutional framework is important if it provides the economic legal framework with lasting power and hence with the required credibility for people to act (invest, etc.) on the basis of their permanence. Institutions are relevant only if they impose constrains to the behavior of actors. They should diminish: i) the discretion of the government in the administration and implementation of policy objectives; and/or ii) the ability of the government to undertake unilateral changes in policies. These conditions are similar to those stated by Levy and Spiller (1994) as the mechanisms that provide credibility and effectiveness to a generic regulatory framework. Hence the main purpose of these mechanisms should be to restrain arbitrary administrative action. As Levy and Spiller point out, to chose this route is to chose policy credibility over policy flexibility (this replicates the choice at the level of macro policy between rules and discretion). As is well known in the discussion of the literature, there are different opinions as to the weights to assign to both objectives. Those that have utmost confidence in public institutions prefer flexibility since they expect these institutions to pursue the population's welfare at every point in time. Hence they find these restraints unnecessary, sub optimal, and possibly also arbitrary and undemocratic.

However, the politician at the center of the political economics literature is an opportunist that maximizes his own welfare function (a "devil"), a far cry from the "saint" who maximizes the social welfare function that is at the center of much institutional analysis. The issue at hand is not to discuss whether politicians are usually saints or devils. What is important is what institutions are optimal given the existence of both. If we accept there are both saints and devils, then the key issue is that the saints actually benefit from institutions that limit the damage devils can make if elected, even if those institutions also limit saints when they are in office.

Hence the type of institutional reforms that are important for our purposes are those that limit government discretion and make drastic shifts in policies difficult, thus providing a more stable and credible investment environment. The issue we will discuss is, then, in which way and to what extent the structure and organization of institutions impose constraints upon governmental action in Chile.

POLITICAL ECONOMICS: THE THEORETICAL LITERATURE

The key question is what policymaking institutions produce better policy outcomes. According to Persson and Tabellini (1998), "..(t)he general approach of the political economics literature is to explain deviations in observed economic policies from a hypothetical social optimum by appealing to specific incentive constrains in the decision problem of optimizing policy makers. The positive analysis focuses on identifying the relevant incentive constrains, while the normative analysis focuses on institutional reforms which may relax them."

One important issue is how to solve the lack of credibility created by incentives to be time inconsistent (i.e. promise something and then renege on that promise). This problem is due to the awareness of these incentives to be time inconsistent, and the lack of credibility of promises that behavior will not be influenced these incentives. Institutions can solve this credibility problem1. The problem here is that desirable policies may suffer from lack of credibility when policy decisions are taken sequentially over time (under discretion) and the government lacks a non distorting policy instrument. Lack of credibility may lead to economic policies or results that are sub optimal (e.g. equilibrium average inflation or wealth taxation may be too high). This "lack of credibility" problem has implications for institutional design. In particular, it makes it desirable to restrict the discretion of the policymaker. This insight has been exploited in monetary policy to advocate the benefit of an independent and "conservative" central bank. It also explains why it may be convenient for a Central Bank to issue debt in foreign currency (or in the case of Chile, in UFs). In sum, institutions that tie the hands of politicians ex ante may result in better policy and better outcomes.

International competition is another institutional device for coping with credibility, because it can reduce the ex post incentives to unilaterally increase tax rates or inflation. Countries that are more open tend not to adopt sub optimal policies because negative effects are widely enlarged by open trade and capital accounts. Inadequate policies would imply that exports can compete less successfully abroad and imports can compete more successfully locally, and/or investors just go somewhere else. See Sapelli (1992) for an analysis of how openness of the economy constrains the growth of the state's influence in the economy, in the case of Uruguay.

One can also argue that institutions that provide these constrains, moderate political conflict and policy extremism and hence result in better outcomes. The literature in this respect has some normative implications that are useful to our examination of the Chilean case.

The premise is that institutions "matter". They matter in the ways they constrain (or do not constrain) behavior. The Constitution and other institutions must lay down some fundamental aspects of the rules of the game, and these must not be easily changed. The incentives are not only provided by the rule itself but also in how easy it is to change it. For example, once an independent central bank has been set up, its effect on credibility depends on how costly it is to change the institution ex post2. However, the staying power of institutions need not be very long to be effective. In the model that dominates the literature on credibility issues in monetary policy, for example, what is needed is a high cost of changing the institution within the time horizon of existing nominal contracts. On the other hand, this implies that legal changes that may appear harmless could have important long term effects. For example, if costs of changing institutions are lowered, then there will be effects as the population realizes there has been an increase in the temptation to renege on the institution. Merely lowering the costs may generate a loss of credibility and lead to a loss in welfare. The real question is what costs are sufficiently high to make the institution credible. For institutions to be important all we need is that one can assume that an institution with "enough" costs can be designed3.

In the case of the design of monetary institutions to tackle the distortions created by credibility problems and political cycles4, the stylized facts show the importance of a good design. Cross country data for industrial countries show a strong negative correlation between measures of central bank independence and inflation, but no correlation between output or employment volatility and central bank independence. Thus, central bank independence seems to be a free lunch: it reduces average inflation at no real cost.

In the case of fiscal policy there are again a number of stylized facts that give credence to the assumption that institutions matter. In this case large deficits and debts have been more common in countries with (see Persson and Tabellini 1998) :

i) proportional rather than majoritarian electoral systems;
ii) coalition governments and frequent government turnover; and
iii) lenient rather than stringent government budget processes.

The correlations between policies and political institutions suggest that political and institutional factors play an important role in shaping fiscal policy. Although the literature concentrates in analyzing credibility problems in the taxation of capital, the credibility problems are not only found here. These problems should be considered the norm rather than the exception when policy is analyzed in a dynamic context.

In the case of capital taxation the dynamic inconsistency is very clear. Since the ex post optimal tax structure differs dramatically form the ex ante optimal structure (as ex ante and ex post elasticities differ substantially), it is optimal for the government to renege on its announced tax structure and ex post dramatically increase taxes on capital5. If the government is unable to commit, the economy is trapped in a third best, or worse, allocation6. The resulting policy uncertainty is another consequence of the discretionary policy environment.

Although presently there are, in several countries, institutions designed to raise the credibility of monetary policy, in the case of fiscal policy, such reforms are less effective. This is so in part because the tasks of a sovereign legislature cannot be easily constrained. Nevertheless, some institutional devices could mitigate the problem. As discussed in Persson and Tabellini (1995) capital controls or international tax agreements that limit tax competition exacerbate credibility problems and could thus be counter productive.

With regard to debt policy, the theoretical models used in the literature7 yield the empirical prediction that political polarization and political instability lead to larger debt accumulation8. Political instability causes government to behave myopically. The result can be applied to any intertemporal aspect of public policy: the level of public investment, the implementation of tax reforms, etc. Which institutional reforms would address this problem? A natural idea is to reform the electoral system to make majority governments rather than coalition government, more likely. The centralization of decision making power could also solve the problem. Although it could be abused, the problem could be mitigated by checks and balances.

CHILE: ECONOMIC POLICY DURING THE TRANSITION

It is widely recognized that the democratically elected government inherited a very successful economy in 1990. This stands as a major contrast to other experiences of transition in the region. It is very important since economic growth can reduce frustrations derived from distributional inequalities by increasing mobility and provide the material base for social compromises. A long record of successful economic performance can strengthen general beliefs in the effectiveness and legitimacy of the system as a whole.

However, legitimization of the free market economic model was crucial. The political actors where conscious of this issue. To make the permanence of the market economy viable, the government had to gain the support for policy continuity. Moreover, this had to be done quickly. In this respect, an important political decision was to address during the first year of the administration two critical economic reforms: a package aimed at funding new social programs (for which there was much pressure and also the fear that satisfying these pressure would imply losing fiscal control), and a reform of the labor law (a law that had been much criticized by union leaders and the possibility of its reform was generating much uncertainty). Also, government officials explained that these two pieces of legislation were the only important changes to the economic model of the military government. It is not clear whether the latter was a credible commitment, but the awareness of the importance of institutional stability is notable, and by tackling these issues early, the government tried to minimize possible negative effects on private investment associated to policy uncertainty.

This was deliberate, as said by Edgardo Boeninger, Minister of the Presidency under Aylwin, (Conference at the World Bank, April 25, 1991): "The administration first undertook the task of implementing reforms producing either uncertainty or cost increases in the economy (tax reforms and changes in labor legislation). These have now been mostly completed, enabling the government to guarantee full stability of the rules of the economic game for the rest of its term, thus facilitating dynamic behavior by business."

There were also fears that import tariffs would be increased selectively; and that an increase in fiscal deficits and wages would lead to a surge in inflation, and capital flight. These areas of conflict were also resolved. Possibly the most notable signal of commitment to the free trade policy was a cut in import tariffs (when fears where of an increase, this had notable effects on expectations).

The new administration avoided the conflict between macroeconomic stability and increased expenditure in the social sectors by changing the composition of government spending toward social spending and implementing a tax. National agreement between the government, the CUT (Central Unitaria de Trabajadores, the main trade union) and the CPC (Confederacion de la Produccion y el Comercio, the main employers association), allowed for a 21.9 percent rise in the real minimum wage (March 90-92)9. It was important to be cautious in not implementing an unsustainable wage increase. The experience of other countries in Latin America helped Chile. Aggressive wage adjustment above productivity gains proved short lived in Argentina, Brazil and Peru in the 80s. After a few years wages were substantially below the starting point in all these countries. This is what Dornbush and Edwards (1991) have called the populist cycle. It is clearly in the interest of workers and trade unions to prevent such a pattern. As the agreement was implemented and economic growth and external conditions continued to be favorable, real wages increased on average by 3.7 % annually during President Aylwin's administration and the minimum wage by more than double that. At the end of 1993 unemployment was below 5%, almost 2 percentage points lower than in 1989. And public social spending rose by almost 50%. Favorable external conditions helped to satisfy most of the requirements in terms of wages and social spending while not affecting growth, inflation or the BOP.

As was mentioned above, an early and significant decision was slashing import tariffs by one third in June 1991 (from 15% to 11%). There was clearly no better way to signal the government's commitment to the goal of an open economy than deepening the process of opening the economy. However, the benefits of this strategy were in some ways jeopardized by the strategy of bilateral agreements which increased effective protection for many sectors and abandoned the worthy goal of giving all sectors the same degree of effective protection. This has political economy consequences, since once one accepts that tariffs are not uniform, lobbying for differential tariffs becomes legitimate.

The commitment to an open economy was probably helped by the fact that, after twenty years of a very open trade regime, the constituency in favor of low tariffs is large, as the export sector grew from 15% of GDP in 1965 to 35% in 1990-94.

Monetary policy

It was not enough to have an independent Central Bank. The authorities thought that the Bank needed to give a signal of its independence and its commitment to low inflation. The new authorities inherited a complicated situation regarding monetary policy. Chile had been enjoying high rates of GDP expansion in 1988 and 1989 (about 9 percent per year on average), accompanied by what many considered an unsustainable rate of expansion of private money (about 40 percent per year) and domestic expenditure (11 percent per year). With money growing at 40% and GDP at almost 10% it was not surprising inflation was increasing towards 30%. Nonetheless, it is important to note that the economy was slowing since the third quarter of 1989 and hence it is not a clear cut issue whether a tightening of monetary policy was necessary. This inaugurated a phase of procyclical policy changes by the Central Bank authorities that have recurred during the 1990s. However, what the Central Bank did at that point appears more uncontroversial than the policy it has followed since then. This was the first test of the new authorities of the newly independent Central Bank. The Central Bank had a reputation to establish, and could not risk making a mistake on the side of softness. It wanted to signal it would be committed to keeping inflation low and hence had to be tough, so tightening of monetary policy was in order.

In early 1990 the newly independent Central Bank engineered a sharp monetary contraction aimed at reducing aggregate expenditure. Ex ante real interest rates on ten year Central Bank bonds were risen form less than 7 percent to 9.7 %. The Central Bank accumulated 2,4 billion dollars in reserves in 1990, and sterilized the effects by selling more bonds. The adjustment was costly in terms of GDP (GDP growth dropped to 3,3% in 1990) but served the monetary authority as genuine proof that it was willing to pay a cost to fight what they considered an overheated economy and thus to preserve macroeconomic equilibriums.

Fiscal policy

Another traditional conflict was over the allocation of expenditures and taxes. But a sharply increased awareness of the costs of budget deficits and money creation contributed to moderating the spending demands10. The budget rules (see below) also have helped keep the process running smoothly.

Fiscal revenue increased significantly during 1990-93 due to: the tax reform of 1990, a strong expansion of economic activity and imports, a relatively high price for copper and a reduction in tax evasion. This, together with low international interest rates made it possible to increase social spending and have a budget surplus of between 1.5 and 2.5% of GDP every year. However, there was a quasi fiscal deficit due to the Central Bank's losses in current open market operations and inherited losses form the rescue operation of banks in the early eighties. The fist Concertación government contributed to a reduction in this quasi fiscal deficit by making extraordinary prepayment to the Central Bank. This allowed the government to support the public sector surplus in a politically viable way, by increasing expenditure through debt prepayments. This was important because it would have been politically costly for the government to show a high non financial public sector surplus, given the pressures for further expenditure.

Labor legislation reform

The government had two major goals with this reform:

1) to balance negotiating power, which the authorities thought was biased in favor of employers, and

2) to legitimize the labor legislation that was largely opposed by trade unions

The reform finally approved by Congress was an intermediate position between the old law and the original project of the executive, as the government had to negotiate with the main opposition party. In this respect, institutional reforms that gave the opposition more congressional representation than would have been warranted by a proportional representation rule (see below) appear important in increasing the difficulty of reforming institutions and contributing to stability in the rules of the game.

The labor reform eliminated the right of employers to fire workers with no expression of cause but allowed a general cause of "needs of the company".

Severance payments remained at one month per year of work but the maximum was increased from five to eleven months (there was no maximum in the original reform project) and workers could appeal in court. If the courts found the dismissal unjustified then severance payment was increased by 20%.

The other reforms applied to collective bargaining and unionization. Groups of non-unionized workers were allowed to participate in collective bargaining. Negotiations were allowed at the industry level, but only if there was an agreement between employers and employees. Another important change included allowing unlimited strikes, eliminating the 60 day maximum.

THE INSTITUTIONAL FOUNDATIONS

We have not discussed the institutional framework that may have contributed to the broad continuation of economic policy. This is what we will analyze now. Regarding institutional reforms, the state inherited by the democratic government was very different from the clientelistic state that existed in the 60s and early 70s. It was a more autonomous state, in which technocratic considerations have greater policy weight. As Velasco (1994) says: "The demise of the clientelistic state is expressed in the institutional arrangements that emerged: that include the uniform tariff which limits the scope of private sector lobbying for import protection; the technification of the state's activities as regulator; the privatization of social security and consolidation of changes in budgeting procedures, both of which limit the amount of lobbying on fiscal matters; and the independence of the Central Bank that both ends the partial control that private sector interests had on credit policies and also limits severely the ability of the bank to finance fiscal deficits. These institutional reforms changes the nature and policymaking capacity of the Chilean state and therefore the way resources are allocated in the Chilean economy."

This institutional setup, to have consequences, needs to be perceived as difficult to reform. It is credible only if these measures can't be reversed with the stroke of a pen. In some cases (as in the prohibition against non tariff barriers) the new legislation took on a quasi constitutional status that created large legal and practical costs associated with attempted reversion (this prohibition is contained in the Central Bank Organic Law, any amendment of which requires a 3/5 majority in Congress).

Tariffs are the subject of a uniform tariff across all goods and a ban against non tariff barriers. Both of these changes severely limited the space available for producers to lobby for protection. The need to discuss tariff changes in Congress brings the issue out of the back rooms of bilateral political deals and into the open arena of competitive politics. And the existence of a common tariff makes it more difficult for given sectors to ask for special deals11. The reforms limited the access of business groups to state protection and the benefits of relative price manipulation by the government.

Consider also the area of pensions. Under pay as you go system, the incentives were all there for congressional representatives and pressure groups to demand increases in current benefits to be financed by borrowing (or taxes to future generations with no current political representation)12. With the creation of an individually funded system, the range of social security issues on which the government can be lobbied is much narrower, and the political incentives point in the direction of preserving the accumulated savings of individuals, not spending the money away.

Two other aspects of the current institutional environment are of great importance. One has to do with the budget process and the second with the independence of the Central Bank. While the 1925 Chilean constitution was quite presidentialist and endowed the executive with ample powers in many areas, governments often found it hard to negotiate and obtain approval in a timely and reasonably effective way, of the governments annual budget. The current budget procedure legislation establishes mechanisms that greatly limit the scope for vested interest lobbying on fiscal affairs. First, all bills involving the allocation of public resources can be initiated only by the executive; the role of Congress is limited to approving, amending or turning down. Second, there are strict limits on the time allocated to congressional discussion of the budget. If sixty calendar days after the reception of the bill the legislature has not approved an alternative to it, the executives proposal is automatically approved (see below).

Fiscal institutions

Baldez and Carey (1998) examine the Chilean fiscal institutions. A key hypothesis of the paper is that the Chilean budget deficits should be low relative to regimes with other budget procedures. The hypothesis reflects the insight of a spatial model that concludes that on each policy dimension, the Chilean budget process favors whichever institutional actor (congress or president) that prefers less spending. If the president prefers less spending on a particular item, then his proposal sets the spending ceiling, whereas if congress prefers less, it can amend to its ideal level. Formally by prohibiting Congress from creating new programs or transferring funds among programs, this procedure limits the potential for legislative logrolls that push spending up. Baldez and Carey also assume in their paper that tax and revenue policy is fixed, so that the direct effect of policy makers on deficits is through expenditures.

Constitutional provisions

Article 64 of the Chilean Constitution establishes the procedure for making government spending policy as follows:

"The Budgetary Law Bill must be submitted to the National Congress by the President of the Republic at least three months prior to the date on which it should become effective; should it not be passed by Congress within sixty days of its date of submittal, the proposal submitted by the President of the Republic shall enter into force. The National Congress may not increase or diminish the estimate of revenues; it may only reduce the expenditures contained in the Budgetary Law Bill, except for those established by permanent law.

Estimation of the returns of resources stated in the Budgetary Law and other resources established by any other proposed law shall be the exclusive right of the President, following a report to be submitted by the respective technical agencies. Congress may not approve additional expenditures by charging them to the funds of the Nation without indicating, at the same time, the sources of the funds needed to meet such expenditures. In case the source of funds granted by Congress were insufficient for financing any additional expenditures approved, the President of the Republic upon promulgating the law, subject to favorable report from the service or institution through which new returns are collected, countersigned by the Comptroller General of the Republic, must proportionately reduce all expenditures, regardless of their nature."

The President's agenda control over fiscal policy, moreover, is not limited to the annual budget bill; it extends generally to all spending and tax policy. Article 62 states, in part: "The President of the Republic holds the exclusive initiative for proposals of law related to ... the financial or budgetary administration of the State, amendments to the Budgetary Law...". The President of the Republic shall also hold the exclusive initiative for: "Imposing, suppressing, reducing or condoning taxes of any type or nature, establishing exemptions or amending those in effect and determining their form, proportionality or progression. The National Congress may only accept, reduce or reject the services, employment, salaries, loans, benefits, expenditures and other related proposals made by the President of the Republic".

The Chilean fiscal year is concurrent with the calendar year. Thus, under Article 64, the President proposes a budget to Congress by the end of September and Congress, in turn, must promulgate a budget law by November 30, one month before it is to take effect. The first public sector budget created under the current process was passed by the newly elected President and Congress in 1990.

The final institutional change that we will discuss has to do with Central Bank independence. The 1980 constitution prohibits the Central Bank from financing government deficits by purchasing treasury securities. In addition in December 1989 an organic law was passed intended to guarantee the insulation of the monetary authority from the pressures arising from the executive. At the same time, Central Bank independence put a definitive end to the Central Banks age old vulnerability to private sector demands for direct credit.

In sum, in the current institutional setup, the power of businessmen, public enterprise managers, middle class groups linked to the state apparatus, and trade unions, to lobby for special treatment has been curtailed. The dire predictions of chaos, populism, fiscal deficit, inflation and capital flight did not materialize to a large extent due to the new institutional setup.

THE RELEVANCE OF POLITICAL INSTITUTIONS FOR ECONOMIC POLICY: THE REFORMS TO THE
ELECTORAL AND POLITICAL RULES

The influence of social groups is a complex function of the way the legal and institutional setting provides or limits opportunities for influence. Political institutions, like markets, structure the incentives to political action and can have profound effects on the relative standing and power of different groups. Although governmental elites and state institutions are pivotal to policy determination (see Grindle 1989), their influence on the government is mediated by the structure of political institutions. Hence, this structure will influence the capacity of the government to act.

What is paradoxical about institutional setup is that when reform is required you first need one that permits reform and then, immediately, one that locks those reforms in, making them credible. Once a certain institutional framework permits a deep reform (and the requirements for such a framework have been under discussion for many years, in particular in international organizations), the key question is whether that same institutional framework provides guarantees that the reforms will in turn not be overturned some time soon13. This is a problematic paradox: the institutions that permit reform may not provide a credible long term horizon to the new rules of the game. Hence the need to give signals of the permanence of the new rules once they are in place. This was the major challenge facing the new democratic government in Chile. The consolidation of reform, involves stabilizing expectations around a new set of incentives and convincing economic agents that they cannot be reversed at the discretion of individual decision makers. A key question is whether the institutional setup inherited by the democratic government provided this set of incentives. This paper argues it did.

The party system and electoral rules

The study of inflation (Haggard and Kaufman) suggests that a number of democratic regimes, and in particular those based on catchall two party systems, have a successful record of macroeconomic management. On the other hand, fragmented party systems increase the risks of unstable coalitions, populist appeal, favoritism, and wide swings in policy. Fragmented party systems can result in perverse incentives that are detrimental not only to macroeconomic stability but to democratic governance as well. An important implication of this analysis is that the engineering of the political and electoral system can have important consequences for the stability of electoral and legislative majorities, and thus for their capacity to mobilize support for credible policy initiatives.

If characteristics of the party system affect the level of state indecisiveness in important ways, then the question is how party systems come to be fractioned to begin with. One important determinant of the number of party factions is the electoral structure.

Most electoral systems give large parties a more than proportional share of seats. The larger this big party bias, the greater the incentive to form electoral alliances. This incentive can be summarized in a number, that is obtained by examining the number of viable parties under a given system. Political scientists estimate the "threshold of exclusion" as the largest vote share that a party can win in a given electoral district and still not be guaranteed a seat. In Chile, this is 33% of the vote14. This threshold is often taken as a rough estimate of the minimum viable size of a party, and taking its reciprocal gives an upper bound on the number of viable parties in a given district. Hence in Chile this number is three. While the percentage of votes remains as is, close to 50/50 between the two alliances, there are no incentives for the third party to appear. However, if one of the alliances were to achieve a dominant majority, there would be incentives for it to split.

The Chilean system is unique in its tendency to over represent second place finishers. Due to this fact, it encourages the formation of broad coalitions. The imperative to coalesce is formidable enough for Chile's traditional multiparty system to perform now much like a two party system.

Adopting the terminology of Sartori (1976) we can say that systems are strong when they provide substantial electoral incentives to coalesce, feeble when they provide little or no such incentives. Systems with low district magnitudes of winner take all seat allocation formulas are strong; systems with high district magnitudes and proportional seat allocations are feeble. Strong systems put meaningful upper bounds on the number of parties. Hence the Chilean system, as it is currently operating could be called "strong".

The second important aspect of the party system is that systems that promote intra party competition for votes/seats, promote in turn more candidate based electoral politics; while systems that hinder intra party competition for votes/seats, promote more party oriented elections. If we consider both aspects (see Haggard and Kaufman), we obtain that the present Chilean electoral system provides similar incentives to the US, promoting few decentralized parties while the 1925 Constitution promoted many decentralized parties (the most fragmented party system possible).

In the context of the classification, the decisiveness of the system lies on whether the Executive and Legislative are aligned. As in the US, the presence of a misalignment produces the phenomenon of divided government. Divided government, imposed by the 1980 Constitution and designated senators, imposes an important degree of indecisiveness in the system, favoring the statu quo. This degree of indecisiveness would have a negative consequence in fiscal policy where it not for the rules for the discussion of the budget that provide the system with a large degree of decisiveness even in the context of divided government (see below). However, the last designated senators were designated by the Frei government (1994-2000), so the balance of power may eventually change. It may become easier for the government to pass laws in the Senate without support from the opposition. Whether this will have a cost in credibility remains to be seen, but if the above analysis is correct, once one important rule of the game is modified, this may have a disproportionate costs in terms of credibility.

Haggard and Kaufman also address the issue of the excessive influence of special interests and which electoral or party rules hinder or promote this influence. The question is what institutional structure will minimize fiscal pork and rent seeking. Following Haggard and Kaufman, the current Chilean system would be classified as one that promotes both the demand and supply of pork (because it is characterized by candidate centered elections and many veto gates). Hence in the long run, if they are right, this issue may become important.

Presidentialism versus parliamentarism

We turn now to the presidential nature of the Chilean system. Shugart and Haggard (1998) conclude policymaking in presidential systems depends on the electoral rules and party systems those rules foster and cannot be analyzed without reference to them. Policymaking is likely to be made more difficult to the extent that electoral laws foster multiparty systems in which presidents have difficulty forming legislative coalitions.

One important characteristic of a presidential system is whether the president has agenda setting powers and exclusive rights of introduction of bills to parliament on certain topics. This power is akin to a veto power. South Korea, Taiwan, Brazil, Chile, Colombia and Peru (since 1993) all have constitutional provisions limiting congress's authority to increase the amounts allocated to items in the budget submitted by the president. As Baldez and Carey (1998) show formally, these provisions enhance the capacity of the president to obtain outcomes closer to his preferences. This system in Chile has several favorable characteristics: decisiveness and a tendency to minimize pork, and the tendency to restrain expenditure in every area to the level advocated by the most restrictive of the two powers: executive and legislative. However, this does not guarantee a stable fiscal stance, if both powers agree to an expansive fiscal policy. The constellation that leads to the greatest overall power: decree power, a strong veto, and exclusive powers of agenda setting, is present in Chile on fiscal matters.

One key issue is whether a presidential system provides more credibility to the policy status quo than a parliamentary system. Moe and Caldwell (1994) argue that presidential systems provide protection but parliamentary systems do not. Presidential systems typically establish more vetoes in the policy making process than do parliamentary systems, in which the majority in a single legislative house is usually sovereign. Hence, in a presidential system policies tend to endure. In a parliamentary system, if the majority shifts, the new winners can quickly change the existing rules and hence are able to rule unencumbered by preestablished rules. Some think this is more democratic (and it may be) but it has clear efficiency consequences. Since it makes it more difficult for the state to commit to a policy rule, investment may be lower, interest rates higher, etc.

In Chile, the President's main powers are in the realm of budgetary policy. Only the President may introduces legislation concerning taxes, expenditures, public debt, social security schemes, creating new public services and collective bargaining. Also, in budgetary expenditure or employment legislation, the legislature may not increase any proposed appropriation. The legislature, however, may delete or reduce any item in the proposed legislation. Chile's Constitution requires also a sort of balanced budget, as it prohibits the legislature from approving an expenditure increase without determining the sources of the funds needed for such additional expenditures. Also the President is the only party that can provide an estimate of revenues.

3. CONCLUSIONS

This paper argues that the institutional setup was key to Chile's successful transition and the continuation of economic policy. Running through much of the literature on economic reform is the recognition that it is important to get institutions right. The technocratic approach to this problem is the creation of independent agencies: independent central banks, autonomous regulatory agencies, independent judiciaries, and so forth. The Chilean process is remarkable in its concern to appropriately design both big and small institutions so to provide an adequate framework for decision making by the private sector. Aside from those mentioned above, there are innovative budgetary rules, that include exclusive presidential initiative, and innovative election rules for Parliament. All those reforms worked in the direction of preserving the statu quo and go a long way in explaining the continuance of economic policy from the military to the democratic government.

* The author wishes to thank the comments from numerous colleagues, participants at the LACEA meeting in Santiago, October 1999, and attendants to the workshops at the Pontificia Universidad Católica de Chile. In particular, I wish to thank Alejandro Foxley and Rolf Lüders. A piece coauthored with Alejandro discusses some of the issues treated here (Foxley and Sapelli 1999). Of course neither necessarily shares the views exposed here, and all remaining errors are my sole responsibility.

** Profesor, Instituto de Economía, Pontificia Universidad Católica de Chile. email: csapelli@volcan.facea.puc.cl.

1 Persson and Tabellini (1998) review this literature, and this section is based on it.

2 Some critics (such as McCallum 1996) argue that the institutional remedies discussed in this section do not fix the dynamic inconsistency that is at the core of this literature. The criticism is correct in that the institutions are assumed to enforce a policy which is ex post sub optimal from the incumbents government point of view. Hence there is always a temptation to renege on the institution.

3 Jensen (1996) studies a model in which a government can renege on the initial institution at a cost. In this setting, institutional design improves credibility but does not remove the credibility problem completely.

4 The formal theoretical literature on central bank independence starts with Rogoff (1985) with his analysis concluding that society is better off electing a conservative (tough on inflation) central banker. But the treatment of society's problem as a principal agent problem is anticipated by Barro and Gordon (1983).

5 Hence in a simple two period model, the government finds it optimal ex post to set either a fully expropriating capital tax rate of 100% or a tax rate sufficiently high to finance all of public consumption with capital taxes, driving labor taxes to zero. Anticipating this, nobody invests. This equilibrium is disastrous: since there is no investment, a large tax on labor will have to be levied as labor is the only available tax base. The government can't do anything to avoid this outcome. No promise to tax capital at a rate lower than 100% would be believed since it would be optimal for the government to ex post renege on the promise.

6 Absent a commitment technology (i.e. an institution), policy is driven by private expectations rather than the other way around. If investment ex post is large, then the tax rate on each unit of capital is small, but if it is low, then the unit tax rate may be huge. The ex post optimal policy depends on aggregate investment. But how much aggregate investment there will be depends on expectations about policy. Individual expectations are self fulfilling and as they are not nailed down by any economic fundamental, they can fluctuate widely. If most think that investment will be large, and therefore taxes per unit low, then investment will be large. If most think investment will be low, and taxes per unit high, then investment will be low.

7 See, among others, Persson, Roland and Tabellini (1997).

8 The results described above are derived from models with two political parties. When we introduce coalition government and more than one decision maker, then there are added problems. The results are an example of the familiar common pool argument: as the property rights to future income are not well defined each of the parties only internalizes a fraction of the cost of current spending and debt issue. The result is a solution that departs radically from the cooperative solution.

9 Though an agreement was made at the time to keep future increases in the minimum wage related to gains in labor productivity and expected inflation, this agreement was violated in 1998 when a three year increase in the minimum wage surpassed any credible increase in productivity plus expected inflation.

10 See Edwards (1993) for an explanation of what caused the transformation in economic views: "A combination of factors were responsible, including the failure of the heterodox programs in Argentina, Brazil and Peru in the mid 1980s and the general sense that the state based development strategy of the previous decade had run out of steam. Many considered the heterodox programs of the mid 80s as a final opportunity for structuralism -in its new incarnation as neo-structuralism- to work. Once the programs failure was clear, politicians turned to the stellar performance of Chile. The largely successful Chilean experience became a model for many countries and the collapse of the Soviet Union pulled the rug out form under the Marxist camp that had long advocated replacing markets with central planning. The Spanish experience with economic reform under socialist leader Felipe Gonzalez also became an important factor."

11 For a review of the political economy arguments for a common tariff, see Rodrik and Panagariya 1991.

12 See Alesina and Tabellini (1990) and Velasco (1993) for models in which the possibility of borrowing creates incentives for systematic government deficits. See Tabellini (1990) for an analysis of the politics of intergenerational redistribution through social security.

13 Corbo, Luders and Spiller (1995) argue that the Chilean system has a strong set of checks and balances based on the diffusion of power. This is because its based on divided government. Electoral rules lead to party and regional fragmentation and hence the legislature does not respond to a single party and that parties themselves will not be very homogeneous. Thus, as emphasized by Shugart and Carey (1992), the Chilean system may provide for gridlock as legislation may not pass easily. Given that the initial conditions prior to the resumption of constitutional rule in 1989 were pro market, the potential gridlock may translate into a strong signal of stability of those conditions.

14 Both the Chamber and Senate are elected from two member districts. The two member electoral system is unique to Chile. Under the system, parties or coalitions present lists of two candidates, and voters indicate a preference for one candidate within one of the lists. The votes of all lists are totaled and the two seats allocated: the first seat is assigned to the preferred candidate from the list with the most votes, after which the list's vote total is divided by two. If this quotient is still higher than any other lists vote total, the second candidate gets the second seat. Otherwise, the second seat goes to the first candidate on the second place list.

REFERENCES

Alesina, A. and G. Tabellini (1990), "A Political Theory of Fiscal Deficits and Government Debt" Review of Economic Studies, 57 (July) 403-14         [ Links ]

Alesina, A. and R. Perotti (1995), The Political Economy of Budget Deficits. IMF Staff Papers March 1-37         [ Links ]

Alesina, A. and R. Perotti (1995), Budget Deficits and Budget Institutions. NBER WP 5556         [ Links ]

Alesina, A.; Osler, N. Roubini and Swagel (1996), Political Instability and Economic Growth, Journal of Economic Growth 1:189-212.         [ Links ]

Baldez, L. and J. Carey (1998), "Budget Procedure and Fiscal Restraint in Post-Transition Chile" Chapter Five in Political Institutions and the Determinants of Public Policy: When Do Institutions Matter? by Haggard and McCubbins (eds.).         [ Links ]

Barro, R. and D. Gordon (1983), A Positive Theory of Monetary Policy in a Natural Rate Model, JPE 91: 589-610.         [ Links ]

Barro, R. and D. Gordon (1983), Rules, Discretion and Reputation in a Model of Monetary Policy, Journal of Monetary Economics, 101-121.         [ Links ]

Bosworth; R. Dornbush and R. Laban, eds. (1994), The Chilean Economy: Policy Lessons and Challenges, The Brookings Institution.         [ Links ]

Boeninger, E. (1992), "Governance and Development: Issues, Challenges, Opportunities and Constraints." In Proceedings of the World Bank Annual Conference on Development Economics 1991. Washington D. C., World Bank.         [ Links ]

Chamley, C. (1986), Optimal Taxation of Capital Income in General Equilibrium, Econometrica 54: 606-622.         [ Links ]

Corbo,V.; R. Luders and P. Spiller (1995), "The Institutional Foundations of Economic Reforms: The Case of Chile" Mimeo, Pontificia Universidad Católica de Chile.         [ Links ]

Cox, G. and M. McCubbins (1998), "Political Structure and Economic Policy: The Institutional Determinants of Policy Outcomes" Chapter Two in The Origins of Liberty: Political and Economic Liberalization in the Modern World, Drake and McCubbins (eds.).         [ Links ]

Dornsbush, R. and S. Edwards (1991), The Macroeconomics of Populism in Latin America, University of Chicago Press.         [ Links ]

Drake, P. and M. McCubbins (eds.) (1998), The Origins of Liberty: Political and Economic Liberalization in the Modern World, Princeton University Press.         [ Links ]

Edwards, S. (1993), Latin America and the Caribbean: A Decade After the Debt Crisis. The World Bank .         [ Links ]

Foxley, A. and C. Sapelli (1999), "Chile's Political Economy in the 1990s: Some Governance Issues", chapter 9 in Chile: Recent Policy Lessons and Emerging Challenges, edited by G. Perry and D. Leipziger, The World Bank, Washington D.C.         [ Links ]

Grindle, M. S. (1989), "The New Political Economy" Working Paper, World Bank, Country Economics Department, Washington D.C.         [ Links ]

Haggard, S. and R. Kaufman (1992), "Economic Adjustment and the Prospects for Democracy" Chapter seven in The Politics of Economic Adjustment Haggard, Stephan and Robert Kaufman (eds.).         [ Links ]

Haggard, S. and R. Kaufman (1992), "Introduction" in The Politics of Economic Adjustment Haggard, Stephan and Robert Kaufman (eds.).         [ Links ]

Haggard, S. and R. Kaufman (1992), "The Political Economy of Inflation and Stabilization in Middle Income Countries" Chapter Six in The Politics of Economic Adjustment Haggard, Stephan and Robert Kaufman (eds.).         [ Links ]

Haggard, S. and R. Kaufman (eds.) (1992), The Politics of Economic Adjustment. Princeton University Press.         [ Links ]

Haggard, S. and M. McCubbins (eds.) (1998), Political Institutions and the Determinants of Public Policy: When Do Institutions Matter? Princeton, NJ.: Princeton University Press.         [ Links ]

Labán, R. and F. Larraín (1995), "Continuity Change and the Political Economy of Transition in Chile", in Reform, Recovery and Growth, edited by R. Dornbush and S. Edwards. University of Chicago Press.         [ Links ]

Levy, B. and P. Spiller (1994), "The Institutional Foundations of Regulatory Commitment: A Comparative Analysis of Telecommunications Regulation" Journal of Law, Economics and Organization. October.         [ Links ]

MacCallum, B. (1996), Crucial Issues Concerning Central Bank Independence. NBER WP 5597         [ Links ]

Mendoza, E.; Ferretti, J.M. and Asea, P. (1996), On the Ineffectiveness of Tax Policy in Altering Long Run Growth. CEPR DP 1378.         [ Links ]

Mendoza, E.; Razin, A. and Tesar, L. (1994), Effective Tax Rates in Macroeconomics: Cross Country Estimates of Tax Rates on Factor Incomes and Consumption. Journal of Monetary Economics, 34: 297-323.         [ Links ]

Moe, T. and Caldwell, M. (1994), "The Institutional Foundations of Democratic Government: A Comparison of Presidential and Parliamentary Systems" Journal of Institutional and Theoretical Economics. 150/1: 171-95.         [ Links ]

Persson, T. and G. Tabellini (1990), Macroeconomic Policy, Credibility and Politics, Harwood Academic Publishers.         [ Links ]

Persson, T. and G. Tabellini (1995), "Double Edged Incentives: Institutions and Policy Coordination", in Grossman and Rogoff (eds.) Handbook of International Economics, Volume III, Amsterdam, North Holland.         [ Links ]

Persson, T. and G. Tabellini (1999), "Political Economics and Public Finance". Handbook of Public Economics, ed. by A. Auerbach and M. Feldstein, North Holland, forthcoming.         [ Links ]

Persson, T. and L. Svensson (1989), Why a Stubborn Conservative Would Run a Deficit: Policy with Time Inconsistency Preferences. QJE 104: 325-345.         [ Links ]

Persson, T.; G. Roland and G. Tabellini (1997), Separation of Powers and Political Accountability, QJE 112: 1163-1202.         [ Links ]

Persson, T.; G. Roland and G. Tabellini (1998), Towards Micropolitical Foundations of Public Finance, European Economic Review 42: 685-694.         [ Links ]

Persson, T. and G. Tabellini (2000), "Political Economics and Macroeconomic Policy", forthcoming in John Taylor and Michael Woodford (eds.) Handbook of Macroeconomics, North Holland.         [ Links ]

Rodrik, D. and Panagariya, A. (1991), "Political Economy Arguments for a Common Tariff" The World Bank         [ Links ]

Rogoff, K. (1985), The Optimal Degree of Commitment to an Intermediate Monetary Target QJE 100: 1169-1190,         [ Links ]

Sapelli, C. (1992), Tamaño del Estado, Instituciones y Crecimiento Económico, CERES-CINDE, Santiago de Chile.         [ Links ]

Sartori, G. (1976), Parties and Party System: A Framework for Analysis. Cambridge U.K., Cambridge University Press.         [ Links ]

Shugart, M. and S. Haggard (1998), "Institutions and Public Policy in Presidential Systems" Chapter Three in Political Institutions and the Determinants of Public Policy: When Do Institutions Matter? by Haggard and McCubbins (eds.).         [ Links ]

Tabellini, G. (1990), "A Positive Theory of Social Security" NBER WP 3272.         [ Links ]

Velasco, A. (1994), "The State and Economic Policy: Chile 1952-1992", Chapter 8 in Bosworth et al. The Brookings Institution.         [ Links ]

Von Hagen, J. (1992), Budgeting procedures and fiscal performance in the European Community, Economic Papers 96, EC.         [ Links ]