15/3? PSD Occasional Paper No. 7 September 1995 Bolivia and the Slowdown of the Reform Process Juan Antonio Morales ~ TThe World Bank Private Sector Development Department Private Sector Development Department Occasional Paper No. 7 Bolivia and the Slowdown of the Reform Process Juan Antonio Morales September 1995 The author would like to thank Brian Van Arkadie, John Page, Izak Atiyas, Leila Frischtak, and Stephan Haggard for their comments. This paper has been cleared for inclusion in the occasional paper series by R. Shyamn Khemani, Manager, Competition & Strategy. The views expressed are those of the author(s) and should not be attributed to the World Bank. This paper forms part of a forthcoming volume, Leila Frischtak and Izak Atiyas (eds.), "Governance, Leadership and Communication: Building Constituencies for Economic Reform." The World Bank Private Sector Development Department ii Contents 1. Introduction ...................................................... 1 2. S.D. 21060 and Its Background. 5 Overview of Political Developments and the State Capitalism Model of the Revolution of 1952 .5 Hyperinflation .7 Stabilization and Reform with S.D. 21060. 8 3. State of the Structural Reforms ...................................................... 11 Tax Reform .................................................... 11 Trade Reform .................................................... 12 Financial Sector Reform .................................................... 13 Incentive Regimes and New Investment Codes .................................................... 14 Pending Reforms .................................................... 15 Overall Measures of Adjustment Success .................................................... 17 Measures of Success in Fiscal Adjustment .................................................... 18 General Measures of Fiscal Stance .................................................... 18 The Structure of Public Spending .................................................... 19 4. Statecraft in the Reforms ...................................................... 25 Initial Weakness of the Government .................................................... 25 Businessmen and Technocrats in Power .................................................... 27 The Weakening of Organized Labor .................................................... 28 5. Governance Capacity and the Second Generation of Reforms ................. ................... 29 Governance Capacity and Fiscal Reform .................................................... 29 Problems with Privatization and Social Security Reforms ................................................ 33 6. Concluding Remarks ...................................................... 37 References ...................................................... 39 iii Al 1 Introduction 1.1 Bolivia was one of the early adjusters of the 1980s, and the success of its stabilization program of August 1985 in stopping hyperinflation has received considerable attention. The program went beyond stabilization and included a broad set of economic reforms, encompassing the liberalization of the markets for goods, money and credit, and labor. The stabilization program and these first reforms were packaged in the omnibus Supreme Decree 21060. Later, S.D. 21060 became the symbol of the new development model for Bolivia. Bolivians would refer to the liberal model in force after 1985 as the model of S.D. 21060. 1.2 In the aftermath of S.D. 21060, the government designed and implemented, in a very orderly way, an economic policy rich in transformations. Victory in the fight against inflation, a problem well-defined and clearly identified by the public, increased the mandate of the government in other policy areas. 1.3 Bolivia's experiment of the 1980s was not only successful in controlling inflation, but was also was a forerunner in the adoption of structural reforms. Equally significant, changes in government did not imply breaks in economic policy. All administrations, despite differences in personality and previous visions, have followed the development model outlined in the reforms of 1985. 1.4 After demonstrating extraordinary zeal in pursuing reform in 1985 and the first few years of the program, the momentum seems lost. I hypothesize that this happened because the new reforms demanded greater governance capacity, understood in the sense given by Frischtak (1994, p. vii), "as the ability to coordinate the aggregation of diverging interest and thus promote policy that can credibly be taken to represent the public interest." A main objective of this paper is to look at the rmissing political and institutional conditions that are needed to forcefully renew and sustain the reform process. 1.5 The most important, although still preliminary, result that emerges from the study is that the Bolivian experience lends support to the hypothesis that the dynamics of stabilization and the reforms that immediately followed differ from the complementary structural adjustments. Once the situation is ripe, stabilization measures, however harsh and undemocratic, can be taken without losing the public's support. When the emergency disappears and things go "back to normal," the design and, to a greater extent, the enactment of the complementary reforms become I 2 Bolivia and the Slowdown of the Reform Process more complex. A major aim of this paper is to contrast the two different settings in such a way that we can isolate the main features of governance capacity. 1.6 The stabilization and the first structural reforms, which were indistinguishable during several months of the stabilization package, will be labeled, for convenience, first- generation reforms. The stabilization and the first reforms were designed to: (1) restore macroeconomic stability; (2) correct relative prices to make them closer to market-determined prices; (3) not unrelated to the point before, open the economy to foreign trade; and (4) remove the bigger obstacles to the smooth functioning of markets. These measures were mostly of a macroeconomic nature. The complementary, or second-generation, reforrns are related more to a redefinition of the roles of government and the private sector, and of the national government and the local governments. Privatization and decentralization within government are central to this redefinition. Also, the complementary reforms are intended to foster institutional change to enhance savings and investment and to strengthen international competitiveness. The reform of Social Security and the laws to guarantee private investment stand out in this area. 1.7 The tax and budget reforms cross both generations of reforms, but there are distinctive differences between the two generations. The content of the first fiscal reforms was defined by the urgent need to narrow the budget deficit. This was achieved with easy-to-collect taxes and easy-to-reduce expenditures. The second fiscal reforms take a longer view. A great deal of effort is given to changes in the composition of expenditure, assuming that there is a ceiling. Also, significantly more attention is paid to the efficiency and equity issues in the tax structure, which were somewhat neglected in the first generation of reforms because of the overwhelming need for revenue. 1.8 The distinction between stabilization and the subsequent first reforms, and the second generation of reforms, can also arise (as in Naim 1994) from the way they were formulated and implemented. Naim prefers to identify the stabilization and the first generation of reforms as Phase I reforms, and our second generation of reforms as Phase II. The Phase I reforms were undertaken in the midst of a severe crisis by a small number of powerful ministers and the president of the Central Bank, advised by high-level technocrats, and isolated from political pressures. The economic theory behind the Phase I reforms is well-established, and utilizes the lessons of previous experience in the country and elsewhere. The measures are packaged in decrees and executive orders, in Naim's words (1994, p. 8), "hard-to-decide but simple-to-execute. " 1.9 The second-generation reforms are significantly more complex. Their requirements in statecraft, political adroitness, and administrative talent will be examined in detail in the pages that follow. 1.10 The market liberalizations were seen by the government of 1985 as a constituent part of--and indistinguishable from--the stabilization program. Only afterward did they acquire their own identity as structural reforms. Many doubts lingered of the permanence of the market liberalizations after the first signs of stabilization. The prevailing view in the public and in some parts of the government was that the adjustment measures were of a transitory nature. Yet there Introduction 3 was also the conflicting perception that the hyperinflation was not only a temporary, albeit severe, disorder, but the most telling symptom of the failure of a development model, and thus that there was a need for extensive reforms with a long-term outlook. The government rapidly seized this argument. Equally important, it was a convenient contention to bring in crucial foreign aid. 1.11 Once it became clear to the public that the government was really committed to stop inflation, its capacity to formulate and implement reforms increased. The government could then shape a national project and offer it to the public and to the community of foreign donors. It is argued in the text that increased governance capacity was not the outcome of a previously set vision, but rather the result of the feedback between government and the public. Adjustment was essentially a dynamic process, with the contents of the program changing each time new information became available, either on the position of the public or the foreign donors. 1.12 The first-generation reforms could be delivered in clear and simple rules., The public understood the messages and accepted them. Even when the reforms introduced regulations, they didn't stir special resistance, because they were perceived as universal. Rules, when seen as universal, are less likely to bring opposition than measures that are perceived as creating obvious and undeserving winners. 1.13 The first-generation reforms conveyed signals of fairness, and this was very important for their implementation. Even if they entailed costs, the so-called "social costs," they were lower than the costs of macroeconomic disarray. Similarly, the market liberalization appeared to place in the open widespread, but hidden, practices. More transparency was seen as an increase in fairness. Everybody could benefit from what had previously been benefits only for the few with privileged information. When regulations were imposed on some sectors--for example, banking--they did not tamper with prices or dictate the allocation of productive activities. Instead, they took the form of rules of prudence, which were understood as preserving 2 the common good of the sector, even by those directly affected. 1.14 This paper is organized as follows. The first section surveys the conditions preceding S.D. 21060 and the contents of the stabilization package of 1985. The next section offers a description of the state of the reforms, and ends with some quantitative results. In the third section, I look at the political competence of the reforrmist governments in dealing with the adjustment measures and coping with the opposition. The fourth section considers capacity governance, with an emphasis on the fiscal accounts and the difficulties met in two of the second-generation reforms, privatization and Social Security reform. The final section offers some concluding remarks. 1 To the informed, they looked like simplistic rules of thumb, such as "govemments should not spend more than they receive." 2 There is a public good argument to justify the acceptance of the prudential rules. Everybody in the sector could find the prudential rules desirable, but only if everybody else was accepting them. The coordinating role was left to the govemment. 4 Bolivia and the Slowdown of the Reform Process 2 S.D. 21060 and Its Background 2.1 It has become customary in the current Bolivian political science literature to place the beginning of the state-led development model in the Revolution of 1952, and its end in the structural changes brought by the set of measures encompassed in Supreme Decree 21060 of August 29, 1985. This section sketches the main political and economic developments between these two landmarks and provides a thorough description of the reforms included in S.D. 21060. Overview of Political Developments and the State Capitalism Model of the Revolution of 1952 2.2 The Revolution of 1952 is a watershed in Bolivia's history.3 The revolution, conducted by the Mr. Victor Paz-Estenssoro and his party, the Revolutionary Nationalist Party (henceforth, MNR, the Spanish acronym), marked the beginning of a development model with strong state leadership. A high share of public investment in GDP and in total investment--not only in infrastructure but also in the agroindustrial and manufacturing sectors--became a permanent and prominent feature of the model. 2.3 Of equal importance with public investment was the establishment of a complicated system of incentives and sanctions to influence private sector production and investment decisions. The main intervention instruments were the allocation of heavily subsidized credit and tax incentives. There were also protective barriers to shelter domestic industry from competitive imports, although they were not as important as in the other Latin American countries, mainly because Bolivia's geographic position provides natural walls to trade. 2.4 The overexpansion of state involvement in the economy was a source of job patronage, as well as clientelism in contracts of public works and purchases of the government and the state-owned enterprises (SOEs). In the past forty years, Bolivia has had the highest share of public employment in the total urban labor force in Latin America, except for Cuba. In addition, after 1952 a segment of the middle classes emerged whose wealth and income resulted from the predation of public resources. Corruption became common, and its scope augmented compared with previous years. 3 See, for example, Sachs and Morales 1988 and the references therein. 5 6 Bolivia and the Slowdown of the Reform Process 2.5 The MNR stayed in power during twelve eventful years. It was overthrown by the military in 1964. The internal strife in the MNR and its worn-out power were as much behind the coup as the ambitions of the military. Despite the demise of the MNR, the state capitalism model was continued by all governments between 1952 and 1985, military and civilian alike, independent of their political orientation. 2.6 General Hugo Banzer's government stands out among the military governments, not only because it lasted seven years (1971-78), but also because it gave a strong impetus to the modernization of the economy. The heyday of the state capitalism model was reached during his reign. This seems paradoxical in view of his conservative and strongly anti-Communist stance. Ironically, the possibility of obtaining loans in private capital markets in that period gave a further impetus to state involvement in the economy. 2.7 Despite his notorious disrespect for human rights, Banzer's administration and the other long-lived authoritarian governments enjoyed the support of ample segments of the population, particularly in the middle and upper classes, with their calls for order. These authoritarian governments had some legitimacy and clearly exerted power over society. 2.8 For short periods (1970-71 and 1978-82), all indications of governance capacity were lost. During a populist military government in 1970-71, lack of social discipline manifested itself in illegal occupations of private property, wildcat strikes, and many attempts at coups d'etat. The second period, 1978-82, was characterized by interim civilian presidents and a succession of military governments. The nadir of this critical period was in 1980, when a military junta, with close ties to the cocaine traffic, took power. 2.9 Democracy returned to Bolivia in the last quarter of 1982, after eighteen years, with only short interruptions by authoritarian governments. The first democratic government (1982-85), headed by Mr. Hern.n Siles-Zuazo, was backed by a coalition of leftist parties, including the Communist Party of Bolivia. The labor unions initially favored Siles-Zuazo, but later in his term withdrew their support and were among his more violent opponents. 2.10 Siles-Zuazo inherited a very difficult economic situation, prompted largely by the foreign debt crisis. Unable to address the economic predicament and all the pressures that went with it, he had to resign the presidency one year before the end of his constitutional mandate. It was during his term in office that hyperinflation erupted. By mid-1985, Bolivia's economy was in complete disarray, and the accompanying social and political unrest threatened democracy. 2.11 President Siles-Zuazo had to call anticipated elections, which took place in July 1985. After the elections, Congress chose Mr. Victor Paz-Estenssoro as president. He came with the support of his aging populist party, MNR. As mentioned above, the MNR was the party of the Revolution of 1952, when the state capitalism model started. In a turnaround, Paz- Estenssoro and the MNR decided on a radical transformation of the development model, putting in place very ambitious structural reforms. 2.12 The MNR signed a pact, the Pact for Democracy, with the Democratic Nationalist Alliance (henceforth, ADN, the Spanish acronym), a rightist party led by General Banzer, to S. D. 21060 and Its Background 7 carry on the liberal revolution. Since this initial signing, pacts have become a permanent feature of Bolivian politics. 2.13 Paz-Estenssoro was succeeded in 1989 by Mr. Jaime Paz-Zamora. He also formed a coalition government between his formerly leftist party, the Revolutionary Leftist Movement (MIR, the Spanish acronym), and ADN. Paz-Zamora pursued the policies set by his predecessor. The change in MIR's stance was as surprising as the change of course of the MINR. 2.14 The current president of Bolivia, Mr. Gonzalo Sdnchez de Lozada, began his presidency in August 1993. He governs with his party, the MNR, and three smaller parties; in this group, the Free Bolivia Movement (MBL), a dissident segment of intellectuals of MIR, stands out. Sanchez de Lozada was the Minister of Planning (actually, the Minister of the Economy) during the government of Paz Estenssoro and the main architect of the changes. Hyperinflation 2.15 Bolivia suffered a dramatic episode of hyperinflation between 1982 and 1985. There is a professional consensus that the hyperinflation was prompted by the accumulation of large fiscal deficits between 1975 and 1981, financed with foreign loans. The general situation of the Latin American debt, and the particular exposure of Bolivia, triggered an external crisis in 1982 that almost immediately led to hyperinflation. Numerous disruptive factors converged. First, accumulated trade surpluses were needed to service the foreign debt. The available instrument was the depreciation of the currency, which not unexpectedly was strongly resisted by the workers. On the path to a higher real exchange rate, strong inflationary pressures emerged. Second, and more important, the financing of the fiscal deficit with foreign loans was replaced with credits of the Central Bank, which required excessive issuance of money. Since the government could no longer roll over its debt to the foreign creditors, and there was no domestic market for government paper,4 the Central Bank loans became almost the only source of financing of the fiscal deficit. 2.16 As the public finance models of inflation predict, this mode of deficit financing interacted with the public's demand for money, and resulted in high rates of price growth. In addition, inflation itself further aggravated the fiscal deficit through Olivera-Tanzi effects. 2.17 Bolivia found itself very rapidly in a high and accelerating inflation path. Government attempts at control by fixing prices and the exchange rate were futile; worse, they became counterproductive. Black markets soon appeared. The premium on the black market for foreign exchange over the official rate was especially notorious: at times, it went up to 1,300 percent. Together with supply shortages, this led to frequent incidents of severe social unrest. 2.18 Contrary to the diagnosis of inflation in neighboring Argentina, Brazil. and Peru, for most of this period the government, its economists, and the public were convinced that the 4 The domestic market for government bonds disappeared in the early 1930s, in the aftermath of the Great Depression. In 1987 the Central Bank started offering certificates of deposit to domestic savers. Open market operations with Treasury bills started in 1994. 8 Bolivia and the Slowdown of the Reform Process pillar of stabilization was fiscal correction.5 Between 1982 and mid-1985, there were at least six attempts to stabilize. Except for the first attempt, all were "orthodox" in the sense that fiscal and monetary restraints were the main intermediate targets (the ultimate target, of course, was the control of inflation). They all failed. Their failure was not attributed to an incorrect conception, but to having been halfhearted measures, with a gradualist approach to a condition that required a shock treatment. 2.19 By 1982 the illegal trade in cocaine had become a major disruptive factor to politics and economic management. The failures of the Bolivian government to take action against the drug traffic, as evaluated by the United States government, directly interfered with the design of economic policy between 1982 and 1985. For instance, independently of the technical merits of the successive stabilization plans, crucial financial support from the United States was withheld. Cocaine continues to play a major role, even after the reforms of 1985, both in Bolivian politics and in economic policymaking. Stabilization and Reform with S.D. 21060 2.20 Three weeks after his inauguration, President Paz-Estenssoro announced a stabilization plan in S.D. 21060. The plan was prepared, in great secrecy, by no more than seven persons. Also, the process of carrying out the program was the responsibility of only a few people. 2.21 Exchange rate unification was central in the program. This measure was supported by very tight fiscal and monetary policies. The fiscal measures were especially strong. On the revenue side, the price of fuels was hiked by a factor of seven. This increased the tax collections on fuel consumption by the same factor. In addition, the state-owned petroleum enterprise, YPFB, was obliged to deposit all of its sales proceeds daily in an account in the Central Bank. Subsequent use of the deposit was then tightly controlled by the government. The deposits worked as a powerful brake to the expansion of money; later, a significant portion of the deposits was transferred to the Treasury. To this day, taxes on fuels, particularly transfers from YPFB, are among the most important sources of Treasury revenue. 2.22 On the expenditure side, the measures were also drastic. Approximately 10 percent of the labor force in the public sector lost their jobs. The policy of dismissals was extremely crude, with no consideration of age or reemployment possibilities. Wages and public sector investment were frozen for an initial period of six months. To give a clear message to the public of the government's commitment to fiscal correction, the budget was managed on a cash- flow basis in the first years of the stabilization program. 2.23 To mitigate the costs of the adjustment, the government created the Emergency Social Fund (ESF) in 1987. ESF has received considerable attention and praise for its accomplishments in the international financial institutions (IFIs). It was very efficiently run, and 5 The inability to formulate a heterodox stabilization plan reflected, to a large extent, the lack of domestic analytic talent. Orthodoxy had the advantage of being a well-known recipe. S. D. 21060 and Its Background 9 for a while it became an example for other countries to follow. Yet, like all narrowly targeted policies, ESF suffered from political isolation. The very poor are not the main losers in adjustment; it is the politically vocal lower-middle-classes and unionized workers that suffer the higher drops in both take-home income and government transfers in an adjustment process. There was no mitigation program for them, except through larger than usual severance payments. 2.24 The markets for goods and factors, except land, were greatly liberalized. Price controls were lifted, import tariffs were lowered, and most quantitative restrictions on imports and exports were eliminated. 2.25 Liberalization seemed at odds with the prevailing thought on stabilization. On purely technical grounds, the onus of stabilization had to fall on fiscal and monetary restraint; liberalization was not necessary, and could even be counterproductive. The foreign trade liberalization could--and did--deteriorate the trade balance. The deregulation pushed steeply upward, in an once-and-for-all movement, the prices of wage goods, which fueled labor requests for higher wages. Had it not been for the extraordinary resistance of the government to pressures to raise wages, the program would have capsized. 2.26 The program checked inflation very rapidly, unified the exchange rates to the market-determined rate, and substantially reduced the fiscal deficit, at least in the first years. On purely technical grounds, a less drastic plan would probably have achieved the same results. The overkill essentially had an announcement value. The government needed to differentiate its program from the previous ones that, although similar in content, had failed. The package of stabilization cum liberalization, a bit by accident, integrated correct economic policy ingredients and the right political announcement features to convince the public of the government's commitment to stop inflation. 2.27 An unorthodox but crucial step was included in the program -the suspension of payments on the foreign debt to commercial banks until a definite arrangement could be reached with them. Loan service to official lenders continued. 2.28 The IMF accepted the stabilization plan, although it had some disagreements on the management of the exchange rate and the suspension of the commercial debt service. The World Bank offered its support to the program from the very beginning. More generally, the external agencies have played a very important role in the stabilization and subsequent reforms. Bolivia has enjoyed (sometimes endured) their technical and financial support since the first measures in 1985. 10 Bolivia and the Slowdown of the Reform Process 3 State of the Structural Reforms Tax Reform 3.1 Once the high inflation seemed to be receding, the reconstruction of taxation appeared to be the next step necessary to consolidate the program. Remember that a main victim of the hyperinflation had been the tax system. Besides the inherent deficiencies in the previous tax structure, Olivera-Tanzi effects had destroyed the tax administration. A new tax law was passed in June 1986 to enhance the stabilization package. 3.2 The tax reform of 1986 featured only eight taxes, if the special treatment to small taxpayers is excluded. The most important tax, by far, is the value added tax (VAT), which currently has a uniforn rate of 13 percent (it started at 10 percent). Wages and income from assets receive a similar treatment to that of any sale subjected to VAT. (Note that there was no tax on profits but on presumed income.) They bear a complementary VAT, from which VAT payments made on expenditures can be credited. This tax carries the same rate as the VAT on personal saving. It can be observed in tables 6 and 7 that the real value of VAT collections and its share in the current revenue of the National Treasury have been steadily increasing. 3.3 Following the launching of tax reform, a great effort was devoted to enforcement and administration. The registers for taxpayers were updated, a mechanism for cross-checks of VAT was created, and the system was fully computerized.7 Many, particularly small, businesses were closed by the authorities for fiscal fraud. The contribution of tax reform to the fiscal position was undoubtedly significant, yet the most important taxes to that end were the "taxes" on fuels, which were included in S.D. 21060. 3.4 Was the tax reform of 1986 necessary on intrinsic technical grounds? The subject is debatable, because once inflation had abated, a reversed Olivera-Tanzi effect would have increased revenues. Also, the previous tax system was not without merits. It is true that a myriad of taxes (400?) existed before 1986, but the ones that raised revenue in significant amounts were no more than ten. The remaining 390 were a nuisance, but they did not really impair tax admninistration, nor were they a source of corruption. The most important criticism 6 The tax law of 1986 was fully reformed in December 1994. Between 1986 and 1994. it had suffered some minor changes. 7 Bolivia now has one of the most modem tax registers in region. 11 12 Bolivia and the Slowdown of the Reform Process that can be made of the tax structure before 1986 is that the weight of taxes on foreign trade (tariffs, and on exports of the mining and oil sectors) was too high. 3.5 It is also true that tax compliance was very low during hyperinflation, but this was not necessarily caused by an increase in corruption in tax administration. This effect was, again, a byproduct of inflation and demoralization in the internal revenue services and in the public enterprises selling services. The employees felt that to enforce the tax laws was not worth the trouble, given that what they collected from taxpayers and for service fees was so little in real terms. Tax reform was needed essentially because of its announcement value. The government had to send the message that rebuilding tax collections had a top priority in its agenda of fiscal correction, the latter being the most important component of the stabilization package. Trade Reform 3.6 Both economic and political causes explain Bolivia's success in trade liberalization. Among the former, the low degree of industrialization stands out. Among the latter, one can identify the correct assessment of the forces of the potential opposition, as well as tenacity in government. 3.7 The steps taken to liberalize trade were forceful. S.D. 21060 eliminated almost all quantitative restrictions and lowered tariffs. Later, further reforms led to a tariff structure consisting of flat ad valorem tariffs, with a rate of 10 percent, except for a small list of capital goods, whose rate is 5 percent. 3.8 Trade liberalization was very quick and, as with many of the first reforms, met little hostility. The opposition came from businesses in the manufacturing sector and from large agriculturalists. The National Chamber of Manufacturers hired a consulting company to compute effective rates of protection, and based on this technical report pressed for changes in the tariff structure. Similarly, the powerful Eastern Chamber of Agriculture (henceforth, CAO, the Spanish acronym) lobbied for changes. The government adamantly refused. 3.9 It must be said that the pressures were not very strong, given that there were conflicting views among the diverse sectors. On the one hand, almost everybody wanted the "model" to succeed, avoiding a return to the ominous years of 1982-85. On the other hand, while they perceived liberalization as costly, they were uncertain of the scope of the prereform protection, and even more unsure of the proposals from their umbrella organizations. In addition, many sectors were accustomed to competing with contraband imports, and what they resented was not so much the lowering of tariffs on their final products, as the increase in tariffs for their inputs. They felt that the fall in effective rates of protection was caused by the uniformed tariffs rather than the lower tariffs for competitive imports. 3.10 Contrary to the expectations of many Bolivian economists, output in the manufacturing sector strongly and positively reacted to the stabilization program.8 The beneficial 8 Unfortunately the growth of the manufacturing sector was not reflected in the overall GDP rate of growth, because of the small initial contribution of the sector. State of the Structural Reforms 13 effects of the reduction in uncertainty more than compensated the dismantling of trade barriers. The government correctly anticipated this reaction. Still, results varied across subsectors. In some branches there were many closings. The manufacturing sector, because of its smallness, traditionally had little voice in influencing policies. Enterprises in many sectors opted for exit, given that their voice was unlikely to be heard. 3.11 CAO's lobbying was more effective. Their associates suffered from higher tariffs on imported inputs and from agricultural imports whose competitive edge was provided by the very rapid exchange rate depreciations in neighboring Peru, Argentina, and Brazil. The government did not yield to the demands of the CAO, but it went to significant lengths to compensate CAO's associates. Since many of them were exporters, a subsidy of 10 percent of the free on board (FOB) value of exports was granted, allegedly a drawback for taxes on imported inputs. Moreover, transportation fees in the state-owned railroads were reduced, to CAO's benefit. In 1991 the export subsidy was lowered to 2 percent for agricultural exports, a rate that is closer to a true drawback than the previous higher rate. 3.12 The most significant concession, however, was the reestablishment of lines of credit, among them an important IDA loan, labeled "Eastern Lowlands Project." The WIs went full force to help in the development of agricultural exports, particularly of soybeans.9 This program partially silenced CAO's demands. Financial Sector Reform 3.13 Another area where considerable statecraft was exerted was in the reform of the financial, mainly banking, sector. The banking sector had known a considerable rate of growth between 1964 and 1982. Their lending operations were made increasingly from sources other than their deposits, either long-term loans refinanced at the Central Bank or borrowing abroad, especially in the second half of the 1970s. 3.14 The long-term loans (called "development loans"), discounted at the Central Bank at highly subsidized interest rates, caused great inefficiency in the allocation of scarce credits and were also a source of quasi-fiscal deficits. Still worse, the borrowers of these loans were notoriously delinquent.10 The state-owned development banks exhibited the worst delinquency rates. The beneficiaries, especially in the agricultural areas of east Bolivia, came to believe that the loans were grants, and did not need to be repaid.'1 Access to highly subsidized credit was one main manifestation of clientelism in the old regime. It was actually a perverse instrument of wealth distribution to favor the privileged groups that captured the state. 3.15 The macroeconomic instability of 1982-85 brought significant distress to the balance sheet of the banks. After the stabilization, banks rapidly recovered deposits, almost all 9 CAO continues to be a powerful lobby. Its efforts in the recent past have been addressed to measures of debt alleviation for their highly indebted associates. '( The Central Bank was, in addition to its monetary role, a second-tier bank. It lent to the commercial banks from funds obtained abroad, normally from the IFIs or foreign govemment agencies and by printing money. " The loans to the cotton growers in the eastern region of Santa Cruz in the late 1970s epitomized this situation. 14 Bolivia and the Slowdown of the Reform Process denominated in dollars, by offering very high interest rates. They increased their lending rates pari passu with those for borrowing. The high rates rapidly deteriorated their loan portfolio, however, and by mid-1987, two years after the stabilization program, Bolivia faced a banking crisis in which four banks had to close. 3.16 Thus, a clear case for prudential regulation and more supervision appeared. Several steps were taken in that direction. First, the minimum reserve requirement for dollar time deposits was raised to 10 percent (it had been 0). Second, banks had to abide by a rule of a minimum ratio of equity capital to loans, in line with the Basle accords. Third, the supervision of banks was severed from the Central Bank and established in an independent Superintendency of Banks, accountable to the legislative branch. Later, as a component in the package of financial reform, the state-owned banks were closed. 3.17 The domestic banks are now more regulated and supervised than ever in their history. The Superintendency of Banks rapidly established its power, which helped to attenuate the banking crisis of 1987-88. Many of the financial sector reforms, in spirit if not in letter, have been incorporated in the Banking Law of 1993. This law resulted from long negotiations in Congress; the lobbying of the Association of Bolivian Banks had significant influence, but it did not change the main tenets of the policies set between 1987 and 1992. 3.18 Development credits are now inter-mediated by private financial institutions. The Central Bank auctions the credits among them. While there are some technical criticisms of the auction process, the principle that all allocation of resources should be market-based is indeed correct. The auctions, as a market mechanism, also provide a signal of fairness that contrasts with the administrative allocations used in the past. Incentive Regimes and New Investment Codes 3.19 Around 1990 the government authorities started to think that the liberalization of goods and factor markets was a necessary step to promote investment. but it was not sufficient. A new investment code and sectoral codes for the investments in minimm and oil were enacted by Congress in 1990 and 1991. 3.20 The new investment code adds few elements to the prevailing economic policy. It reiterates the guarantees for investment that are in the Constitution and the Civil and Commercial Codes. It tries to cement in legislation the reforms that had been ruled in adrninistrative acts of the executive, such as Supreme Decrees, that legally are transitory in nature. It has to be seen as a precommitment device to reduce the likelihood of policy reversals. Although it is true that laws can be changed, in the current institutional context it takes a two-thirds majority to repeal a law; the reforms undertaken since 1985 have thus been given more permanent support through the investment code. It is important to underscore that the code does not offer tax concessions, as used to be the case in previous codes. 3.21 The laws on mining and hydrocarbons are important to the attraction of foreign investment. The treatment of the property rights of the concessions was clarified, and a profits State of the Structural Reforms 15 tax, set at a very low rate, was introduced to replace the previous system of output taxes. The new mining law also allows the formation of joint ventures between the state-owned mining corporation COMIBOL and private enterprises. In hydrocarbons, the possibility was opened to extend the system of production-sharing between the state-owned oil company and foreign firms that had been in place for many years to include joint ventures--that is, to schemes of risk sharing. With these reforms, Bolivian legislation in nonrenewable natural resources is among the most liberal in Latin America. 3.22 The formation of joint ventures with COMIBOL has been far from smooth because of the opposition of the miners. They have been able to block almost all attempts. Although COMIBOL has been in disarray for many years, the government could not convince the union of miners to acquiesce to the joint ventures. The promise of very high severance payments has produced little result to date. The mishandling of the labor problem in the state- owned mines appears as a clear governance failure. Pending Reforms 3.23 Bolivia still has to undertake some important and difficult structural reforms, including the areas of privatization; the Social Security system; increasing the flexibility of the labor market; decentralization of public education and health; liberalization of the legislation on land tenure; and the independence of the Central Bank. For the sake of brevity I will concentrate on the two most controversial reforms: privatization and Social Security reforms. 3.24 In 1991 the executive obtained passage of a law of privatization for small SOEs from Congress, but no more than twenty-five of over one hundred have been privatized to date. Even if the small SOEs were insignificant in employment and investment, their privatization has met strong opposition. 3.25 The urgency of the reforms of the Social Security system cannot be underestimated. The system as it stands, with its twin features of a pension fund and medical insurance, could not be in a sorrier state. Pensions are dismally small and are paid late. Similarly, the medical services are of a very poor quality. Yet opposition to change is so strong that the authorities do not feel confident enough to send the draft laws to Congress. 3.26 The government of Sanchez de Lozada, more forcefully than its predecessor, is trying to give a new momentum to the reforms. It has two innovative proposals: capitalization and popular participation. The idea behind capitalization, in brief, is to convert the state enterprises into joint-stock companies, then increase their capital by issuing new shares and selling them to the private sector. The new shareholders will control the companies, regardless of the shares they own. Once the companies are capitalized, the government shares will be distributed, free of charge, to all adult Bolivians, in a pattern similar to that followed in some countries of Eastern Europe. The distribution of shares, however, will not be direct. They will go to the pension funds, in which all Bolivians are supposed to be enrolled. Capitalization is actually a convoluted process to circumvent opposition to privatization of the large SOEs. 3.27 Popular participation is an ambitious plan to decentralize the national government's social services to the local governments and to enhance participatory democracy. 16 Bolivia and the Slowdown of the Reform Process At first, funds will be channeled to the municipalities, which will be given the task of providing health, education, and other services of lesser importance. It is expected that municipalities will be able to raise a portion of the necessary taxes to provide for those services locally at a later time. 3.28 The capitalization and popular participation proposals have been approved in Congress and have become law. The implementation of the laws, however, is lagging. Table 1: Bolivia's Key Indicators after Reform (percent) Indicator 1986 1987 1988c 1989c 1990c 1991c 1992c Real GDP growth -2.5 2.6 3.0 2.7 2.7 4.1 2.0 Real per capita GDP growth -4.5 0.5 0.8 0.6 0.6 2.0 -0.1 Consumer price changes (Dec. to Dec) 66.0 10.7 21.5 16.6 18.0 14.5 10.5 Terms of trade index (1980 = 100) 85.5 69.6 61.1 67.8 59.2 49.8 43.4 Ml growth (Dec. to Dec.)a 82.9 39.3 30.9 -2.4 28.3 25.2 19.0 Interest rate on dollarized time deposits 15.0 15.6 15.0 15.1 13.4 10.1 11.2 Real exchange rate (December 1987 = lOO)b n.a. 100.0 85.7 85.1 127.5 127.6 125.2 Service on public and publicly guaranteed 29.6 24.7 41.9 27.7 32.8 27.3 n.a. debt as percent of exports of goods and services Percent of GDP Private fixed investment 4.7 4.0 4.2 4.1 4.3 4.3 5.6 Consolidated NF-public sector deficit 2.8 7.7 6.5 5.1 3.3 3.6 4.7 Exports + imports 32.3 29.8 29.2 35.2 39.9 34.6 33.8 BOP current account deficit 10.3 11.8 11.1 8.5 8.3 9.5 14.3 Total external debt 107.6 113.1 100.8 84.7 92.1 81.2 81.7 Net transfers on debt 3.3 2.2 1.0 3.0 -0.6 0.6 n.a. Total official developmentaid 8.3 7.4 8.9 12.1 13.3 12.1 11.8 Memo GDP at current prices (US$ millions) 3,903 4,306 4,422 4,509 4,499 5,014 5,263 Exports of goods and services 685 667 690 891 998 942 798 (US$ millions) Notes: n.a.= Not available. a Growth of Ml in domestic currency. b: Increase in index indicates increase in competitiveness. Preliminarv. Source: Author's elaboration with data from Central Bank, National Institute of Statistics. Ministry of Finance, and the World Bank. State of the Structural Reforms 17 Overall Measures of Adjustment Success 3.29 It can be observed in Table 1 that the achievement in controlling inflation (measured by percentage changes in the Consumer Price Index) is truly remarkable. The inflation rate for 1993 was 9.5 percent, and for 1994, 8 percent. Similarly, the evolution of the real exchange rate has been favorable. This result was brought about by the combination of a correct domestic policy and external developments, such as the high exchange rate appreciations in important trade partners. Bolivia is now a reasonably open economy, as shown by the (exports + imports) ratio to GDP, despite a poor export performance. 3.30 The other indicators of economic performance have been modest. Bolivia has been running very high trade deficits for several years, and their effects on the domestic economy have been attenuated only by substantial amounts of foreign aid. The contribution of gross foreign savings to the formation of capital is very substantial, possibly higher than in any other country at the same stage of development. The high BOP current account deficits are partially explained by debt service. Despite the substantial relief negotiated from the largest creditors, foreign debt continues to be very high in relation to the size of the economy. 3.31 Only since 1991 has the per capita real GDP growth reached the range of 0-2 percent. ' More troublesome than the low GDP growth rates are the very low rates of investment of the private sector, which are even lower than in some of the hyperinflation years. Several factors explain the low investment rates. 3.32 First, with a more open economy and fundamental changes in the international context and in our close neighbors, the effects of the lack of infrastructure and of skilled labor are felt more strongly than in the past, when the economy was more closed. In other words, the deficiencies in basic infrastructure and skills showed more sharply once the economy opened. 3.33 Second, Bolivia's traditional exports have faced very low prices since 1985, well below historical levels. The fall in the export prices of minerals, especially tin, and of natural gas has caused a huge drop in the terms of trade; the index in 1992 was only 43.4 percent of the 1980 index. The resulting loss of income has affected the saving rate, and therefore the investment rate. The substantial flow of foreign aid has not been able to compensate for this. Also, the low export prices have affected the decisions of foreign investors. 3.34 Third, the high real interest rates from 1986 to 1993, partially caused by the stabilization effort, inhibited the investment of the small and medium-size enterprises that can only borrow in the domestic market. Also, the high interest rates pulled entrepreneurial talent from production activities toward financial speculation or commerce with a rapid turnover of inventories. 3.35 Fourth, the taxation of enterprises in the tax law of 1986 was clearly deficient. A presumed income tax was imposed on business, with their net worth as reported in their books as the tax base. The tax penalized capital-intensive industries, while enterprises intensive in intangible assets, such as the services sector, went almost untaxed. In addition, enterprises had to pay the tax regardless of the rate of utilization of installed capacity and current profits. 12 Preliminary estimates for 1994 show a per capita real GDP growth of over 2 percent. 18 Bolivia and the Slowdown of the Reform Process 3.36 Last, there is an enduring perception that some of the reforms may be reversed, which would punish investments already undertaken and irreversible. The incomplete credibility of the reforms is based on two factors. The slow growth and low investment feeds back on the investment decisions through an expectation channel. The recurrent social unrest, caused by the high underemployment rates and the low wages, awakens fears of political instability. Investment is usually a cumulative process, where sufficient previous investment is needed to bring more investment. Otherwise, new investors fear that taxation would fall out of proportion on the capital expenses of the few that have invested. This creates a coordinating role for the government; its absence can be thought of as a shortage of governance capacity. 3.37 The qualitative effects of the reforms, however, are probably more favorable than the numbers indicate. There is no doubt that economic policy is now significantly more orderly than before. In addition, the country has developed more resilience to external shocks. One proof of this is the robustness of the anti-inflation program in the face of the severe external shocks caused by the extraordinary fall in the terms of trade, as previously mentioned. Measures of Success in Fiscal Adjustment General Measures of Fiscal Stance 3.38 The best picture of the Bolivian fiscal stance is given by the accounts of the consolidated nonfinancial public sector (NFPS). The structure of the NFPS, as well as its relation with the financial public sector, is shown in Figure 1. The budget of the consolidated NFPS merges the accounts of the general government and the SOEs. 3.39 Table 2 shows that, except in 1986, in the immediate aftermath of the stabilization program, the overall fiscal deficits (deficits of the consolidated nonfinancial public sector) have continued to be high, certainly higher than expected for a country that has stabilized inflation. In 1990 and 1991 the deficits came down to more acceptable levels, but again in 1992, and especially in 1993 (not shown in the table), there was a significant slippage, with the deficit reaching 5.8 percent of GDP. The deficits in the past two years were not as much caused by a decline in tax effort as by an expansion of expenditures, mainly for investment. 3.40 A partial explanation of the high overall deficits is given by the interest payments that continue to weigh heavily in the fiscal accounts. If the fiscal effort is measured by the primary balance, the results between 1986 and 1991 are not that worrisome. Since 1986 there have been some modest surpluses or small deficits (1987 provides an exception). It must also be mentioned that since 1986 there generally have been surpluses in the current account balance, although they were not large enough to finance the public sector's investment program. A structural (and undesirable) feature is that most capital expenditure is externally financed. State of the Structural Reforms 19 Figure 1: Structure of the National Budget I ICentral administration I I ICentral I I I Igovernment