The Neoliberal Shock and Demise of Democracy

Clearly, with the July 1990 inauguration of President Alberto Fujimori, a former rector of the National Agrarian University and a Japanese immigrant of humble origins, Peruvian politics crossed into completely uncharted territory. Just as the magnitude of the prevailing social violence and economic collapse would have been unthinkable a decade earlier, so too would the election of a candidate that had virtually no ties to the country's established political parties and criollo elite. Indeed, the coalition of traditional conservative parties that had rallied around Mario Vargas Llosa was stunned by its defeat at the hands of an outsider that had been outspent by a ratio of more than 60:1 (Grompone 1991). The final tally in the second round of voting, which gave Vargas Llosa 37.6 percent to Fujimori's 62.4 percent suggested an even more profound political realignment than had occurred in 1983. However, the partyless nature of this political shift, Fujimori's lack of clearly stated positions, and the unexpected tenacity of the newly elected president, would pose new opportunities and constraints for political economic management in the 1990s.

At face value, Fujimori drew most of his support from the country's silent majority: low income voters, small business owners, workers in the informal economy, and evangelical Protestants. He also benefited from the longstanding acrimony between Garcia and the APRA's presidential candidate, Luis Alva Castro, as Garcia began channeling clandestine support to the Fujimori campaign once the public opinion polls began to show that Alva Castro was in striking distance of placing second in the first round of voting, and hence, could face Vargas Llosa alone in the second round vote (Schmidt 1996: 341-342). In all, Fujimori captured the majority of the country's poorest districts in both the urban and rural areas, while Vargas Llosa mainly carried the middle and upper-class districts within Lima and the other major cities (Roberts and Arce 1998; Marcus-Delgado 1999). Given the trauma of hyperinflation and an all-out civil war, this electoral triumph of poor over rich, and countryside over city should perhaps not have been such a surprise. Rather, it was further confirmation of the widening chasm between the state and civil society (Crabtree 1998:20), and a wake-up call concerning the utter failure of the standing political parties to properly mediate between the two. In hindsight, at least two other key factors contributed to Fujimori's 1990 victory, and to his decade-long incumbency that followed.

The first was the transformation that had occurred within the Peruvian military over the course of the 1980s, by virtue of the increased role that the various branches of the armed forces had come to play in combating the country's guerrilla insurgencies and exploding drug trade (Mauceri 1996:136-141). Whereas President Belaunde had sought to appease the armed forces in the newly democratizing context of the early 1980s, through generous budget allocations and by eventually granting the military full latitude in the counter-insurgency effort, President Garcia pursued his security policy through the co-optation of senior officers by way of special economic incentives and promotions (Obando 1998:196-197). By the late 1980s, both of these civilian strategies toward the military---benign disengagement and selective co-optation---had faltered amidst the political and economic chaos that Garcia's misguided heterodox program had inflicted on the country.

In response, the military drew up the "Green Book," its first sweeping plan of action since the early days of the RGAF (Cameron 1997:51). But in contrast to the RGAF's nationalist-reformist project, military leaders of the late 1980s called for the implementation of market reforms with solid backing from business, government and the military, as well as "a strong government that would last 20 years if necessary" (Crabtree 1998: 21-22; Obando 1998). While Fujimori's last-minute ascendance in the 1990 race took everyone by surprise, including the military, the armed forces could not have found a better candidate to fulfill their long-term vision of Peru's transformation based on economic liberalization and tight political control. Because the president's initial base of electoral support was no more than a disperse social movement, Fujimori was quick to consolidate the military's support for his administration. By forging a partnership with the military, Fujimori was able to accomplish those tasks of economic recovery and political pacification that had eluded his immediate predecessors, and hence to prolong his tenure in office beyond anyone's expectations.

The second trend that contributed to Fujimori's longevity had to do with the deep scars that hyperinflation and civil war had left on the national psyche. In this sense, the president's eventual conquering of inflation and the capture and imprisonment of Sendero's core leadership by late 1992 went a long way toward securing his support amongst voters of all political persuasions. This was true despite the fact that Fujimori had jettisoned his earlier promises of gradual reform and the pursuit of a soft economic landing, and instead launched a stringent orthodox stabilization program just ten days into office. Similar to other market reformers in the region, the sheer magnitude of the economic crisis left the incoming Fujimori administration with little choice but to implement a serious program of macroeconomic stabilization and structural adjustment. However, in contrast to other leaders, such as Argentina's Carlos Menem and Brazil's Fernando Henrique Cardoso, Fujimori went so far as to suspend congress and the constitution in April of 1992 rather than negotiate his reform measures through the democratic channels upheld by these other civilian presidents.

Although formal democratic rule had been restored by the time of Fujimori's 1995 re-election, it was ironically these heavy-handed politics, bolstered by the economic recovery of the early 1990s and the fairly recent shadow of a tumultuous past that won the president his second term. During those economically slack periods when Fujimori's support had faltered, the president resorted to the selective distribution of subsidies and other economic incentives (Graham and Kane 1998), and to the masterful use of public opinion polls to revive his following (Conaghan 1995). To this day, there is debate as to whether Fujimori's decision to temporarily bracket democratic politics in favor of deepening market reforms and heightening the counter-insurgency campaign was one of necessity. With the military solidly in his camp, the business community in his back pocket, and an electorate willing to sacrifice short-term gain in exchange for political stability and economic growth over the long run (Stokes 1998), it appears that Peru's democratic rupture of the early 1990s was mainly a matter of personal preference on the part of the president.

While military might, a risk-averse civil society, and the president's own political cunning all go a long way toward explaining Peru's current democratic lag, less obvious are the factors that have contributed to the comparatively successful economic trends that appear in Table 6.1. As the table shows, Peru held its own along with Argentina and Chile during the 1990s in terms of its annual average rates of growth of GDP, and gross domestic investment. Similarly, Peru's annual rates of growth in employment and real wages outpaced much larger and more developed economies, such as Argentina and Mexico. As tempting as it may be to attribute this impressive turnaround to the Peruvian military's recipe of free markets and quasi-authoritarian rule, this chapter points to a quiet process of state reconstruction and institutional reform that occurred over the course of the 1990s. This explanation may seem paradoxical, given that Fujimori had become best known for his fierce independence and resistance against institutionalizing his own support base or the state's ties with civil society. However, having inherited Latin America's most volatile political economy in 1990, the president had little choice but to overhaul those state institutions which were crucial for economic recovery.

TABLE 6.1 GOES HERE

This included, for example, the renovation of longstanding financial entities like the Central Reserve Bank and the Ministry of Economy and Finance that were essential for the successful completion of the macroeconomic stabilization effort. Another key aspect of institutional reform in the early 1990s was the reinvention and/or creation of new state entities geared toward achieving the longer-term goals of economic restructuring along market lines. The purpose of these longer-term institutional goals was, first, to modernize and sustain the state's revenue stream at levels that would allow for the proper provision of essential public goods while at the same time reducing the prohibitive levels of government-held debt; and second, to rationalize the new market-based development strategy through the design of institutions that sought to guarantee property rights and to promote a competitive business environment. In essence, the buoyant trends in Table 6.1 are testimony to a pattern of "autonomous" institutional reform that has entailed the siphoning off of strategic pockets of the public sector and the management of these units as if they were private entities (Nunes and Geddes 1987; Johnson 1987; Keefer 1995).

Over time, it was precisely the autonomous nature of these state agencies that would put them at odds with the office of the executive. Which is to say, the institutional underpinnings of Peru's economic recovery were as much inadvertent as they were a purposive part of the designated economic strategy under Fujimori. Just as the RGAF was directly responsible for organizing those base groups within civil society that later emerged as the main catalyst for the transition to democracy in the late 1970s (Sanborn 1991), so too has the Fujimori administration infused new technocratic vigor and professional life into state agencies that can carry the reform project forward into the post-Fujimori era. Because this modernization process has yet to fully penetrate the sectoral ministries, the state bureaucracy at large, or the legal-juridical apparatus, the degree of institutional reform achieved thus far constitutes a necessary, but not entirely sufficient condition for more dynamic, equitable and sustainable growth in Peru. This latter set of reform tasks goes to the very heart of the challenges that the next administration faces, and to the unfinished business that remains with regard to rendering market reform and democratic politics more compatible in Peru.

The chapter begins with a review of Peru's track record in implementing the "First Phase" of market reform, which consisted of the macroeconomic stabilization and structural adjustment measures advocated by the "Washington Consensus" (Williamson 1990). A following section analyzes the process of administrative and institutional reform that was undertaken simultaneously with these first phase measures. A third section examines the extent to which Peru has advanced in the area of "Second Phase" market reforms, which include a range of initiatives meant to correct for earlier instances of market failure (e.g., rising inequality) and to increase public input and accountability with regard to the reform process. This final section also explores the phenomenon of post-reform elections in Peru, which encompasses the period from 1995-2000. Although comparative analysis shows that voters in Argentina and Chile, for example, have now passed the presidential torch to opposition coalitions who nevertheless vowed to uphold and strengthen market reforms, in Peru no such coalition has come to the fore. This chapter suggests that it will be difficult to bring essential second phase reform tasks to full fruition, not to mention Peruvian democracy, in the absence of those broader leadership coalitions and deeper institutional ties between the state and civil society that have now emerged in these neighboring countries.

The "First Phase" of Market Reform

As the analysis in the preceding chapter showed, by the end of Garcia's term Peru had basically reneged on any of the stabilization, liberalization or privatization commitments that government officials had made to the multilateral lenders in the earlier part of that decade. With the country having defaulted on service payments for its public and private external debt, it took Fujimori just ten days after his inauguration to realize that he had, in fact, zero room to maneuver in implementing a gradualist reform strategy. Fujimori's campaign platform, for example, had originally called for administered prices, a downsized but activist state, and an economic strategy geared toward fostering labor-intensive microindustries (Stokes 1996:61). But like other leaders elected at this time on similar platforms, such as Venezuela's Carlos Andres Perez and Argentina's Carlos Menem (Corrales 2000a), both of whom moved immediately in launching market shock programs, Fujimori followed suit. Hence, virtually overnight, Peru went from being the rogue state in ongoing assessments of the region's track record on market reform (World Bank 1989; Williamson 1990), to one of Latin America's model reformers.

Stabilization with a vengeance: As can be seen in Table 6.2, by 1990 the arrows on growth, investment, and inflation all moved radically in the wrong direction. Similarly, Peru's current account deficit continued to hover at an all-time high, and the overvaluation of the exchange rate exacerbated this trend. Public finances had literally collapsed under the thrust of the heterodox program, and the assessments on the damage inflicted by the country's various guerrilla insurgencies were now running in the range of US$22 billion (Manzetti 1999: 232). At the outset of the Fujimori administration more than 50 percent of the population was classified as living in poverty, and half of those fell into the category of "chronic and extreme" poverty (Sheahan 1999:108). With the gutting of those technically qualified white collar workers within the public sector work force from the late 1970s-on, and the greater infiltration of patronage appointments from the ranks of the main political parties during the 1980s, Peru's available pool of public sector expertise had also hit an all-time low.

TABLE 6.2 GOES HERE

Thus, similar to the other market reformers discussed in chapter one, once the decision was made to finally buckle down with a strict stabilization program, the policy package moved forward with little debate and just a handful of advisors. To the extent that a policy debate did occur, it was in the form of "A Plan for Economic Stabilization and Growth" that was generated by a group of the country's most talented economists in conjunction with Prof. Jeffrey Sachs at Harvard University. Published by the Brookings Institution in 1991 as a collection of essays and proposals for Peru's macroeconomic stabilization and longer-term recovery (Paredes and Sachs 1991), this distinguished group's effort to contribute to the domestic policy debate turned out to be a gesture ahead of its time. Although academic economists and private sector consultants had actively shaped the policy reform debate in Argentina, Brazil, Chile, and Mexico during the 1990s, in Peru such links between the government and the professional economics community were still comparatively weak (Conaghan 1998).

Hence, the Fujimori team took the path of least resistance and opted for the Washington Consensus blueprint based on liberalization, privatization, and deregulation that had been advocated by the multilaterals since the outbreak of the 1982 debt crisis (Gonzales 1998). It is interesting to note that the Paredes-Sachs plan also fell firmly within the Washington Consensus camp. However, the Paredes-Sachs proposal emphasized some important policy nuances, especially concerning the use of the exchange rate as a price anchor, and it raised the question of the proper sequencing of market reform measures and the need to provide an adequate social cushion for adjustment. Perhaps under the best of times, these policy nuances would have found an attentive audience within the office of the executive and the state bureaucracy. But in Peru of 1990, and in light of the urgent need to rebuild the country's ties with the multilaterals and international investors, the president opted for the reform strategy that he and his advisors deemed most likely to entice foreign capital back into Peru.

On the stabilization front, the main challenge was to rationalize macroeconomic policy making after years of haphazard management of exchange rates, interest rates, fiscal targets, public sector prices, and state borrowing. The first crucial step was to purge the country of hyperinflation, which was approached by way of a tight monetary policy, draconian spending cuts aimed at reducing public sector deficits, and the unification of the multi-tiered exchange rate. Virtually overnight, government-controlled prices and subsidies were lifted on everything from gasoline to utilities, to sugar, rice and medicines. Between August 1990 and February 1991, emergency taxes were introduced, a new managed floating exchange rate system was established around the unified rate, domestic credit had been tightened, and interest rate ceilings had basically been lifted (Paredes 1991:301).

While similar measures had brought a swift end to Bolivia's hyperinflation several years earlier (Pastor 1992), Table 6.2 shows that this was not the case with Peruvian hyperinflation. A main difference between Peru's approach and that of Bolivia's, or the inflation reduction strategy launched simultaneously in Argentina in 1991 (Pastor and Wise 1999b), was the Fujimori team's insistence on moving to a floating exchange rate system at the outset. Whereas these other programs relied on devaluation and the fixing of the exchange rate to provide a nominal anchor for price stabilization (Corden 2000), the fate of Peru's stabilization depended on the ability of the financial authorities to properly manage the new floating exchange rate regime. As Table 6.2 shows, the combined reliance on a floating exchange rate, tight credit, and market-driven prices and interest rates, created further pressures for inflation and exchange rate appreciation. The Central Bank's attempts to reverse these trends through intervention in the foreign exchange market turned out to be part of the problem: a haphazardly managed "dirty float" greatly increased the recessionary costs of the program without fully rectifying the initial problems.

In order to quicken the pace of economic adjustment Fujimori brought in Carlos Bolona as Minister of Economy and Finance in February of 1991, a Ph.D. economist with strong neoliberal credentials. As one of the few local technocrats who approximated the image of a new generation of highly trained economic ministers that had been appointed in Argentina, Chile, and Mexico during this period (Dominquez 1997), Bolona forged ahead with the implementation of deeper structural reforms and the renegotiation of the country's external debt. During the first half of 1991, tariffs on trade had been reduced to a maximum of 25 percent; the capital account of the balance of payments had been liberalized; numerous labor market regulations left over from the military regime had been eliminated; land tenure laws were amended to offer a broader scope for private initiative; the tax code was broadened and simplified; and, the sale of some twenty-three state-owned enterprises (SOE's) had been announced (Paredes 1991:313-315; Bolona 1996).

The proper timing and sequencing of such measures had been an ongoing subject of debate within the multilaterals and the academic community (Edwards and Van Wijnbergen 1985; Edwards 1990), and by 1990 the general consensus was that the adoption of structural reforms of this magnitude in the midst of the initial stabilization plan could detract from achieving the goals of the latter. On the other hand, the abrupt implementation of such reforms could readily disarm the political opposition and send a strong signal concerning the credibility and irreversibility of the government's commitments. The economic team proceeded apace, apparently convinced that the benefits of launching the entire range of structural reforms simultaneously, along with the mending of Peru's relations with its external creditors, would outweigh the potential threat to the stabilization effort.

After unilaterally resuming payments to service the debt, the government initiated negotiations for settling the US$1.85 billion in arrears it had run up with all the main multilateral lenders during the 1980s. This was a hurdle that had to be overcome before Peruvian policy makers could reschedule some US$8 billion in debt with the consortium of western country debtors known as the Paris Club, not to mention another US$9.2 billion in commercial bank debt. However, the president's decision to take matters into his own hands in April 1992---by closing the congress, suspending the constitution and dismantling the judiciary---threw a quick wrench into any of the planned debt rescheduling. The Paris Club, in particular, insisted on a credible timetable for the restoration of democratic rule as a precondition for the rescheduling of Peru's debt.

The civilian coup and its aftermath: As the economic program moved into its second year, there was still some question as to whether the particular strategy in place would be the one to stabilize inflation and trigger an economic recovery. As Table 6.2 shows, growth did not credibly resume until 1993, and inflation did not fully reside until 1995. At the same time, the various guerrilla insurgencies and violent civil-military confrontations that had endured for more than a decade still showed no signs of abating: estimates suggested that more than 26,000 lives had already been claimed. It was ostensibly this security threat that prompted the April 1992 civilian coup, and the heightened tensions between the executive and the legislature over the former's effort to further increase the military's role in eradicating the guerrillas. But Peru's democratic rupture also had do to with the president's attempt to carve out a role for himself as an almost purely "managerial executive."

Having launched his presidency with a fairly eclectic cabinet and group of advisors, Fujimori switched within the first year to surrounding himself with independents that were non-threatening and readily amenable to his point of view. This was especially so when it came to ministerial appointments. Simultaneously, the president forged an ever-stronger political alliance with the armed forces, and the National Intelligence Service (SIN) in particular. By early 1992, it had become apparent that the president had a low threshold for the usual congressional checks on executive power, which form the cornerstone of all democracies. Although both Belaunde and Garcia had taken full advantage of their constitutional privilege to legislate by executive decrees during set periods specified by congress, both also promoted their policy proposals through the majority party coalitions that each held at the time. Fujimori, in contrast, showed little interest in coalition building, policy debate, or the legislative process. Rather, even though the Fredemo bloc in congress consistently voted with Fujimori's Cambio 90 coalition to secure the executive's economic program, Fujimori went out of his way to avoid trading legislative favors and patronage within the congress.

The president's preference for ruling by executive decree was confirmed in April 1992, when he issued Legislative Decree 25418 which announced that he would do just that. The decree, which officially dissolved congress, the 1979 constitution, and the court system, also created the "Government of Emergency and National Reconstruction." Although a shock to international observers, hindsight shows that Fujimori's new blueprint for governance was very much in line with the military's political prescriptions. Moreover, the warning lights concerning this radical decision had actually been blinking for several months prior to the coup. In late 1991, for example, the president submitted a package of 126 executive decrees to the congress, just before the expiration of his constitutional authority to do so. The bulk of this package consisted of the ambitious program of structural reforms analyzed below, which congress approved.

However, congress was less willing to go along with the remaining measures, which sought to limit the freedom of the press and to greatly extend the military's power in fighting the counter-insurgency effort in the country's numerous emergency zones. These unresolved conflicts, combined with the president's rejection of social expenditure requests put forward by congress, prompted the latter to pass a February 1992 law geared toward limiting the executive's powers. Thus in April, after less than two years in office, the president showed his true political colors: a preference for military cronies, and a stronger sense of obligation to the military than to upholding democratic principles. After turning to the likes of SIN director and former army captain, Vladimiro Montesinos, and General Nicolas Hermoza Rios, head of the Joint Command of the Armed Forces, as his closest advisors, Fujimori created a quasi-authoritarian civilian-military regime that would dominate politics well into the next century.

Again, international actors did not take kindly to the notion of a coup, civilian or otherwise, and it was external pressure that most compelled Fujimori to begin going through the motions of restoring democratic norms. At this point, politics and economics entered into an even more complicated dance, which entailed three sets of elections between April 1992 and November 1993, as well as a third wave of executive decrees which sought to deepen the process of structural reform (Bolona 1996; Gonzales 1997). In November 1992 delegates were selected for a Democratic Constitutional Congress (CCD), which produced a new constitution that the country voted on by referendum exactly a year later. In the midst of the CCD, nationwide municipal elections were held in January 1993.

Interestingly, although public opinion echoed resounding support for the president and his rash actions, the vote in each of these elections was considerably more cautious. The Fujimori coalition won just over half of the eighty seats that had been put forth for the CCD, and the 1993 constitution passed by just over 52 percent of the vote. Despite the capture of Sendero's leader and mastermind, Abimael Guzman, by a small elite counter-terrorist squad in a Lima hideout in September 1992, the Fujimori forces received little bounce from this victory in the January municipal elections. In fact, after sinking like a stone in the public opinion polls, the Cambio 90 candidate for mayor of Lima withdrew from the race just prior to the election.

Thus, the neoliberal strategy was quickly pushed forward against a highly ambivalent political backdrop. The president garnered broad public support by hammering away at the ineptitude and moral corruption of the country's traditional legal institutions and political parties. But subsequent voting patterns suggest that the results of his political reform were not entirely embraced. One problem was the highly contrived rules that governed the CCD, in particular the president's unilateral stipulation that delegates to the CCD could not run for elected office for two subsequent terms. Though later rescinded, authoritarian gestures such as these did little to authenticate the CCD as an institutional conduit for the revival of democratic norms. The new constitution itself cast similar doubts, in that it expanded the president's powers to dissolve congress, to declare national states of exception, and to promote military personnel without congressional oversight (Stokes 1996:66; Gonzales 1997:44-45).

Even more telling, in terms of the president's personal ambitions, was a new constitutional clause that allowed for the executive's immediate re-election to a consecutive term, and then further re-election after the lapse of one term. Thus, as of late 1993, President Fujimori also became incumbent Fujimori, a turn of events that rendered solid economic performance and favorable public opinion ratings of the utmost importance for the president's political survival. The following sections trace the processes of structural reform, state reconstruction, and political control that together contributed to Fujimori's election to a second term in 1995.

Economic Recovery and the Deepening of Structural Reforms

After the loss of more than 20 percent of GDP from 1988-1992, the first convincing signs of economic recovery appeared in 1993. As can be seen in Table 6.2, that year marked a turning point in terms of the shift toward positive growth rates, as both private and public investment began to recuperate from the rock bottom levels to which each had fallen during the Garcia era. Moreover, for the first time in a decade, the annual inflation rate was below 50 percent. Peru's economic recovery quickly took on the features of other market reformers that had forged ahead with similarly ambitious measures. On the upside, because of the steep losses in personal income and the high levels of idle capacity that prevailed at the outset of the reform effort, there was considerable room for economic expansion.

On the downside, the combination of rapid unilateral trade liberalization, tight monetary policy, and the continued appreciation of the exchange rate threw Peru's external accounts into disequilibrium (Jimenez 1995). While capital repatriation and a burst of privatization-related FDI helped to finance the current account deficit, the external gap increased nearly threefold between 1991-1995. Peru thus joined step with other Latin American countries that found themselves in similar straits in the wake of launching deep market reforms. Policy makers in Chile, and eventually Mexico and Brazil, have moved to rectify these problems, for example, by adjusting the exchange rate in a more competitive direction and by simultaneously promoting an export-led model of growth (Wise 1999).

In contrast, the Fujimori administration stood out for its insistence on a hands-off strategy, and chose instead to assign the task of economic adjustment primarily to market forces. Although this theme holds steady across the main components of the structural reform program reviewed below, it does not detract from the integrity of this early stage of market restructuring in Peru. Rather, the problem of policy inflexibility rears its head later, when challenges from the external sector demand a more nuanced or hands-on approach to market reform, but policy responses move in the opposite direction. In other words, although the institutional bases necessary to orchestrate a cohesive and effective economic policy response were finally being laid, an increasingly dominant president and his inner circle apparently came to associate political survival with a strict adherence to the economic status quo that had been established during this first phase of market reform in Peru.

Trade liberalization: In setting the baseline for this study, chapter two (Table 2.1) showed that, with an average tariff of 25 percent in the late 1950s, Peru's economy was far more open than that of the other top five countries in the region. The reversal of this liberal trade stance over the next three decades was such that, by the end of the Garcia administration in 1990, the tariff structure had splintered into 56 different rates ranging from 10-110 percent; within this framework, 539 import items were banned outright (Rossini and Paredes 1991: 284-285). Thus, the trade regime inherited by the Fujimori administration fit the classic Latin American mold under ISI: commercial policy had largely fallen prey to those special interests that stood to lose most from the lowering of tariffs. The damage over the long run was a policy bias that favored industry over agriculture, fostered a distorted and inefficient manufacturing structure, and caused exports to stagnate. Between 1988-1990, Peru's industrial production had contracted by 30 percent (Abugattas 1998: 64), and the country's exports per capita were lower overall than they had been in the 1950s (Rossini and Paredes 1991: 285).

Despite steady pressure from the multilaterals to reduce trade barriers, policy makers had departed from this highly protectionist strategy just once---between 1978-1982---during the transition to civilian rule. Alas, more than two decades ago it had been recognized that ISI had run its course in Peru. By 1981, the average nominal tariff had been brought down to 32 percent, and 98 percent of all registered items could be imported freely (World Bank 1985: 48). On the export side, under the thrust of a depreciating exchange rate, export incentives applied through the CERTEX system, and increased participation in the Andean Pact, this period saw a fifteen-fold increase in manufactured exports (Sheahan 1999: 50). However, with the advent of the 1982 debt shocks and the gradual appreciation of the exchange rate, this brief experiment in trade liberalization came to an end. Protectionist interests again prevailed, arguing that tariff reductions had been implemented with little consultation or warning, and without taking into account the many distortions in Peru's markets for land, labor and capital (Rossini and Paredes 1991: 280-281).

In the wake of 1990's shocking hyperinflation, the Fujimori team made it clear that the options for "debating" protectionist favors with the private sector were now virtually nil. Having dispensed with one trade minister who had begged to differ, the administration moved swiftly in completing the bulk of the trade reform measures by March 1991. By this time, a dual tariff structure had been established, under which 87 percent of tariff items were subject to a 15 percent tariff and the remaining 13 percent to a 25 percent tariff (Bolona and Illescas 1997). Moreover, the web of non-tariff barriers that had evolved over the course of the 1980s was eliminated. With the exception of weapons and illicit goods, there were no bans on the import or export of any goods and services, and all taxes and subsidies on exports had been removed (Vega 1997: 44).

Ironically, whereas Peru had once set the pace in upholding protectionist trade and investment norms within the Andean Pact, it now led the way as a champion of liberalization. In this sense Peru had quickly joined step with other emerging market countries in the region, such as Chile and Mexico, where unilateral trade liberalization had similarly been embraced as a tool for both promoting macroeconomic stabilization and as a means of forcing competitive changes at the microeconomic level (Wise 1998). But unlike these other reformers, which were more aggressive in coordinating macroeconomic policy with export promotion, Peruvian policy makers remained steadfastly committed to a hands-off economic policy. While trade liberalization had promoted efficiency and competitiveness within the domestic industrial sector, exports as a percent of GDP continued to stagnate.

Fiscal shock: The data reviewed in chapter one on the Latin American public sector showed that by the mid-1990s Peru had recuperated from the outright fiscal collapse that had occurred a decade earlier. The rationalization of public finances was such that tax revenues had doubled from the depths of the Garcia administration, where they had fallen to just 7 percent of GDP, to nearly 16 percent by the end of Fujimori's first term. At the same time, public expenditures were brought more realistically in line with tax revenues: the government deficit was less than 1 percent of GDP from 1990-1994, and public debt as a percent of GDP---while still high by regional standards---had been reduced from nearly 50 percent in the mid-1980s to 32 percent in the mid-1990s. Three underlying factors contributed to Peru's fiscal overhaul, which was considered by some to be the most successful component of this first wave of structural reforms under Fujimori (Durand and Thorp 1998).

First, was the recognition, once and for all, of the essential role that fiscal adjustment plays in sustaining a macroeconomic stabilization program to the extent that it is credible at home and abroad (Arias 1991). In the wake of hyperinflation, fiscal soundness was all the more important, as policymakers and economic agents could no longer hide behind inflation in the setting of completely unrealistic revenue targets. Second, were the deep institutional changes that underpinned the tax reform that was carried out between 1991-1994, including a complete revamping of the national tax agency (SUNAT) and a strengthening of its capabilities in all areas of revenue collection (e.g., streamlining of the tax structure, creation and updating of tax rolls, technical staff training, computerization, etc.). Third, the increase in tax revenues was bolstered by the more stable base for economic recovery and higher growth that had been established as a result of the implementation of the entire package of structural reforms reviewed in this section.

As impressive as these advances have been, little inroads were made in shifting the tax structure toward greater reliance on direct taxation or taxes on income and property. Apart from the fierce political resistance by powerful economic groups to these more progressive tax categories, fiscal reform along these lines was further hampered by the very small percentage of Peruvians that were actually in the habit of paying taxes. To the government's credit, the number of income tax returns filed rose from less than 200,000 in 1991 to more than 423,444 in 1994, yet 85 percent of the revenues collected came from less than 3,000 large taxpayers (Durand and Thorp 1998: 221). Thus, while the tax structure had been simplified to just five taxes (income tax, assets tax, excise duties, a housing and urban tax, and a value-added tax or "VAT"), revenue collection depended disproportionately on indirect taxes and the tax on consumption (VAT) in particular. Peru is not unique in this respect, as the increased reliance on the more regressive VAT tax is a regionwide trend in the 1990s (ECLAC 1998: 67-71). But as the VAT came to account for 44 percent of all taxes collected by 1994, Peru's shift toward greater regressivity was compounded by pre-existing inequities that were intrinsic to the tax structure at the outset of the reform (Arias 1991: 211).

Privatization: The fiscal collapse of the late 1980s also rendered privatization inevitable at this point. With the SOE share of GDP having peaked at more than 10 percent under Garcia, operating losses amounting to more than $500 million by 1989 could no longer be sustained (Manzetti 1999: 65). While the usual suspects within the Congress and the private sector lobbied for the same gradualist approach that had thwarted all earlier attempts at privatization in Peru, this time the executive's assertive elite-level decision-making style prevailed. In late 1991, with strong backing from Finance Minister Bolona and the multilaterals, a new privatization law had been passed, and within a year the government's newly created privatization commission (COPRI) had begun to quickly unload state assets. In contrast to the footdragging and political charades that had surrounded previous efforts at privatization, COPRI staff moved forcefully in selling off some US$5 billion in state assets by the end of 1995. During Fujimori's first term, for example, seventy-two privatizations had been completed in such sectors as electricity (Edelnor, Edelsur), telecommunications (ENTEL-CPT), mining (Tintaya, Cajamarquilla), industry (Cementos Lima), and banking (Banco Continental). What explains this abrupt departure from past practices?

Certainly the liberalization of Peru's foreign investment regime and the overall economic recovery were important contributing factors, but it was the creation of COPRI and the long overdue professionalization of the privatization strategy that accounts for these inroads. First, with the creation of COPRI, the legal guidelines for privatization were finally in place, the most important being an explicit set of instructions for assessing the value of individual firms and concrete procedures for divestiture. Second, the technical and financial expertise of COPRI's staff was such that the agency actively involved itself in the restructuring of companies to make them more attractive to potential buyers. Third, apart from better preparing state firms for sale and assisting in the preparation of proposals for financial backing, COPRI started its sell-off campaign with smaller firms which are easier to unload; the strategy with the larger firms was to begin by offering stock options in a more piecemeal fashion. This latter approach represents a major difference from past efforts, which hinged on grand privatization schemes that simply never paid off.

This is not to say that privatization's enemies had been completely neutralized, or that the strategy was free of problems. Along each step of the legislative way congressional foes sought to exclude key sectors such as mining and petroleum, which set the stage for a bruising fight over the privatization of Petroperu during the 1995 presidential campaign. This, in turn, helped to sour public opinion against the government's plan to sell off every last firm in its portfolio. Other problems, such as the challenge of raising sufficient funds to restructure the large number of SOE's that were in a semi-liquid financial position, meant that potential buyers had to be guaranteed lucrative returns for some time to come. This made it all the more difficult to apply the kinds of anti-trust criteria which typically prevail in the OECD countries: by the mid-1990s, the average percentage of common shares held by the three top shareholders in Peru's ten largest non-financial privately owned domestic firms approached 60 percent, compared to 38 percent for Chile or 12 percent for the U.S. (World Bank 1998: 80). In the absence of a sufficient competitive policy to guide the process, the lack of transparency and solid financial information eased the way for new private owners to set above-market prices for public goods previously provided by the SOE's (e.g., transportation, electricity, telephones).

Financial reform: On the foreign front, financial reform meant completing the unfinished business of renegotiating Peru's outstanding debt and interest arrears with the multilaterals, the Paris Club and the commercial banks. As a result of Garcia's partial moratorium on external debt service payments, Peru's debt had grown at an average annual rate of 6.8 percent between 1985-1989, as opposed to a 2 percent annual rate of increase for the rest of the region during this same period (Larrain and Sachs 1991: 228). The magnitude of the problem was such that policy makers' hands were completely tied without the standard IMF seal of approval, and to obtain this would obviously involve settling past scores with the multilaterals. Because the statutes of the IMF and the World Bank stipulated that no new credit could be obtained from these institutions by a country that had fallen into arrears on previous loans, policy makers began repairing Peru's damaged borrower status by negotiating with the IMF through its Rights Accumulation Program (Bolona 1996: 216).

Between 1992-1995, Peru's adherence to its Rights Accumulation Agreement with the IMF allowed for an initial US$873 million long-term loan from the IMF with a three-year grace period; a bridge loan from the US Treasury and Japan's Export-Import Bank to pay back US$1.76 billion in arrears to the IMF and the World Bank; and, having cleared its arrears with the multilaterals, Peru was granted access to the Fund's Extended Facility Program 1993-1995 for balance-of-payments support, as well as another US$2.35 billion in fresh loans from the World Bank and the IDB (Manzetti 1999: 237). Thus, it took Fujimori's entire first term to restore Peru's good standing with the multilaterals. This, plus the reinstallation of democratic political procedures by 1995, then paved the way for the renegotiation of the Paris Club debt in 1996 and the rescheduling of the country's commercial debt under a Brady deal in 1997. The latter, financed jointly by the multilaterals and the Peruvian treasury, restructured US$10.6 billion in arrears on principal and interest that dated back to 1983, when the debt crisis first erupted (Kisic 1998: 51).

The remainder of the financial reforms involved a complete redefinition of the government's presence in this sector (Paredes 1991: 314; Kisic 1998: 48). A new legal framework was written that simplified the rules of the domestic financial system and deregulated financial markets (e.g., the lowering of marginal reserve requirements, the liberalization of interest rates and elimination of ceilings on rates, and a reduction of the tax on debits). At the same time, the capital account of the balance-of-payments was liberalized, and this included the opening of the banking system to foreign interests and the freeing of foreign exchange transactions. These measures, along with the launching of the privatization program and the Foreign Investment Promotion Act of 1991, contributed to the development of the Lima stock exchange. Having long been hampered by poor transparency, a burdensome regulatory backdrop, and the lack of modern financial instruments, the combined effects of financial sector reform helped to increase the capitalization of the Lima stock exchange from just US$800 million in 1990 to US$19.5 billion in 1997 (Manzetti 1999: 275).

Social compensation: Up until this point in the analysis, Peru's social trajectory had been one of increasing polarization of income against the backdrop of periodic improvements in poverty reduction and real wages, with the hyperinflation of the late 1980s wiping out most of these gains. For example, by July 1990 working Peruvians were earning just half of what they had earned in July 1985, and in the wake of the August 1990 adjustment package this purchasing power was again cut in half (Gonzales 1993). As noted earlier, the social fallout from the Garcia period was a welfare deficit that had exploded into crisis proportions. The unmet basic needs of the population---including everything from food and housing to education and health care---were such that 55 percent of all households had fallen beneath the poverty line by 1991, compared to 38 percent in 1985 (Sheahan 1999: 108). Given that much of the violence over the preceding decade had been poverty-inspired, the multilaterals strongly encouraged the Fujimori administration to implement a safety net scheme to help offset the burden of further adjustment.

After a first year of false starts, the president side-stepped the relevant line ministries that had long proved inadequate in delivering basic public goods and created the National Fund for Development and Social Compensation (FONCODES) in August 1991. Similar to compensatory social programs that had been implemented simultaneously with sweeping market reforms in Bolivia, Chile, and Mexico, FONCODES relied on a combination of traditional social welfare relief and on new demand-based criteria requiring that communities generate specific proposals for assistance (Graham 1994). Through FONCODES the administration targeted desperately

needed compensatory resources toward poor communities, the bulk of which went toward economic infrastructure (road maintenance projects, irrigation, deforestation) and social infrastructure (health facilities, schools, basic housing). As with the other safety net schemes being implemented in Latin America at this time, FONCODES was designed to temporarily alleviate poverty and thus to provide some political breathing space for sustaining market adjustment. With FONCODES' per capita expenditure averaging just US$12 from 1991-1994 (Graham and Kane 1998: 86), the program was never meant to resolve longstanding structural inequities in Peru.

Fujimori's simultaneous accomplishment of these two goals, increased political capital and short-term poverty relief, comes through clearly in the process of implementing FONCODES. Especially after the 1993 constitutional referendum, where the president lost in all departments outside of Lima, FONCODES became a main venue for channeling public resources to those regional districts where the government had fared particularly poorly at the polls. As many of these districts had long been neglected by the central government, one result of the president's politically driven allocation of social resources was the direction of scarce funds to regions with extremely weak social indicators. This image of greater inclusion at the local level was bolstered by an overall increase in social expenditures, from 16 percent of government spending in 1990 to 40 percent in 1995 (Sheahan 1999: 125). This, in turn, helped deliver Fujimori's victory in all but one regional department (Loreto) in the 1995 presidential election. However, this fortuitous mix of politics and temporary social relief also distracted from the need to formulate a cohesive long-term strategy for poverty reduction and income distribution in Peru. By 1994, 48 percent of the population was still living below the poverty line (Sheahan 1999: 108), and "for the average household, 1997 incomes were lower than those of 1975" (Webb 2000: 280) were.

Reinventing the State

By definition, the sheer magnitude of market measures that had been introduced during Fujimori's first term implied a fundamental shift in the balance of public and private influence over the economy. One way of evaluating the changes that had occurred is to revisit the two main arguments that were introduced in chapter one and that have been woven throughout the last four chapters. The first argument concerned a cluster of trends related to state-led development, which had become cumulatively more problematic since Peru's embracement of a developmentalist strategy in the early 1960s. In brief, this consisted of: 1) an over-reliance on external borrowing, which allowed for the chronic mismanagement of key macroeconomic policies; 2) an increasing dependence on SOE's to carry out the state's development tasks, but against the backdrop of weak administrative structures; 3) the cultivation of an ambiguous relationship with the private sector, which claimed to lack confidence due to the state's regulatory intrusions and poor macroeconomic skills; and, 4) heightened poverty and inequality, due to the failure over time to assign priority to distributional policies as an integral part of the developmentalist model.

As the data in Tables 6.1 and 6.2 suggest, by the mid-to-late 1990s the ambitious market reform effort had worked to reverse these trends on all but the distributional front. The destructive link between external borrowing and reckless macroeconomic management had finally been broken: in the wake of the 1997 Brady debt restructuring deal, Peru's outstanding external debt had been renegotiated and reduced and inflation had stabilized to single digit levels. As for the SOE sector, the privatization strategy had been revamped to the extent that nearly 190 privatization transactions had been completed by 1998, bringing in around US$8.7 billion in revenues. Moreover, gone were the days of the big "white elephant" public investment projects, as the private investment share of GDP nearly doubled over the course of the 1990s and the privatization process itself generated commitments for another US$7 billion in private project investment (see Table 6.3). This remarkable improvement in macroeconomic policymaking, along with the construction of a modern regulatory framework to guarantee property rights, helped to reduce the wariness between public and private actors that had prevailed since the military's state capitalist experiment (Durand 1998: 271-276).

TABLE 6.3 GOES HERE

The second argument from chapter one concerns the nature of the institutional backdrop that frames the economic policy making process. Four main institutional variables were introduced---bureaucratic autonomy, the modernization of state economic and planning agencies, the stability and coherence of the leadership coalition, and the extent to which predictable rules of the game have been established for the intermediation of economic policies through peak organizations that are recognized by the state. The remainder of this section explores the ways in which institutional reform has advanced in Peru since 1990, particularly with regard to the first two variables just mentioned. The analysis identifies three main patterns of institutional change: the streamlining of the Peruvian state during the Fujimori era; the reconstruction and renovation of the country's main economic institutions; and the increased reliance on bureaucratic autonomy in the 1990s through the creation or overhaul of numerous state agencies involved in revenue collection, regulatory oversight, and the delivery of essential public services.

Again, the underlying argument here is that, while laying the necessary groundwork for the impressive economic turnaround of the 1990s, the confinement of institutional reform to the state's internal organizations will not be a sufficient condition for sustaining the economic success of Peru's market model. Rather, as the recent experiences of Argentina, Chile, and Mexico have shown, reform sustainability requires broader coalitional support and the strengthening of institutional ties between the state and those civic organizations that represent the ultimate stakeholders in the reform process. In all three countries, the widening commitment to sustain market reform has become a main focal point around which domestic politics have become more open and competitive. At the same time, citizens' demands for greater accountability and better results have heightened. While difficult to imagine even a decade ago, open economies and competitive politics are now chasing each other in a virtuous circle in these other emerging market countries. But not so in Peru, which since 1992 has remained an outlier case in the sense that the leadership coalition has virtually closed ranks and the executive has gone to great lengths to avoid forging more cohesive and reciprocal ties between the state and civil society.

Streamlining of the Peruvian state: In Peru, "streamlining" has entailed both a haphazard retrenchment of the public sector which began under Garcia's watch, and a more purposeful restructuring of the Peruvian state after 1990 as fiscal reforms and the privatization program were launched. On the side of haphazard cuts was the steep compression of public spending from a peak of US$1,059 per capita in 1975 to just US$178 by 1990, a decline of 83 percent (Webb 1991: 2-3). Between 1987-1990 the public payroll contracted by 75 percent, but the burden of adjustment fell mainly on real wages as very little workforce downsizing actually occurred. More than ever before, the public sector offered de facto employment relief for a rapidly shrinking middle and lower-middle class. In the mean time, the state's capacity to supply the most basic public services was drastically reduced. Nowhere was this more apparent than within the SOE sector, which on the eve of the privatization drive was running an astonishing US$2.5 billion in annual losses (Manzetti 1999: 248).

Obviously, the more purposeful effort at streamlining in the early 1990s would require the rationalization of the public sector, but also the revival of the state's ability to provide an acceptable level of public goods. On the side of rationalization, the greatest strides were made in the SOE sector, where close to half of the state's 250 assets had been sold by 1998 (Manzetti 1999: 262-264), and the government had met 85 percent or more of its privatization goals in such sectors as telecommunications, finance, industry, and fisheries. Similarly, a handful of state development banks were either sold or closed down, as was the Foreign Trade Institute and the National Planning Institute---the very symbol of populism in Peru. Despite the announcement of further cuts in public sector employment and the intention to consolidate and/or seriously pare down the central government ministries (Bolona 1996), these other aspects of streamlining have yet to take off. In 1996 the government did propose a new "Program to Modernize the Public Administration," to be run out of the Ministry of Economy and Finance (MEF), which ostensibly would have tackled these tasks. Yet, in the end the president was not willing to expend the political capital that would be required to simultaneously downsize and reform the public bureaucracy. As a result, the program died in less than a year.

Although the SOE workforce was cut from 140,000 to 50,000 between 1991-1994 (Manzetti 1999: 262), Table 6.4 shows that over the course of the decade total public sector employment held steady as central government layoffs were offset by hiring at the regional level. Likewise, the organizational contours of the central government have not changed substantively since the rapid expansion that occurred post-1968. As can been seen in Table 6.5, there are currently sixteen line ministries plus a council of ministers, basically the same structure that has prevailed since the Velasco era. Some minor changes have been initiated, such as the merging of the Ministry of Housing and Construction with the Ministry of Transportation and Communications (Bolona 1996: 243), and the Ministry of Energy and Mines, in particular, has been overhauled, downsized, and made more efficient.

TABLES 6.4, 6.5 GO HERE

The two main departures from the past include, first, the designation of the Ministry of the Presidency as a super-ministry in 1992, along with the Ministry of Economy and Finance (MEF). Throughout the 1990s, these two ministries, plus the Ministry of Defense, accounted for the mainstay of central government expenditures. A second overriding change has been the cultivation of numerous enclave or autonomous agencies involved mainly in revenue-generation, service delivery, and regulatory oversight. These public entities stand apart from the national budget, and are instead financed through separate fund (FONAFE) located within the MEF that the agencies themselves generate through user fees and other service charges. As Table 6.5 shows, these FONAFE-financed entities staff far fewer employees, and the average remuneration is 2-6 times higher than the salaries paid by the central government ministries. It should come as no surprise, then, that this leaner and more modernized strand of the Peruvian state has been associated with highly professional service delivery in the 1990s (consumer protection, revenue collection, market regulation), while the services provided by the traditional line ministries (education, health, infrastructure) continue to lag far behind.

Renovation of state economic institutions: As seen in the previous chapter, Peru's two main economic institutions, the Central Reserve Bank (BCRP) and the Ministry of Economy and Finance (MEF), had become mere shells of their former selves by 1990. The supposedly autonomous BCRP had lost all credibility, most obviously because of its failure to fight off hyperinflation, but also because the Garcia administration had rotated the Bank's directorship four times in five years---a reflection of the extent to which the Bank had been overrun by party politics and misguided executive ambition. However, once Fujimori and his advisors had made the decision to attack hyperinflation by launching a market shock program, the renovation of these two key economic policymaking entities quickly became an essential part of this effort. Overall, Peru's public administration suffered from four main shortcomings when the Fujimori team was handed the reins of government, and these two institutions were no exception.

First, the very structure of government, in terms of the degree of centralization, autonomy, and the functions performed across agencies, was completely unproductive and inefficient. Second, there was little financial accountability with regard to how public resources were dispersed and few checks to discourage the diversion of state funds for illicit use. Third, personnel policies worked against the modernization of the public sector, as salaries were far too low to retain employees with the proper skills, and civil service tenure offered few incentives for improved worker performance or career advancement. And fourth, there were no mechanisms to deter officials from engaging in arbitrary or capricious behavior, and few significant consequences when they did. From the analysis thus far it is clear that the Fujimori administration's track record in addressing these shortcomings was selective and uneven. But again, given their immediate importance for stabilizing the economy and sustaining market reforms, the BCRP and MEF were overhauled in ways that directly addressed these weak spots.

With regard to the BCRP, important legal steps were taken to restore the Bank's autonomy from domestic political intrusions and to reinforce its overriding mandate to preserve monetary stability (Velarde and Rodriguez 1998; Kim 1999). The 1993 Constitution did retain the same seven-member Board of Directors (three appointed by congress and four by the executive, including the Bank's president), as well as the stipulation that the BCRP president serves a five-year term parallel to that of the Peruvian executive. Although the recent historical record suggests that a staggered term between the Peruvian executive and the BCRP president would be more likely to promote central bank autonomy, Fujimori saw this differently. Nevertheless, according to the new constitution, neither the BCRP president nor the board of directors could be removed at the whim of the executive, as this now required a 2/3 congressional vote triggered by evidence that the parties concerned had failed to uphold the Bank's monetary policy mandate. Other important BCRP reforms concerned the establishment of a budgeting system separate from the national budget, with the General Comptroller and the Superintendency of Banking and Insurance (SBS) designated as the overseers of the BCRP's financial management, and the application of private sector labor law to all BCRP personnel (Kim 1999).

The MEF was similarly strengthened in ways that both supported the short-term goals of macroeconomic stabilization, and the longer-run objective of carving out a new role for government as that of market regulator and facilitator of private initiative. As one of two "super ministries," the MEF on average has accounted for approximately 30 percent of all central government expenditures over the past decade. The agency has been staffed with a highly qualified upper cadre of managers and analysts since 1991. Although central government salaries, including those at the MEF, have remained low compared with those paid by the autonomous agencies, administrative talent has been retained at the MEF through multilateral financial assistance (e.g., the United Nations Development Programme) that supports a more competitive salary structure at the top.

Since the early 1990s, the MEF has taken the lead in reforming public financial management in Peru (Shepherd 2000). By 1997, the budget was prepared on a programmatic basis, and a new budget framework law had introduced greater flexibility and transparency (Peru's national budget is now published on the Internet). Substantial progress had also been made in shifting to an Integrated Financial Management System, which means that all public funds are due to pass through the same budget. As the manager of the separate FONAFE fund through which a number of the autonomous agencies are financed, the MEF has promoted the delivery of higher-quality public services by applying centralized accountability mechanisms and contractual performance indicators to these agencies.

A main challenge now is for the MEF to take a stronger lead in expanding the scope and depth of financial, administrative, and personnel reform throughout the line ministries. On this count, the MEF's failure to advance the aforementioned Program to Modernize the Public Administration is telling. The initial objectives of the program, "to rationalize, clarify, and consolidate the activities of the ministries, to deregulate administrative and procurement rules, to introduce a performance measurement system, and to allow ministries to apply private-sector law" (Shepherd 2000: 6), are every bit as relevant today as when they were proposed by the government in 1996. The fact that this more penetrating reform proposal was so easily defeated, by legitimate fears about the negative ramifications for labor-downsizing but also by executive refusal to confront entrenched bureaucratic politics throughout the central government, suggests the limits to further institutional reform in the absence of assertive executive leadership.

The other designated super ministry, the Ministry of the Presidency, had been created by the Garcia administration in 1986 as a main venue for promoting miscellaneous APRA projects and then deactivated as that party exited office in 1990 (Marcenaro 1996). The ministry was resurrected by legislative decree immediately following the 1992 civilian coup, and went on to quickly capture around 23 percent of the central government budget by 1995 (Graham and Kane 1998: 85). Since its revival, the ministry's formal objective has been to promote a sweeping set of social programs: social development (health, food assistance), infrastructure (schools, housing, water purification), and regional development. Informally, it has served as the much larger vehicle through which the president has secured political support by channeling public resources to districts where voters have strayed. Organized around three vice-ministers who oversee each of these functions across nearly twenty sub-programs, the lion's share of expenditures are captured by just one program within each vice-minister's bureaucratic domain. Within the social development program FONCODES dominates the budget, under infrastructure INFES (a school-building program) is the main focal point, and in the regional development category INADE (the National Development Institute) leads the way.

No matter that each of these functional areas already occupied budget lines in various other central government ministries when the Ministry of the Presidency was brought back to life. Like his predecessor, Fujimori sought an institutional locus for an ambitious social program that he could personally take credit for at election time, and for which there would be little budgetary oversight or procedural accountability. This he clearly found in the Ministry of the Presidency, which even Fujimori's most polite critics refer to as the president's own "mafia." The use of this ministry as a main conduit for a more targeted social policy has been indisputably successful for accomplishing the twin goals of re-election and short-term poverty relief. However, this arrangement has only reinforced the inefficiencies and indiscretions that continue to plague Peru's central government structure. It has also worked to postpone debate over the urgent need for a longer-term social policy that would tackle inequality through improved access to a much better quality of education and healthcare services.

Bureaucratic autonomy: Ever since the political and economic meltdown of the 1960s, successive administrations in Peru have recognized the need to shield policy makers from undue populist and clientelistic pressures. Until 1990, this realization had been consistently manifested through the insulation of the executive and small elite-level working groups around the president. Except for a brief interlude during the early 1970s, insufficient effort had been made to nurture strategic sectors of the state bureaucracy to carry out executive policy preferences. This changed dramatically during the Fujimori era, as the degree to which autonomous public agencies have been established and utilized effectively in Peru is perhaps unparalleled in the region (Shepherd 2000: 5). The political impetus for this partial transformation of the public sector remains somewhat of a puzzle, particularly in light of Fujimori's avowedly anti-institutional stance on all other matters, not to mention the very half-hearted efforts that have surrounded the reform of the line ministries and bureaucracy at large.

The literature on bureaucratic delegation, which analyzes this phenomenon from the standpoint of both the OECD bloc and the developing countries, offers three insights with regard to this strong reliance on autonomous agencies in Peru. First, apart from the chaotic economic backdrop against which Fujimori was elected, this was the first civilian administration since 1963 that did not command a majority vote in the Peruvian Congress. This is the opposite scenario that awaited other civilian presidents who had inherited similarly daunting reform challenges in the early 1990s, such as President Carlos Salinas in Mexico and President Carlos Menem in Argentina. Relatively secure about their support bases and tenure in office, the impulse of these other executives was to delay in delegating authority to independent public entities such as central banks (Mexico) or regulatory boards (Argentina) for fear that such bureaucratic reforms would diminish their political control over the policy making process. In contrast, Fujimori, with no real political party to call his own, and no guarantee of a "natural" legislative coalition to back him in Congress, pursued the autonomous agency route as the only certain means for controlling his market reform agenda.

Hence, although market reforms were quickly perceived as the only viable option when Fujimori first took the helm of government, policymakers faced the very pragmatic question of where to start in implementing them. The Peruvian bureaucracy may have been completely decimated by 1990, but it was still enshrouded in endless procedural rules and red tape. Thus, a second insight from the literature concerns the recognition that market reforms will flounder in the absence of a proper institutional base. As Philip Keefer (1995: 25) has observed, the potential benefit of autonomy is "that it insulates agencies from the influence of different government entities that attempt to use their oversight capacity to distort agency decisions in favor of narrow interests. The agency is relieved of the burden to balance every technical decision against the parochial concerns of a multitude of entities with oversight responsibilities." In a very real sense, then, the cultivation of a cluster of autonomous agencies to act as the standard-bearer for implementing and sustaining market reforms became part and parcel of the reforms themselves.

A third insight on bureaucratic insulation concerns policy efficacy. As voters in Peru have virtually abandoned party affiliation and instead cast their ballots according to the government's ability to visibly improve the delivery of essential public services (Stokes 1997; Graham and Kane 1998), the more efficient outputs from autonomous agencies have been a main factor in enhancing Fujimori's electoral prospects. In short, they have enabled the president to vastly improve the productivity and efficiency of fiscal spending, while simultaneously attending to the needs of key constituencies. There is, however, a distinct downside to this new link between political survival and the development of autonomous public agencies in Peru.

For example, in other bureaucratic settings autonomous entities have more commonly been used as the surest way to render policy reforms irreversible (Boylan 2000: 5-7). As such, they tend to be governed by commissions comprised of public and private representatives who serve staggered terms and are appointed by congress (Evans 1995; Keefer 1995). Not so in Peru, where autonomous agencies are under the direct control of the president, who can hire and fire agency staff at will. Thus, although the president has delegated authority to these autonomous agencies as a means of guaranteeing the success of his own policy goals, he has done it in such a way that renders them "relatively easy to create and easy to disable" (Shepherd 2000: 7). While tremendous strides have been made in the way of efficiency, transparency and service delivery, the future viability of this autonomous agency approach will depend heavily on the extent to which they can be integrated into a broader legislative and managerial framework.

In sum, as Table 6.6 shows, Peru's autonomous agencies can be roughly grouped into four categories: financial regulation, tax administration, other miscellaneous regulatory bodies, and new independent agencies which relate directly to the goals of sustaining market reform. Despite the diversity of these activities, the independent agencies share the mandate to manage economic change at the point of service delivery (Keefer 1995; Wilkins 1999). As such, they operate with varying degrees of managerial autonomy and are not subject to the normal budgetary process through which line ministries obtain resources. In contrast with the central government ministries, these autonomous agencies are subject to clear performance requirements and bottom-line limits. At the end of the day, however, they are still under the president's direct control.

TABLE 6.6 GOES HERE

"Second Phase" Reform, First Phase Politics

By the end of Fujimori's first term it is safe to say that the economic rules of the game, in terms of macroeconomic behavior, the dismantling of the institutional bases of statism and protectionism, and everyday business practices in Peru, had been radically transformed. In other words, a first generation of market reforms had basically been completed against a political-administrative backdrop that resonated with the management strategies adopted by other market reformers in the region during this time: "presidents and technocratic economic cabinets were able to design and implement changes in macroeconomic rules with relatively little interference from the rest of the political system or the public sector" (Naim 1994: 35). Having implemented a somewhat generic "Washington Consensus" market reform program in the early 1990s, political leaders and policy makers in Latin America began to turn their attention to the consolidation of these reforms from the mid-1990s on. While all market reformers faced continuing challenges in the areas of macroeconomic management, income distribution and the modernization of state organizations, the more specific content of "second phase" reforms varied according to a given country's political, social, and institutional legacies.

For example, by the mid-1990s one of Argentina's most pressing second phase challenges was the deregulation of domestic labor markets, while in Mexico the 1994 peso crisis had driven home the urgent need for a sweeping reform of the banking sector (Pastor and Wise 1999a; Kessler 2000). For Brazil, the rationalization of fiscal policy had been delayed to the point where this was the main trigger in setting off a massive run on the exchange rate in January 1999 (Cardoso 2000). While the specific tasks may vary, they present quite similar challenges in terms of the complexities and difficulties of following through on second phase reforms. First, because the gains from the policies now required (civil service reform, administration of justice) are subtler than the obvious benefits of macroeconomic stabilization, while the pain (downsizing, loss of power and access to patronage) is more concentrated. Second, apart from the need to involve a much larger chunk of the central government in the reform process (especially the ministries of education and health), the implementation of this next round of reform requires debate and negotiation with those most affected. In other words, second phase market reforms require a more inclusive and accountable style of politics.

For Peru, it is the failure to shift to a more open and participatory mode of politics that constitutes the most glaring reform gap, as a highly insulated executive and an ultra-elite decision-making style have been frozen in place for an entire decade. Again, this contrasts with the political trajectory of those market reformers mentioned above, where patterns of executive autonomy and bureaucratic insulation that characterized the implementation of market reforms have given way to much greater levels of political competition and accountability in the consolidation phase. This is certainly true in the cases of Argentina, Brazil, and Chile, as attested to in the most recent round of national elections that took place during 1998-1999 in all three countries; even in semi-authoritarian Mexico, the seventy-one year political grip of the single-ruling PRI party was finally broken by the opposition's victory in the July 2000 presidential elections. There, as elsewhere, the combination of deep market restructuring and internal state reform laid the groundwork for a political opening that not even the power-hungry PRI could prevent.

There is no reason to expect that the same political transition could not eventually occur in Peru, but as with the other phases of political economic development analyzed in this book, it appears that the country will continue to march forward at a beat that differs from that of its Latin American neighbors. Below, I examine the ways in which state-society relations have evolved under Fujimori's long tenure, with an eye toward identifying both the constraints and the possibilities for Peru's transition to the same competitive political mode that has taken root in other countries now in the midst of consolidating second phase market reforms.

The leadership coalition: As the dust began to settle on the mass of market measures that had been implemented during Fujimori's first couple of years, it became apparent that the post-reform leadership coalition looked nothing like the elite party-controlled cliques that had ruled the country from 1980-1990. While the executive's style of governing Peru continued in the same insulated and autocratic vein, his two main allies---domestic business and the military---had not sustained a presence within the upper echelons of government since the return to civilian rule in 1980. Hindsight shows that the role of each in the leadership coalition lent credibility to the Fujimori administration at the very moment when public confidence in national leaders had sunk to an all-time low. However, as the 1990s wore on, the very narrowness of the leadership coalition, and the president's intent on indefinitely prolonging his own incumbency, tipped the balance of power uncomfortably on the side of the military. Not only did this disproportionate influence of the military puncture any myths about the depth of Peru's democratic transition, just as importantly, the lack of accountability and proper legislative oversight did not bode well for the ability of the ongoing Fujimori administration to advance second phase market reforms.

How is it that an executive like Fujimori, with no mentionable ties to the business community at the outset of his presidency, was able to forge a viable partnership with the Peruvian private sector when all of the other attempts reviewed in this study had failed? This remarkable alliance must be understood from the standpoint of Fujimori's own pragmatism, as well as the institutional strengthening and internal reform that occurred within the Peruvian business community during the early 1990s. The trend toward modernization of key state agencies must also be factored in here, as this provided the private sector with unprecedented guarantees in the way of property rights and policy continuity. With regard to the president's role in this alliance, once the decision had been made to proceed with a market shock program back in 1990, Fujimori quickly grasped the need to bring the domestic private sector on board. In contrast to Belaunde's constant waffling on market measures, or Garcia's highly erratic behavior when it came to spurring private initiative, Fujimori toed a steady line in the announcement and implementation of market reforms.

Whereas the 1990-1992 congress had been put off by this managerial executive-style, and the aloofness it implied, the private sector found some comfort in it. This is not to say that business was enamored with the entire program, but the consistency and clarity of the new economic rules, and the positive signals from higher growth and lower inflation helped forge a viable working coalition. Even before his inauguration in 1990, Fujimori began meeting with the representatives from CONFIEP, the private sector's umbrella organization, and during his entire tenure as president gave the closing speech at the annual conferences of CONFIEP and CADE---the yearly meeting of business executives (Durand 1998: 270-272). Through the course of the 1990s, the president also appointed prominent business leaders to direct FONCODES and to key ministerial posts, for example, at MEF, the Ministry of Industry, and the Ministry of Foreign Affairs. The Fujimori team was also able to move past the collective action gridlock that had long plagued government-business relations by offering some compensatory relief (labor market flexibilization, lower taxes, political access) to domestic entrepreneurs, while still adhering to the tight fiscal and monetary targets that had been agreed upon with the IMF.

For the private sector's part, domestic business had come a very long way from the days when bank owners chained themselves to their desks in protest of Garcia's 1987 attempt to nationalize the country's financial institutions. This, in fact, turned out to be a catalyzing moment for the CONFIEP, as it marked the point at which a highly heterogeneous set of domestic business interests began to set aside their differences and unite to become a peak organization proper. Since the Garcia fiasco, CONFIEP has attracted generous external support to the tune of US$2 million annually from USAID and the multilaterals, which over the years has helped to finance office space, support staff, conferences, and publications. Since the mid-1990s CONFIEP has also used its multilateral support to create a business think tank (IPE) and a Productivity Institute, both of which have generated studies to better inform policy debates and to clarify the business sector's position on specific issues.

This external support has served CONFIEP well in several respects. First, these resources contributed directly to the modernization of CONFIEP, in terms of broadening its membership base and professionalizing its internal rules and procedures for sectoral representation. For instance, CONFIEP recognized the need to bring small businesses into its fold, as this sector generates the bulk of employment and constitutes by far the largest group of firms in Peru. But because it has always suffered from informality, poor credit access, and weak representation, it is also the most susceptible to government co-optation. Finally recognizing that there is greater strength in numbers, the Confederation of Small Businesses (CONAMYPE) was formed and officially incorporated into CONFIEP in 1995 (Durand 1998: 268). At the same time, CONFIEP's presidential term was expanded from one to two years so as to maximize on a given leader's learning curve, and the committee rules were reformed to encourage policy consensus as opposed to policy vetoes. Second, although CONFIEP is still the locus of considerable conflict within and between different economic sectors, it has matured to the extent that it is has become an effective venue for mediating and arbitrating conflicts of interest between its members. Because of these various changes, business interests now play an assertive role in the policy process, rather than the mainly reactive stance of the 1980s.

In many respects, the Peruvian private sector has turned the same corner as other business groups across the region. As market reforms were delivered as a fait accompli in the early 1990s, private actors have had little choice but to adjust. In the process, economic policy debates have shifted from the state-versus-market mentality that prevailed in Peru up until the Garcia period, to debates over how to better adapt to more intense patterns of competition. But like their Latin American counterparts, Peru's business class is fully aware that market reforms have now cast them as the main economic protagonists, and that their leverage over the government has therefore increased. For the most part, the Fujimori administration has delivered on the kinds of reforms that matter most to the private sector (privatization, the liberalization of financial markets and freeing of restrictions on FDI), and business reciprocated by supporting the 1992 coup and the president's bid to run for re-election in 1995.

But this support was offered as a quid pro quo for a timely reinstallation of congress and the judiciary, and for the revival of rotating elections and constitutional rule. By the late 1990s, government-business relations were once again strained by the president's failure to fully follow through on the promised democratic transition. In particular, Fujimori's questionable bid to run for yet another five-year term in 2000 led to executive actions that impinged directly on the business community. First, because of the president's growing tendency to retaliate against those who questioned or opposed him, for example, in the form of tax harassment or costly state-sponsored litigation; and second, because the external sector reacted to Fujimori's questionable electoral ambitions in 2000 as it had to the 1992 coup---with threats of economic sanctions and suspension of multilateral support, which understandably invoked bad private sector memories from the 1980s.

As for the military's strong presence in Fujimori's leadership coalition, up until late 1992 the military's high profile in national politics could perhaps be justified by the demands of quelling a long-running guerrilla insurgency and the drug trafficking that had become closely linked to the country's various guerrilla factions (Alvarez 1998: 123-124). But the president's continued reliance on military advisors like Montesinos and Hermoza Rios for the remainder of the decade was both a reflection of his own unease with the give and take of competitive politics, and the utter opportunism of those advising him. Both Montesinos and Hermoza Rios had come forth when a very politically inexperienced Fujimori took command in mid-1990, and once the guerrilla insurgencies and narcotics trade had been deftly brought under control, they were there for the president at each subsequent critical juncture: the pre-coup battles with congress and the 1992 coup itself; the president's 1995 re-election campaign; Peru's 1995-1998 border conflict with Ecuador; the resolution of the 1996-97 hostage crisis at the Japanese Embassy in Lima; and, a prompt response to renewed damage from floods caused by El Nino in 1998.

Yet, the flip side of this "loyalty" to the president was the increased politicization of the Peruvian military under Joint Chief Commander Hermoza Rios, with SIN director Montesinos pulling most of the strings behind the scenes. In this sense, the Fujimori team picked up on Garcia's earlier strategy of co-opting the military and carried it to new extremes. On the one hand, the military was granted greater autonomy in battling guerrillas and drug traffickers, and despite the escalation in human rights abuses and extra-legal deaths at the military's hands, very few officers were brought to justice; on the other hand, the president had issued a new military law in November of 1991 which eliminated the time-honored system of merit-based promotion and rotating appointments that had earned the Peruvian armed forces a reputation as one of the region's more professional military institutions. Whereas the previous system called for the rotation of the Joint Chief Commander each year, based on seniority and a transparent set of performance criteria, Fujimori installed the mediocre but loyal General Hermoza Rios as Joint Chief Commander and simply renewed his contract every year until 1998.

As the 1990s wore on, those military personnel that displayed loyalty to the Fujimori-Montesinos-Hermoza Rios triumvirate were rewarded with job security, promotions and pay raises. Those that did not fell prey to threats of dismissal and blackmail, based on wiretaps and other forms of information gathering that the SIN director had developed as a counter-insurgency expert under Garcia (Obando 1998; Bowen 2000:62-68). It was this new perverse system of incentives and disincentives that prompted a failed coup attempt against Fujimori in November 1992, as those officers that prided themselves on military professionalism sought to regain control before the Fujimori team had completely gutted the armed forces of all institutional integrity. The triumph of Montesinos and Hermoza Rios, not to mention the president's eagerness to retain their support, once again put Peru at odds with the rest of the region.

In contrast to other civilian leaders, who had sent the armed forces back to the barracks and worked hard to keep them there, Fujimori actively sought the military's entry into his leadership coalition (Hunter 1997), and empowered his military colleagues in ways that further delayed the country's democratic transition. It is against this coalitional backdrop that Peru became suspended in what I referred to above as an insulated political mode that roughly correlated with the implementation of first phase market reforms in Latin America; obviously, the shift to a more inclusive politics demanded by second phase market reforms would require that the military drop back from its high profile in national politics, which finally began to occur in the aftermath of the tumultuous 2000 presidential election.

The intermediation of societal interests: Throughout this study, the intermediation of societal interests in Peru has been discussed along two axes--the state-capital-labor axis and the party-congress-executive axis. On the first count, except for the military's anomalous presence in too many facets of everyday life, interest intermediation in Peru of the 1990s has conformed to a regional trend. Like other elected leaders in Latin America, Fujimori constructed the kinds of state-capital alliances that are essential for signaling a serious commitment to private initiative. At the same time, the president and his advisors have offered a sophisticated mix of incentives and compensatory schemes to those who are essential for sustaining market reforms (the private sector) and to those who have been key to Fujimori's survival in office (the mass base of popular sector constituents throughout the country who carried the president to victory in all three national elections since 1990).

The indisputable loser along this axis has been labor, both organized and otherwise. In Peru and in the region at large, the combination of a prolonged period of economic restructuring, company downsizing under privatization, and labor market deregulation has taken the wind out of the unions' sails. Furthermore, like Garcia, Fujimori has focused state resources and political attention on the disorganized and poorest segments of the workforce, and thereby cultivated a vertical relationship between the executive and the most marginalized sectors of the working population. Against this backdrop, after peaking in 1978, union activity and influence in Peru accounted for just 13 percent of private sector workers in 1995 (Thomas 1998: 163). Having gone from one of the most regulated Latin American labor markets in 1990, to one of the most liberalized by 1995 (Burki and Perry 1997: 40-41), this rapid transformation has thus far meant a proliferation of temporary contract work and an urban unemployment rate of close to 9 percent between 1990-1999. When the continuingly high levels of informal work and underemployment are factored in, economic uncertainty and fears of job loss have also hampered the formation of those horizontal organizational ties that characterized Peruvian labor during its heyday in the late 1970s.

Although Peru fits the regional mold in terms of interest intermediation along the state-capital-labor axis in the 1990s, the same obviously cannot be said of the relationship between political parties, congress, and the executive. The inability of political parties in Peru to revive themselves and reform from within, as Argentina's Peronist party has done over the past decade, or to forge a viable horizontal coalition of the kind that eventually unseated General Pinochet in Chile, is confirmed by the paltry percentage of votes that the standing parties have captured during every election in the 1990s. By the time of the 1995 presidential elections the three traditional parties (APRA, AP, and PPC) together captured less than 13 percent of the total vote. This reflects the public's disgust with the abysmal performance of all three parties during their time in elected office in the 1980s; but the decline of traditional political parties in Peru also has to do with the adherence to a majority runoff electoral format for the first time in 1990, and to the design of further rules under the CCD that complicated an electoral system that was already far too complex.

Matthew Shugart and John Carey (1992) have argued that, in contrast to plurality elections as in the U.S., majority runoff systems offer low barriers to entry and hence tend to generate larger candidate lists with much higher levels of uncertainty. Peru's majority runoff format also discourages alliance building and invites the rise of dark horse candidates, Fujimori's ascendance being a case in point; indeed, some have argued that Fujimori may never have won in 1990 had the election been based on a more simple plurality formula (Schmidt 1996). Another interesting counterfactual is the decision of once-major parties like APRA and the AP to boycott the 1992 elections for the CCD. At the time, potential opposition candidates for the CCD were understandably reacting to Fujimori's threat to require that CCD delegates sit out two full congressional terms before seeking re-election; the opposition also objected to the president's plan to pare down the Peruvian congress into a unicameral body with 120 representatives elected under one single district (Cameron 1997: 62-63), which he of course did with ease in their absence. Hindsight shows that by abstaining from the CCD altogether, the traditional parties marginalized themselves right out of the political arena.

Had these parties run for the CCD elections back in 1992 would they now be alive and well? Probably not, as it was fierce internal divisions and a gaping leadership void that thwarted collective action and paved the way for Fujimori's civilian-military political machine in the first place. But the fact that the opposition parties never even tried to assert a presence in the CCD put them further at odds with a median voter that had emerged from the turmoil of the 1980s with a fierce sense of independence and antipathy toward traditional partisanship (Schmidt 1996; Tanaka 1999). Thus, the CCD marked the turning point where political organizing in Peru shifted toward large eclectic movements that coalesced around election time and then scattered until the next election (Cotler 1995: 349). Such was the case with Fujimori's Cambio 90 or the Union for Peru (UPP), which backed the losing presidential candidate and former U.N. Secretary General Javier Perez de Cuellar in 1995. Geared almost entirely toward achieving electoral gains, and low on ideological or programmatic content, these movement coalitions are a far cry from the internal discipline and institutional cohesion that now characterizes political party structures in countries like Argentina, Chile, and Mexico.

In sum, although the absence of authentic interest intermediation along the party-congress-executive axis is indisputable in Peru of the 1990s, it is impossible to ignore the favorable economic returns over the past decade. These trends are contradictory, particularly in light of recent work which attributes successful instances of neoliberal reform in other emerging market countries to the firmer grounding of such reforms along this very axis (Haggard and Kaufman 1995; Corrales 2000a). Similarly, the current thinking on political parties in Latin America implies that an inchoate or collapsed party system such as Peru's would be much less likely to succeed on the economic front (Mainwaring and Scully 1995: 22-23). This gap between political economic theorizing and concrete outcomes has spawned a rich body of research on Peru in the 1990s, which has focused on two main questions: 1) What drives Peruvian politics in the era of market reform; and, 2) What accounts for economic success, given that interest intermediation and policy formulation have circumvented the party-congress-executive axis since the advent of Fujimori?

Most of the literature to date offers society-based answers to these questions. Susan Stokes (1998), for example, analyzes public opinion as a proxy for the president's political support and the perceived success of market reforms. After voicing broad support for the president and his economic program even before it began to pay off, the public shifted from this "intertemporal" posture based on the expectation of future prosperity, to a "normal economic voting" posture in which government support rose and fell more directly with the performance of the economy (Stokes 1998: 1). Yet, the government's mediocre showing in the three elections held during 1992-1993 showed that intertemporal responses to public opinion polls do not always translate into votes. Because Peruvian respondents proved to be quite fickle during this period, others have emphasized the ways in which state resources have been strategically deployed as a way of winning back votes and tying the masses to an executive that has otherwise shunned the institutionalization of such ties (Graham and Kane 1998; Schady 2000).

Coined by some as "neoliberal populism" (Dresser 1991; Roberts 1995; Kay 1996)---the use of inclusive political gestures and targeted material rewards to smooth over the exclusive and regressive impact of neoliberal reforms on the population at large---in Peru this strategy definitely helped to extend the president's incumbency beyond all expectations. But precisely because of its shallow roots in civil society, and its failure to tackle the underlying structural roots of poverty and inequality, Fujimori's neoliberal populist strategy seemed to have run its course by the end of the decade. Despite the continued channeling of generous state resources to those districts and regions where the president sought to lure voters into his camp, he was not able to avoid the lightning bolt rise of opposition candidate Alejandro Toledo in the first runoff vote for the 2000 presidential elections. For fear that not enough votes could be bought, the president and his cronies resorted to the kinds of balloting fraud and dirty campaign tricks that Mexico's PRI party had long been famous for, but which set new unseemly records in post-1980 Peru.

My answer to the above questions is decidedly state-centered, and meant to complement these societal explanations. I have argued throughout this chapter that the internal reform and modernization of those state agencies that are essential to the success of a market strategy constituted a necessary condition for sustaining economic reforms and for stabilizing politics. For the first time in the post-World War II period, at least some quarters of the Peruvian state could be taken seriously. The private sector responded in kind, and forged the closest thing to a government-business pact that the country has yet to see (Durand 1998). For better or for worse, a coalition comprised of the state, domestic capital, and the military has provided the stability needed to restructure the economy along market lines. But the exhaustion of first phase market reforms by the mid-1990s, and the contagion from massive external shocks in Asia, Russia, and Brazil from 1997-1999, have again highlighted that state renovation and stable leadership are necessary but not entirely sufficient conditions for coping with these new economic challenges.

As economic growth tapered off in 1998 and 1999, Peru arrived at the same crossroad that other emerging market countries like Argentina, Chile, and Mexico had arrived at somewhat earlier: full economic revival would require deeper institutional reform in numerous areas (tax collection, the financial system, social policy, the administration of justice, and much greater respect for the rule of law), as well as a more inclusive and encompassing development coalition to move these policies forward. Yet, rather than broaden the reform coalition in ways that would facilitate the wider participation and input demanded by second phase reforms, the president closed ranks in the guise of offering apolitical technical solutions to these more recent patterns of economic stress. A deeper problem is that the everyday concerns of political survival and the burden of constantly defending a decade-long incumbency distracted the ruling coalition from articulating a cohesive development strategy for the longer-term. Whereas this very coalition was once a source of confidence in the country's improved investment ratings, and a signal to the population that things were on track, it has increasingly come to be perceived by domestic and international observers alike as a main source of uncertainty.

"Markets without politics": It was the quest for political stability and continuity in economic policy that enabled the president and his backers to amend the rules surrounding re-election in 1995 and to neutralize political conflict over that decision. Similar concerns had underpinned constitutional amendments to allow for the re-election of the executive to a second consecutive term in Brazil and Argentina in the late 1990s. Like Fujimori, President Menem of Argentina had also sought to further bend the constitution in order to run for a third term. However, it was Menem's own party that reined him in on the grounds that, within a democracy, such fundamental rules as those surrounding presidential succession should not be up for periodic renegotiation. In Peru, obviously, there were no such reliability checks to deter the executive from almost single-handedly reinterpreting these same rules. Even apart from the highly questionable legal justification for a third term, there were at least two other reasons why an extension of Fujimori's stay in office was not in the country's best interests.

First, because little new had actually been accomplished in the way of deepening market reforms during his second term. As re-election in 1995 had afforded Fujimori a much longer time horizon on his tenure in office, the president's appetite for further reform was visibly curbed (Durand and Thorp 1998; Boylan 2000). As a result, privatization slowed considerably, exports were still lackluster and too dependent on raw materials (fishmeal, mining, and services related to the processing of primary goods), and social policy had yet to reach sufficiently beyond the executive's concern for political survival and hence his doling out of immediate adjustment relief. Second, because the prospect of a third term predictably provoked broad opposition (national opinion polls showed that nearly 70 percent of Peruvians felt that it was time for the president to move on), this prompted Fujimori and his congressional allies to exert their will through direct intervention in the country's legal and judicial apparatus. Always at odds, the relationship between the country's democratic transition and Fujimori's prolonged incumbency became even tenser.

Done in the name of maintaining political stability, the executive's rash legal interventions ultimately provoked a series of political crises. For example, when the equivalent of Peru's Supreme Court ruled that a third consecutive term was unconstitutional, congress quickly dismissed the dissenting justices. Various civic organizations then mobilized and gathered 1.2 million signatures in support of a referendum on the matter, only to have congress manufacture a package of questionable laws that killed the referendum. In the period leading up to the 2000 presidential elections, the judicial system, and in particular the main electoral agencies that fall under its jurisdiction (the National Elections Board, the National Registry for voters, and the National Office for Electoral Procedures), had all been tampered with in ways that favored Fujimori's re-election. Any critical voices within the media were subjected to constant surveillance and harassment by Fujimori's foot soldiers within the SIN.

The 2000 presidential election was a grand finale of sorts, at least in terms of Fujimori's efforts to project a technical, apolitical image in what turned out to be a quintessentially political race. Whereas debates over second phase policies such as education, health and the administration of justice had dominated presidential elections in other Latin American countries in the late 1990s, Peru's 2000 contest was notable, not for serious debate, but for the procedural infractions and venomous attacks that emanated from the Fujimori camp. Like opposition candidates in Argentina, Chile, and Mexico, Peru's leading opposition candidate, Alejandro Toledo, had vowed to uphold and strengthen the market program now in place, while also tackling the distributional problems associated with it. Toledo also promised to advance political reforms that would set the country firmly back on a democratic trajectory. Not surprisingly, public opinion polls in Peru had shown for some time that these were the issues that most concerned the electorate.

But the tone of the election quickly fastened onto the ethnic and racial themes---"El Chino" Fujimori versus "El Cholo" Toledo---that had been so crucial in sinking the candidacies of European-style elites like Vargas Llosa and Perez de Cuellar. The difference this time, however, was that there was no frontrunning criollo candidate for Fujimori to scapegoat. Rather, the Stanford-educated Toledo infused the race with an Andean ethnic identity and Horatio Alger rise from poverty that topped Fujimori's own struggling immigrant image. Fujimori's dark horse victory of 1990 also overshadowed any serious debate, as it appeared that Peru's majority runoff system could again wreak unexpected havoc, this time on the president himself. Toledo had gone from capturing a mere 12 percent in the campaign polls in February 2000, to 41 percent of the vote in the first round election. Indeed, not only did Toledo place second to Fujimori in the first round of voting on April 9 and therefore force a runoff race, but the polls showed that the two would run a virtual dead heat in the second round of voting.

The first round election had been wrought with accusations of government-sponsored fraud, and therefore met with the same international disapproval and threats of sanctions that the 1992 coup had provoked. As the Fujimori administration refused to guarantee a fair and clean runoff race, Toledo bowed out of the second round election, vowing to re-enter only when the electoral playing field had been cleared of all irregularities. While perhaps understandable, Toledo's withdrawal from the race at the very moment when he was gathering momentum for a possible victory, caught his supporters off guard. Was this a repeat of the 1992 CCD elections, where the opposition shot itself in the foot by staying off the ballot? Was this the same political opportunity structure which pro-reform opposition coalitions in Argentina, Chile and Mexico had seized in order to beat entrenched incumbents at the polls? Or, given Peru's lack of cohesive political parties, was it premature to expect that competitive politics and market reforms could advance forward along these same mutually reinforcing lines?

Had Toledo stayed on the ticket, it appeared at the time that international pressures, including everything from the suspension of bilateral aid and trading preferences to the loss of multilateral funding, could have prevailed in forcing the Peruvian government to stage an entirely new election based on fair competition and transparency. Without having finished the race to the end, Toledo risked alienating those international actors, including the European Union, the OAS and U.S. State Department, whom he would ultimately need to win a new clean replay of the election. In the end, there was no electoral reform, and thus no real contest; the second-round vote was still held on May 28, but it turned out to be more of a referendum on Fujimori. The results showed that nearly half of the voters did not support the means that the president had deployed to achieve his electoral goals. Fujimori took just 51 percent of the vote and, after failing to secure a majority bloc in Congress, managed to cobble together different independent factions to back him in the legislature. But Toledo won 18 percent of the vote, even though he had withdrawn his candidacy; the fact that voters had invalidated 30 percent of the ballots cast in the second-round runoff election left few doubts concerning the lack of legitimacy for a ruling coalition that had served its political and economic purposes, but simply refused to rotate out in democratic fashion.

As it turned out, just as international actors and local opposition forces in Peru seemed to have resigned themselves to a third Fujimori term, on September 17, 2000, the president suddenly announced that new elections would be held and that he would not be a candidate. Having completed less than two months of his third presidential term, Fujimori's unexpected resignation announcement took the entire hemisphere by surprise. Despite steady pressure from the OAS, the U.S. State Department, and the United Nations for the restoration of authentic democratic norms in Peru---including everything from the reinstatement of the constitutional tribunal, to the staging of fair elections, to the firing of Montesinos---it appears that the impetus for Fujimori's withdrawal was mainly an internal domestic affair. The triggering event was the surfacing of a videotape that fell mysteriously into the hands of the political opposition, and which showed Montesinos blatantly bribing a newly elected member of congress to switch his party affiliation over to the legislative coalition that backed the president. This, combined with Montesinos' cumulative track record of involvement in human rights abuses, drug and arms trafficking, and other forms of electoral fraud, prompted the president to dissolve the SIN and request the resignation of his treacherous security chief and right-hand advisor. Not surprisingly, the military split ranks, with Montesinos and his backers mustering enough incriminating evidence to bring Fujimori down with them, but not enough power to invoke a full-fledged military coup.

So, the apparent winners are those segments of the armed forces that had long been calling for a return to military professionalism and the reinstallation of merit-based criteria for appointments and promotion, and a political opposition that gathered new strength in the face of further revelations concerning high-level government corruption. At the time of writing, new elections are expected to be held during the first half of 2001, although the constitutional mechanisms and procedural details for staging another presidential race have yet to be worked out. At once, Fujimori's declared resignation tarnished the chances of up and coming presidential hopefuls like the recently elected vice president, Francisco Tudela, or Carlos Bolona, who had returned for a second, albeit even shorter-lived, stint as Economy Minister in the wake of Fujimori's May 28 electoral "victory;" and, it offered Peru's democratic opposition a second chance to form the kind of encompassing pro-reform alliance that has prevailed at the polls in Argentina, Chile, and Mexico.

Above, I argued that Peru's most glaring second-phase reform gap has been the failure to shift to a more open and participatory mode of politics. And now, almost uncannily, domestic actors have been given a second chance to seize this political opportunity structure. Already, the two main opposition candidates from the 2000 presidential race, Alejandro Toledo and Lima Mayor Alberto Andrade, have declared their intentions to run again, and the OAS has firmly committed its support in promoting a stable democratic transition in Peru. The ability of opposition leaders, international actors, and the incumbents still in office to agree on a graceful exit strategy is anything but guaranteed. However, the very prospect of imminent political change places the Peruvian economy and entire reform trajectory on much more optimistic footing than the previous option of waiting another five years to breathe new life into the political economy.

Markets without planning: In spite of these political machinations, the Peruvian economy showed strong signs of recovery by mid-2000. Real growth was running at more than 5 percent, annual inflation was projected to be about 5 percent, and the fiscal deficit had been reduced after a counter-cyclical expansion in response to the Asian crises. It appeared that Peru had finally crossed a major threshold in the sense that politics no longer drove economics: sound performance and professional management practices were now credibly grounded in the country's main economic institutions. This was reflected in the response of the external sector once Fujimori had been officially inaugurated, as capital flows and interest rate spreads remained favorable to Peru, and the IMF went ahead with a multiyear lending agreement. However, the president's ratings in national public opinion polls continued to lag, despite this favorable economic news (Schmidt 2000).

The normal economic voting posture described by Stokes (1998) seemed to have run its course by 2000, which could be partially explained by the self-inflicted political shocks described above. But a widening gap between macroeconomic dynamism and microeconomic stress had also shaped Peruvian public opinion in ways not well captured by either the normal economic voting or the intertemporal model. According to the latter, economic recovery in the early 1990s had met with a pessimistic response on the part of the population, as higher growth and real wage increases called up painful memories of hyperinflation. Ironically, in the first stage of market reform, pessimism concerning present performance correlated with optimism about the future (Stokes 1996b). Yet, a decade later, the future had arrived, and the gains from market reform still eluded the majority of Peruvians. Although Table 6.1 shows that Peru outperformed Argentina, Brazil, and Mexico with regard to the growth of real wages and employment expansion from 1990-1999, the data also show that between 1986-1996 the poorest 40 percent of the population continued to hold just 13-14 percent of national income. To put this another way, Richard Webb (2000: 271) notes that in the average Peruvian household, "income either stagnated or fell slightly between 1985 and 1997."

In light of these figures, and given the increasingly sinister political backdrop, the president's lower approval rating in the midst of an economic recovery is probably best interpreted as a distributional response on the part of those who have yet to fully benefit from Peru's high growth rates in the 1990s. In an innovative study of social mobility in Peru from 1991-1996, Carol Graham (2000: 261) analyzes new panel data which show that nearly 38 percent of the respondents (N=676) experienced either upward or downward mobility. Between 1991-1994, about 20 percent crossed above the poverty line, while about 8 percent dropped below it. But the 1994-1996 figures were literally flat: of the 23 percent of respondents that crossed the poverty line, 11.4 percent rose about it and 11.4 percent sunk further into poverty. Again, it is similar distributional trends that prompted demands for second phase reform in the other emerging market countries discussed here, and which ultimately helped to elect pro-reform political coalitions that sought to more aggressively bridge the distributional gap. In Peru, it is safe to say that distributional demands have been muted by authoritarian politics, and in particular the executive's tight control over social expenditures which provide short-term relief to the poor but have not sufficiently tackled the underlying causes of inequality and poverty.

Across the region, the tenacity of these trends flies in the face of neoliberal thinking, which holds that market reforms will expand the national pie and enable a wider segment of the population to gain access to this newly found wealth (Belassa et al 1986; Krueger 1990; Przeworski 1999). Recent research on Latin America's distributional shortcomings in the aftermath of market reforms offers two main explanations for this counter-intuitive outcome. First, income inequality has lingered in the region because of the high levels of asset concentration---both productive and human capital assets---that are still present in these economies (Baer and Maloney 1997; Birdsall and Londono 1997). And second, the failure of liberalization and privatization to penetrate these entrenched dualistic structures is no longer seen as an adjustment lag, but rather as an endogenous feature of market reforms which must be addressed through more assertive public policies (Lustig and Deutsch 1997; Sheahan 1997). In other words, whereas a purist approach to economic liberalization may have been necessary to purge Peru of the worst excesses of its developmentalist past, the region's track record suggests that, over time, microeconomic adjustment will not occur solely at the hand of market forces.

Remarkably, both Fujimori terms passed without the generation of an integrated development plan that reflected the government's policy goals on this very question. Understandably, "planning" had been given a bad name by both the RGAF and the Garcia debacle; nevertheless, it is sobering to think that the Peruvian military has been the only national institution to offer a longer term vision for the country in the era of market reform, and this more than a decade ago in the form of the "Green Book." The Fujimori administration had succeeded in attracting a higher level of professional expertise than any of the preceding administrations reviewed in this study, so the lack of an integrated development strategy which links macro- and microeconomic goals into the medium-term cannot be laid at the feet of technical incompetence. The failure to better harness the talent of this sophisticated policy segment surely had to do with the peculiar circumstances which led to the vetting of executive access by the likes of Vladimiro Montesinos, and to the president's own discomfort with talented technocrats who might upstage him. This situation is world's apart from other emerging market countries in the region, where executives have purposefully armed themselves with large high-profile technical teams which, themselves, have come to signal reform credibility.

In the final chapter I review the piecemeal steps that have been taken to infuse microeconomic dynamism into the Peruvian political economy, and to attack inequality at its roots. In doing so I borrow from the experiences of other market reformers in the region to highlight the ways in which the Peruvian effort could be strengthened. Peru again finds itself at a fascinating juncture. Wittingly or not, the president and his small base of advisors had reinvented the Peruvian state in ways that make it more capable of steering the economy onto a higher growth-higher productivity track, regardless of who occupies the executive office. To tinker with the impressive array of state agencies that appear in Table 6.6 would be to derail these goals. Thus, the institutional incentives work in favor of executive action to insure the longevity of these agencies, for example, by grounding them more firmly within the public administration and through the design of independent governing commissions whose members are appointed by the legislature and who serve staggered terms (Keefer 1995). While this was never Fujimori's style, his successor would do well to carefully nurture these modernized pockets of the state. Not only are they key to the next administration's political survival, but also to the promotion of greater economic stability and more widely shared prosperity in the post-Fujimori era.

Conclusion: Peru's Political Economy at Century's End

Up until 1990, the tendency in Peru had been for each successive administration that prevailed from 1963-on to hand over an even bigger bundle of political and economic problems than it had inherited upon taking office. The crisis of the late 1980s finally put an end to these delay tactics, as the combination of state collapse, hyperinflation, and civil war made it virtually impossible for politicians and policy makers to avoid implementing a long overdue set of reforms. In hindsight, a maverick independent politician such as Fujimori may have seemed the least likely candidate of all to succeed in restructuring the economy along market lines and in reconstructing the Peruvian state. Yet, the magnitude and longevity of the 1980s' crisis, combined with Fujimori's own fortitude, created a unique set of opportunities for the initiation of sweeping political economic change.

Already, Peru's transformation in the 1990s stands on par with that of the 1970s: Just as the twelve-year nationalist military regime put a final end to the country's longstanding oligarchic class, Fujimori's decade in office marked the demise of traditional political parties controlled by ineffectual criollo elites. And, just as the RGAF had radically redrawn the lines between state and market in favor of the former, Fujimori reversed them just as radically in favor of the latter. While the tasks of fine-tuning Peru's market model in a more dynamic and distributive direction, and of deepening the process of democratic transition have been left to Fujimori's successor, the country that emerged from the decade of the 1990s is considerably different from the one that entered it. This is so both in regional terms, and when Peru is assessed according to the macroeconomic and institutional variables laid out in chapter one.

When compared with the rest of Latin America, the data in this chapter confirm that Peru's main reference point for measuring economic performance is no longer an Andean bloc still struggling to implement market reforms, but rather such emerging market countries as Argentina, Chile, and Mexico. Furthermore, whereas the previously identified cluster of adverse trends related to state-led development (reckless borrowing, bloated SOE's, fickle investors, and regressive income distribution) were still fully intact at the outset of the first Fujimori administration, a decade later all but the distributional challenges had basically been resolved. With regard to institutional reform, the modernization of key entities like the central bank and a handful of ministries (Economy and Finance, Energy and Mines, Industry and Commerce), as well as the renovation or creation of a range of highly professional and efficient autonomous agencies, has catapulted Peru into the ranks of the top ten developing countries on such competitiveness indicators as the quality of state management and public spending (Vial and Sachs 2000: 10-11).

There is, however, much more to political economic success than the inroads just mentioned. While I have argued throughout this chapter that Peru's cup is now more than half full on th