Beyond the diffusion of neoliberalism: Embedded reasoning and policy innovation in the origins of the Argentinian currency board

Suffering from high rates of inflation since the 1940s and having experienced two hyperinflations at the end of the 1980s, in 2019 Argentina faced an increase of prices of more than 50%. But in the 1990s Argentina achieved price stability and growth at once. What were the sources of such an immediate change? What lessons can be drawn from this experience? As in the 1990s Argentina adopted profound market reforms, stability could be considered an achievement of neoliberal diffusion. Not only international forces were propelling change; the International Monetary Found (IMF) celebrated the success and invited other emerging nations to imitate it. Nevertheless, this perspective underestimates the importance of the local context in the history of neoliberal reforms and overestimates the coherence of external forces. Through the analysis of testimonies of Argentine and foreign officials and the study of declassified documents from the IMF, this paper argues that stability was only achieved after the adoption of a currency board, which was against the recommendation of most foreign officials. Instead of a simple top-down transposition of ideas, the Argentinian case reveals the importance of the agency of technocrats. Their local innovation not only made the diffusion of neoliberalism possible, but it also turned the currency board into one of the global anti-inflationist recipes recommended to other countries, despite its heavy consequences.


Introduction
Considered an ineluctable effect of growth in emerging economies in the postwar years, inflation has become an endangered species in the economic landscape. In most developed countries inflation rates remained relatively low after the Second World War and were successfully controlled and later reduced after the oil crises' impact of the 1970s. On the contrary, in emerging countries inflation was higher but accepted as a side effect of development until it climbed into three or four digits in the late 1970s and 1980s, encouraging deep reforms. Since then, most Latin American countries have left this experience behind and preserved stability. This is not the case of Argentina. Suffering from high rates of inflation since the 1940s and having experienced two hyperinflations at the end of the 1980s, this country dealt in 2019 with an increase of prices of more than 50%. In view of this long-lasting phenomenon, the exception astonishes: in the 1990s, as most of its neighbors, Argentina achieved prices stability and economic growth at once. What were the sources of such an immediate change? What lessons can be drawn from it?
From the policy diffusion perspective, Argentina can be regarded as an example of neoliberal diffusion, and supposedly there would be a strong relation between international mandates, market reforms and stability. Actually, in the 1990s Argentina adopted profound market reforms in a short period of time and within a democratic regime. Public companies were privatized while the government opened commercial and financial markets and eliminated most state regulations. From this perspective, the experience of the 1990s could show the path to solve Argentinian economic instability today. Not only the international community becomes a driving force for change. As in those circumstances, the International Monetary Fund (IMF) could be regarded as a good advisor and monitor of domestic public policy. Once again, Argentina should make its institutions converge with those prevailing abroad.
But the understanding of political economy decisions is complex and public policy still has a lot to learn from the origins of local neoliberal experiences. The issue has been approached from different angles. On the one hand, many case studies have analyzed the importance of domestic conditions for any change. Inflation in Argentina is only one of the identified opportunities for the adoption of profound reforms, as have been the crises of state interventionism, the recrudescence of political struggles and other severe economic disorders 3 . On the other hand, many researchers have stressed the effect of global forces in the spread of neoliberalism: economic ideas, capital flows, trade agreements and communication technologies 4 . These conditions were certainly crucial for the adoption of neoliberal policies. Finally, different paths to neoliberalism have been systematically compared 5 .
that "policies are formed and institutionalized by the way of traveling" (Chorev, 2009, p. 132). These analyses have provided critical insights for the study of policymaking. First, they placed recent transformations in the context of the longer history of emulation, coercion, learning and competition among different states. Second, the study of institutional mediators offers an observational vantage point to see how policies are shaped and reshaped by the process through which they are translated from one country to next.
In spite of this, certain assumptions persist. Frequently, neoliberalism tends to be interpreted as the result of a transfer of global "economic" ideas or pressures, nuanced by certain local "political" particularities. While unity, premeditation, and efficacy are ascribed to external "economic" forces, diversity, bewilderment, and incompetence or impotency characterize local "political" counterparts. Neoliberal policies tend to be considered as coherent and closed packages to be transported from an external starting point into distant locales; researchers tend to identify only one prevailing diffusion mechanism mediating temporal and geographical settings. On this basis, these perspectives tend to underestimate the importance of local contexts in the history of neoliberal reforms and overestimate the coherence of external forces. As noted, neoliberalism is a plural rather than a unique process (Puello-Socarrás, 2013, p. 27).
Through the analyses of testimonies of Argentine and foreign officials and the study of declassified documents from the IMF, this paper argues that neoliberal reforms gained legitimacy thanks to the fight against inflation, but stability was achieved only after the adoption of a currency board, which was against the recommendation of most foreign officials. Instead of a simple top-down transposition of ideas, Argentinian economic policy is a case of technocrats' agency and policy innovation. After its immediate success and thanks to the support of international organizations, the Argentinian currency board became one of the global anti-inflationist recipes recommended to other countries. Focused on the stabilization of the macroeconomy, local economists and foreign officials neglected both its heavy consequences and the threats of its probable end, which were nevertheless clearly foreseen in the earlier stages.
We argue that the agency and embedded reasoning of economic experts are missing links that can contribute both to diffusion theories and institutional mediators' analyses. The crucial role of economists in internal-external coalitions has been already noted: they are frequently portrayed as "agents of knowledge" that enable the diffusion and circulation of diagnoses, interests, or pressures (Levi-Faur, 2005). While their global integration (through socialization, practices and networks) has been very largely documented 6 , less importance has been given to their agency. Thanks to the analyses of Callon (1998), studies on sciences and technology revealed the performativity of economics. Most of this literature remains focused on economists in the developed world and less attention is paid to the of economists in the South. According to Kaplan (2017, p. :219), even if "technocrats are an important transmission mechanism for explaining economic policy choices", they have received little attention. The Argentinian case shows that economists' innovation not only made the diffusion of neoliberalism possible, it also enacted economic policy on the ground, operating within internal and external opportunities and constraints. Like Bockman and Eyal (2002), we demonstrate that in international dialogue both reformed countries and the international organizations are active and thoughtful agents, acting in a fragmented and pragmatic manner.
The inquiry reconstructs the decision-making process of economic authorities based on personal accounts 7 , documentary sources 8 , press data 9 , and the Minutes of the several discussions that took place on the Executive Board of the IMF for the approval or reviewing of the current Financial Agreements 10 . Taking into account the secrecy surrounding public decision-making and the contradictory positions adopted by local and international high officials, we triangulated multiple sources and did a permanent crosschecking among them.
The paper is organized in four parts. In the first one, we open the black box of neoliberal reforms to differentiate structural reforms and anti-inflationist policy. We show the importance of inflation to legitimize profound changes as well as the limits of structural reforms to achieve stabilization. In the second part, we present the arrival of Cavallo's team to the Ministry of Economy and the crafting of the Argentinian currency board, its  Montecinos & Markoff (2009). 7 After the crisis of 2001, one of the authors interviewed sixty professional economists in Argentina (scholars, state officials, consultants, journalists), several of whom had a direct or indirect intervention in the convertibility decision-making process. We also used the Archivo de Historia Oral of the Germani Institute, where very long interviews with high state officials of the period are available. Other personal accounts were consulted in Coiteux (2003). 8 We analyzed reports from the most important local economic think tanks as well as the Minutes of Argentine Parliament discussions. 9 A comprehensive survey of news about economic policy published between 1989 and 1995 by the most important national newspapers -La Nación, Clarín, Ambito Financiero -was collected and analyzed. 10 The Minutes reproduce the stenographic version of the debate on the Executive Board when considering the approval and/or renewal of a financing agreement, thus constituting a primary documentary source of significant importance -and recent access -to account for the interests of a country in relation to a topic under discussion in the IMF (Nemiña, 2013). These sources were made public only in 2010, making this research possible (IMF, 2010).
Beyond the diffusion of neoliberalism: Embedded reasoning and policy innovation in the origins of the Argentinian currency board predecessors and features. In the third part, we analyze the resistances and objections of foreign and local economists to the anti-inflationist plan. In the end, we show how the stabilizing effect of convertibility gained the endorsement of most foreign and local officials even to the point of becoming one of the global anti-inflationist recipes recommended to other countries. The dramatic social consequences and the tumultuous end of the Argentinian currency board illustrate both the uniqueness of this local innovation and the importance of original neglected warnings by experts about it.
The productive effect of hyperinflation and the limits of structural reforms Under the stress of hyperinflation, the Argentinean electorate did not vote on the neoliberal program purposefully in the 1989 elections. There were two main candidates in the presidential election of 1989. Eduardo Angeloz, from the official party blamed for the current economic failures, the Unión Cívica Radical, was advised by a group of neoliberal economists and promised to undertake with vigor the structural reforms that radicalism had previously rejected. It is hard to separate the electoral opinion on his platform from his party's discredit. The government's weakness favored the Peronist candidate, Carlos Menem, who was supported by traditional trade unions and the most marginalized sectors of the Peronist movement. His campaign revolved around a typical anti-elitist style promising a "productive revolution" and a "salariazo" (a drastic wage increase).
Once elected, the Peronist candidate, to everyone's surprise, immediately implemented structural reforms (Stokes, 2001). Menem's "liberal revolution" resulted in the enactment of two laws, both passed in 1989: the Administrative Reform Act (no. 23696) and the Economic Emergency Act (no. 23697). Within this new legal framework, privatizations followed with dizzying speed 11 . The state began to reduce public spending and the government renounced interventionism, bucking the trend of fixing, freezing and controlling prices and salaries. This turnaround was rounded off by a new diplomatic policy called "peripheral realism" (Escudé, 1997). Menem's government felt that consistent alignment with the US was the best strategy to follow. As Menem's ideological shift left his original allies in shock, he recruited traditional liberal and neoliberal economists as his main advisors and promoters of his agenda. Two factors are fundamental to understand the final success of neoliberal reforms in Argentina: first, the fact that they were implemented by a Peronist leader, using his internal influence and clout over unions (Weyland, 2002) and, second, the context of inflationary trauma facilitated the appointment of mainstream economic advisors (Kaplan, 2018).
Even with these two factors at work and the far-reaching nature of the economic measures carried out, the spiral of inflation did not stop. In spite of the fact that the pillars of the Peronist doctrine had been abandoned, major structural reforms were already in place and the support of the US government was granted, the effort to tame inflation failed. Three economic ministers paraded through office in one year and by 1990 Argentina was experiencing a hyperinflation rate of 2,310%. At the time of the presidential elections in May 1989, the dollar was worth 100 australes, in November 1989, it climbed to 1,000, and by December 1990, to almost 10,000. According to opinion polls, the image of the President was falling and his party risked losing the next legislative election (Clarín, 04/06/91, p. 7). In this context, the government sought new approaches to curb inflation, one of which was convertibility.

Buenos Aires' recklessness…
The arrival of Domingo Cavallo, a Harvard trained economist, to the Economic Ministry in January 1991and, in particular, the enactment of the Convertibility Act several months later -allowed the government to finally stabilize the economy and consolidate the neoliberal order. Far from being at the origin of this story, the support of the international community and the success of its ulterior local allies should be placed at its end.
First appointed as the Foreign Affairs Minister, Cavallo seemed to be consistently opposed to using the overvalued exchange rate as a means to achieve price stabilization, a solution that was widely discussed in Menem's inner circle 12 . His position was made explicit in his press interventions and justified in a book (Cavallo, 1984). According to interviewees, most members of his professional circle were aware of his resistance to dollarization, which, by contrast, was promoted by local orthodoxy as a solution to inflation (Almansi & Rodríguez, 1989). When a bill 11 The national telephone company and commercial airline passed to private hands in less than a year. The reduction of tariffs encouraged the opening of the domestic market. Transportation was deregulated and regulatory commissions for agricultural and industrial production were eliminated. 12 The oldest Argentine convertibility promoter was Eduardo Curia, who paradoxically became later one of the fiercest opponents of Cavallo and his program (Curia, 1999). On the basis of interviews and press analyses, at least two other instigators can be mentioned: the CEMA economists, including Carlos Rodríguez and Ricardo Arriazu, one of the initiators of the 1978 plan.
However, in the beginning of the 1991, the dollar skyrocketed again and put pressure on the inflation rate. There were legislative elections that year and, faced with the dangerous prospect of growing conflicts and electoral defeat, the government chose a drastic approach to stabilize prices (Starr, 1997, p. 109). For that purpose, Cavallo took up a currency board 13 project that had been designed by one of his team members (Roig, 2016). The project was consistent with the monetarist approach of the economic team and also with the commitment to eliminate the monetization of the public sector deficit.
Was convertibility a complete Argentinian innovation? Not at all, in its pure sense. As a guarantee of low inflation, currency boards were used to make trade with the metropolis easier during the second half of the 19th century and the beginning of the 20th, until the beginning of the Great War. From then on, the social and economic costs associated to maintaining the gold standard encouraged governments to have greater control over exchange rate policies. This led to the progressive decay of currency boards in several countries and the creation of central banks (Walters, 1998). After the fall of the Bretton Woods agreements and the advance of the floating exchange rate frameworks, by the 1990s less than ten countries around the globe had currency boards, and these were all small economies, with their trade centralized in very few partners 14 . Thus, in the early 1990s the currency board seemed to the international community to be an old tool, reserved for very small and undiversified economies. Convertibility was not even unknown in Argentina. Vitelli (2004) says that there had been five convertibilities between 1822 and 1929, and the most long-lasting of them was in force between the end of the 19th century and the beginning of the Great War, and it coincided with a period when prices of raw material peaked.
These predecessors do not only anticipate the risks associated with currency boards but also indicate that the Argentinian innovation did not consist in creating something brand new but in adjusting and updating an anachronistic monetary institution. In fact, the 1991 Argentinian convertibility was the first to be proposed in a context of low commodity prices and financial liberaliza-tion. Only external financing could maintain it in a context of trade deficits and irregular FDI flows (Ponsot, 2003).
Taking as a starting point the perception that it was necessary to offer a shock of confidence to stop speculations regarding devaluation of the national currency, the bill postulated the free convertibility of the national currency with the American dollar at an equivalent exchange rate. To emphasize the change at a symbolic level, the name of the national currency was changed from austral to peso, taking on the historical name of the Argentinean currency. The currency also lost four zeros (that is, 10,000 australes converted into 1 peso). That way, the parity 1 peso/1 dollar was established 15 .
Far from being an external imposition or the result of a complex technical layout, the convertibility regime was the response of the local economic authority to solve the problem of inflation radically. According to their testimonies, the economic team worked under two assumptions: first, that financial agents had institutionalized inflation, so any stabilization program had to use the shock effect; second, due to the persistent inflation, Argentines had in practice replaced their currency with the American dollar: the American currency was used as a unit of account and medium of exchange for durable goods. Thus, the officials in charge based their interpretations on the psychology or subjectivity of Argentines, rather than on intrinsic virtues of the model. The IMF executive director for Argentina emphasized the existence of an "entrenched inflationary psychology" (IMF, 1991a, p. 33), while Cavallo spoke of Argentina as a country that "did not have a currency" and explained economic fluctuations saying that "The Argentinean people are cyclothymic. We have a psychological cycle…" (in Liviatán, 1993, p. 37, 51).
Included as a monetary instrument of a broader anti-inflationist program centered on fiscal adjustment, the new monetary system included a fixed exchange rate between the local currency and the dollar, as well as a ban on index-based adjustments 16 . The state was no longer allowed to print money without an equal foreign currency reserve as a means of financing its deficit or encouraging growth, but the regime went even further. From then on, Argentina adopted a dual-currency system, meaning that both currencies -the dollar and the peso -could be used equally. The economic team ensured that the peso was born as "sound", and for this purpose it should be chosen by the people and not imposed by the state; making the dollar circulation 13 A currency board refers to an exchange rate arrangement in which the exchange rate is fixed to an anchor currency and the Central Bank is not allowed to issue national currency unless an equivalent amount of international assets to back it is obtained. This means that the Central Nank is not able to develop an active monetary policy (Walters, 1998). In the Argentinean case, the exchange rate was pegged by law to the US dollar in a 1 to 1 relation, so it was a hard peg scheme. 14 They were Brunei, Djibouti, Hong Kong, Antigua y Barbuda, Dominica, Grenada, St. Kitts and Nevis, St. Lucia and St. Vincent and the Grenadines; the last six ones were members of the Eastern Caribbean Central Bank (Enoch and Gulde, 1998). 15 The Argentine peso was first legally issued on January 1st 1992, so between April and December 1991 the American dollar could be freely converted into 10,000 australes. 16 With the notable exception of public services tariffs, which were linked to the US inflation rate to guarantee a privileged return in US dollars to external private operators (Azpiazu, 2005).
Beyond the diffusion of neoliberalism: Embedded reasoning and policy innovation in the origins of the Argentinian currency board official serving this purpose (Cavallo, 1996). Argentines maintained the right to save, invest, and establish local contracts in dollars, and banks could register and contract operations in both currencies. This relationship between pesos and dollars, established by law, prevented the Central Bank from autonomously taking the role of "lender of last resort". In case of a crisis of confidence, national authorities could not respond by creating the required currency (the dollar). Monetary policy was thus disentangled from the national state's prerogatives and intertwined in foreign currency and external capital fluctuation. Even if it translated into a drastic integration into the financial and commercial global flows, Argentina's convertibility included elements that betrayed both the orthodox doctrine of currency boards and the commandments established by the Washington Consensus (WC). Unlike previous experiences, Argentina's 1991 currency board did not abolish the Central Bank and maintained some margin for discretion on exchange policies. Although the Central Bank adopted the currency board's criteria, it could integrate up to a third of reserves in local sovereign bonds denominated in dollars, valued at market prices 17 . Also, there was no maximum ratio of reserves, above which they would accumulate without the corresponding monetary expansion. Both elements provided a margin to sterilize or expand the monetary base in a discretionary way (Hanke, 2008;Hanke & Schuler, 2002). On the other hand, although according to Ramírez (2013) the convertibility was not unconnected to neoliberal debates since it helped to constrain the government's monetary freedom, a currency board system with a fixed and appreciated exchange rate was against the economic orthodox tide, which favored the floating exchange rate schemes that had become predominant after the fall of the Bretton Woods regime, and it deliberately contradicted the Washington Consensus' recommendation for developing countries, which included having a floating and depreciated exchange rate in order to boost external surplus to facilitate the repayment of foreign debts (Williamson, 1990).

…against Washington's advice
The establishment of an exchange anchor sanctioned by law constituted an economic policy inspired in the public choice framework, which, through the promotion of a depolitization of the public administration and the limitation of demos, was one of the leading theoretical sources of the Virginia School, the most relevant neoliberal school of thought in Political Science during the nineties (Morresi, 2007;Ramírez, 2013) 18 .
But notably, at the very beginning Argentina's convertibility was criticized by both foreign and local economists 19 . Orthodox economists 19 denounced the flexibility margin kept by the Central Bank, by pointing out that it was in contradiction to the postulates of a pure currency board and that it was a source of distrust that could turn against the efforts of the authorities. When, during a seminar at the World Bank, Cavallo said that having a Chicago boy at the Central Bank provided additional confidence in the continuity of the currency board, Hanke, a well-known and orthodox economist, answered that if Argentina had a pure currency board, who was in charge of the Central Bank wouldn't be relevant, since it wouldn't exist at all (Liviatán, 1993, p. 50). Heterodox economists 20 did not support this strategy either, but they stressed other arguments. On the one hand, according to them, pegging the local currency to the dollar would sooner or later reduce Argentinian competitiveness. Open commercial markets would increase imports while the country had already trouble to place its products abroad. On the other hand, constraints on public spending and economic retreat would sooner or later provoke unemployment and poverty.
As a result, convertibility faced not only resistance at an intellectual level but also questionings regarding the political support it would receive. Reflecting the political and academic consensus about suitable exchange arrangements, many crucial actors at the time demonstrated an active resistance to the overvaluation of the local currency as a means for stabilization. Convertibility did not have any support from largest companies, the IMF or the central countries' governments before its announcement. Most local big firms agreed with the structural reforms but resisted any exchange rate anchor with the dollar (Beltrán, 2011;Viguera, 2000;Etchemendy, 2011).
The IMF authorities were visibly reluctant. The IMF expressed doubts based on several grounds. They had been traditionally attached to flexible exchange schemes and to using free market mechanisms to establish exchange rates 17 That limit was later reduced to 10%, but it could be increased in times of crisis, which happened during the Tequila crisis and in 2001 (Hanke & Schuler, 2002). 18 Although they shared the basic assumptions, in the field of economics the version of the Chicago School constituted the dominant perspective that informed neoliberal thought (Denord, 2002). 19 This refers to economists who agreed with the neoclassical theory and/or those who gave express support to the postulates of the WC in the early 1990s (Dequech, 2007, p. 293). 20 Carlos Rodríguez, a referent of the orthodox Center of Macroeconomic Studies of Argentina (CEMA, after its initials in Spanish), warned about the "traps" of convertibility that would allow the government to dangerously go into debt. Also, some members of the liberal group of legislators Unión del Centro Democrático (UCEDE) decided to vote against convertibility because it did not guarantee a "genuine" fiscal discipline (Ámbito Financiero, 22/3/91, p. 7 and 26/4/91, p. 1). 21 Schvarzer (1996, p. 149-150). (Babb, 2007, p. 131). The IMF officials feared that funds obtained from privatization would be used to support the exchange rate instead of go towards paying Argentine's debts (Bonelli in Clarín,04/26/91,p. 16), which is what actually happened. Finally, they emphasized that the 1 = 1 exchange rate was 35% appreciated in relation to the 1990 average, and underlined that that could conspire against the sustainability of the exchange scheme (IMF, 1991b, p. 20, 34).
For their part, in July 1990, in a meeting of the IMF's Executive Board aimed at approving the new Stand-By arrangement, convertibility got a cold reception from the directors representing central countries. With the exception of the German director, who praised the program and the convertible regime, the others showed a cold attitude and stressed the possibility of Argentina being capable of reaching enough fiscal balance to make projections sustainable. The director for Canada, for example, pointed out that "(t)he key element of this new program is fiscal restraint. As we are all aware, a fixed exchange rate does not in and of itself confer credibility" (IMF, 1991c, p. 48).
At the time, Argentina-US diplomatic relations faced growing tensions. The US ambassador had accused European competitors of paying bribes to Buenos Aires authorities to prevent a fair participation of American companies in privatizations. President Bush had encouraged structural reforms but had not guaranteed any financial support for the monetary policy 22 . For this reason, the IMF director for the USA merely supported the agreement, although she said it was important to increase the rhythm of debt payments (IMF, 1991a, p. 7) 23 . The directors for Spain and Japan joined the request; the latter emphasized his doubts regarding the sustainability of the exchange rate 24 and the projections of growth and revenues (IMF, 1991c, p. 47). Finally, the executive directors for Italy, Canada, the United Kingdom and the Netherlands joined the Japanese director in his doubts. While the latter two picked up on the question of the integration of one third of the reserves in sovereign bonds (IMF, 1991c, p. 53, 56), the Canadian director even suggested that it was necessary to design a contingency plan to be put into practice if the currency board turned out to be unsustainable (IMF, 1991c, p. 50).
Since it did not satisfy internationally established canons, neither those of a pure Currency Board nor those of the WC, and since it did not have the initial support of major foreign and international authorities, convertibility cannot be considered a part of an imported economic orthodoxy. Once adopted, it wasn't clear whether those who voted for the arrangement at the IMF or locally supported the fixed monetary system believed that it would endure for a decade, or if they viewed it as a short-term solution to economic problems. President Menem, who presented himself as an enthusiastic believer in the fixed monetary system, insisted on numerous occasions that it would last for decades (Clarín, 06/04/91, p. 7), but the opinion of experts committed to this measure seems less clear, their later testimonies and analyses suggest that they thought it would only be in place for a few months. An analysis of data from the early 1990s reveals that experts predicted that convertibility would have a short life span 25 . Like most heterodox and orthodox groups, Menem's Economic team worried about the social costs and external vulnerability caused by convertibility 26 .

Global alignment and eclecticism for export
Suspicions about convertibility were soon dissipated by its effectiveness. After more than forty years of chronic inflation and twenty years bordering on hyperinflation, the government had accomplished, in just a few months, the stabilization of prices, making Argentina's inflation rate one of the lowest in the world 27 . This effective control coincided with economic growth after more than a decade of stagnation. The electorate supported Menem by voting for his reelection in 1995. Soon, multiple actors wanted to be recognized as precursors, partners or enthusiastic adherents of this achievement.
The Convertibility Plan benefited from the Brady Plan after its adoption. Along with other countries, Argentina ratified an agreement in 1993 that established an extension of its debt maturity with a reduction of actual repayments 28 . As a result, states and local commercial banks once again had access to foreign financing. Encouraged 22 Apart from participants' testimonies, this general support (without financial aid) was publicly quoted and analyzed in the press. Cf. Clarín (04/13/91, p. 11 and 04/12/91, p. 16). 23 Back then, Argentina was slowly mending its external situation, after the 1988 extension; the debt would be definitely renegotiated through the Brady Plan in 1993 (see below). 24 Faced with the insistence on the issue at other meetings later on, the Argentinean director replied that agricultural subsidies in central countries conspired against competitiveness in the local economy. He therefore restricted the debate on the exchange rate to the conclusion of multilateral trade negotiations on this topic (IMF, 1992a, p. 34). 25 In an interview with Coiteux (2003), one of his closest partners, Juan José Llach, affirmed that the economic team was prepared to introduce adjustments but the hostile reactions of financial markets and the public opinion prevented them from doing it. The press also shows that the minister foresaw some exchange rate flexibility (Ruiz, 2005, p. 210). Nevertheless, by 1993, the minister claimed that parity was "untouchable" and would last "an indeterminate number of years". 26 Early local consciousness of these probable negative effects can be observed in parliamentary debates and press analyses. 27 Annual inflation dropped from 2314% in 1990 to 24.9% in 1992, and two years later it was in a single digit. 28 The agreement involved the conversion of $ 23 billion of debts with commercial banks (from a total external debt of around $ 65 billion) to par bonds with interest rates reduced by two-thirds approximately, or discount with a nominal haircut of 35 %. Banks gave up a third of their claims in exchange for securities collateralized with US Treasury bonds. There was an additional reduction of 6 billion dollars by debt-to-equity conversions from the privatization process (Cline, 1995, p. 317).
Beyond the diffusion of neoliberalism: Embedded reasoning and policy innovation in the origins of the Argentinian currency board by declining interest rates and recessions in major western countries, financial flows shifted towards the "emerging markets". Argentina's inclusion in this list allowed it to rely on capital inflows at relatively low interest rates to support the state budget imbalances and the economic recovery (Santiso, 2003). This was reflected in a visible change in the position of the IMF and central countries vis-à-vis Argentine convertibility, which went from underestimation to recognition and, after the Mexican crisis, to acclaim.
In order to prove their support for the process of structural transformations, the directors for the United States and the United Kingdom expressed several times their enthusiasm for Argentine economic decisions. While the former pointed out that he expected to be accepted as a member of the "Amen Chorus of the Casa Rosada" (IMF, 1992a, p.: 12), the latter went further and said, expressing his impressions after a visit to the country and a meeting with the Minister of Economic Affairs: "(T)he secret of success in Argentina […] was […] the combination of structural reform on the one hand and macroeconomic stabilization on the other". And he added: "He would be interested […] on what lessons the Argentine experience might have for other countries […]. (W)ith respect to macroeconomic policy, whether the shock treatment of a currency board arrangement could be repeated in other countries and, if so, what were the necessary conditions for success" (IMF, 1992a, p. 12-13).
Financial capital inflows threatened to overheat the economy, since convertibility forced the Central Bank to change it for pesos, with the corresponding increase of the monetary base. This created a controversy over how to approach the situation, which was taken up in several meetings of the IMF Board between 1992 and 1994. The opinion of the majority, among which France, Spain, Italy, the IMF itself and, later, the USA stood out, was progressively in favor of the use of "relief valves" that originally included increasing reserves above 100% of the monetary base, having open market operations and increasing banking reserves to avoid a rise of the inflation (IMF, 1992a(IMF, , 1992b(IMF, , 1994a. The Board also discussed economic competitiveness. Aligned with the official approach, the majority opinion (which included the IMF and practically all of the G7 plus Spain) was to maintain convertibility, making use of the margin of monetary flexibility considered by the law and to go deeper into structural reforms 29 . Spain and Italy admitted that, considering the problem that the real appreciation of the exchange rate represented -a consequence of fixing a high rate from the beginning, plus the persistence of inflation after the establishment of convertibility --, the normal approach would be to take advantage of the context of growth to depreciate the exchange rate nominally, but since that could have awoken recent and traumatic memories, it was best to give this up and improve competitiveness with structural reforms (IMF, 1993, p. 49, 84). However, Canada and -to a lesser extent -the Netherlands were critical. They both rekindled their objections to the "relief valves", pointing out that they distorted the currency board. Canada went further and insisted on the suggestion that Argentina had to devalue and establish a floating exchange rate system, as most of the world was using, to regain competitiveness (IMF, 1992c, p. 33;1993, p. 80, 81). Towards 1994, the tenacious Canadian opposition was the only critical voice against convertibility. It went from being one of the instruments of the anti-inflationary shock plan to being considered, a few years later, by the US, Italy, Spain and France as the core of the economic plan (IMF, 1992c, p. 23, 34;1994b, p. 19). Even the USA identified the implementation of convertibility as the initial moment of the economic growth in Argentina (IMF, 1994b, p. 15).
The 1994-1995 Mexican crisis was a test for the program and its support. Although stabilizing inflation at one digit and external and internal approval of the economic program led the government to announce that they would not renew the agreement with the IMF as a sign of strength, the acceleration of capital outflows due to the "Tequila Effect" prompted the country to ratify the continuity of the agreement and to ask for an increase of the available funds of IFIs. Also, authorities took advantage of the bank run to move forward on the reform of the banking sector, which included the privatization of several provincial banks and generated a rise in the concentration of the sector and the transfer to foreign ownership (Bleger, 2000). In the middle of the financial crisis, the IMF approved a disbursement of over 1 billion dollars. The majority opinion among central countries was to support the continuity of the convertibility regime.
Going against the tide of the generalized consensus, Germany and Canada criticized the continuity. The former used a musical metaphor that turned out to be prophetic, considering the path that the country followed. Germany suggested that convertibility had become a dilemma similar to the one in the famous song "Hotel California" by The Eagles: "You can check out any time you like, but you can never leave" (IMF, 1995a, p. 9). In that sense, Germany's criticism was that if the mere evaluation of the continuity promoted distrust, that meant that the regime was structurally weak (ibid., p. 54). Some replied that such objection should have been exposed earlier, and the managing director of the IMF said that they had had off-the-record discussions with the authorities regarding alternatives. However, they reached a consensus that continuity was the most reasonable option given the circumstances (IMF, 1995a, p. 55, 56).
Once Argentina navigated through the crisis, the local and external interpretation was that it managed to do so thanks to convertibility, not in spite of it. Although they disagreed on the effects of the experience for the country's future, Argentine economists granted the fact that surviving the Mexican shock had made the convertibility regime stronger. According to Juan José Llach (an official close to Domingo Cavallo), "The Tequila crisis had a negative effect on a social level because it made us believe that we could go on as we did until then, and that was a mistake" (in Coiteux, 2003). The general local support convinced the IMF and most central countries that the convertibility regime was solid and could guarantee growth and modernization. And not just that: other developing countries began to consider the possibility of adopting similar programs.
The support was also almost unanimous at the international level. At the IMF Board meeting by the end of 1995, Germany expressed surprise because the report mentioned neither the exchange rate nor convertibility, and asked rhetorically if by omitting the issue they sought to sink it in oblivion (IMF, 1995b, p. 17). The explicit or implicit answer provided by the rest of the G7 (except Canada) plus Spain and the IMF was that it was not the right time to evaluate alternatives and that the exchange rate was not a problem, given the improvement of the trade balance (IMF, 1995b, p. 13, 34) 30 . The USA went even further and warned that the "currency board regime […] perhaps deserves more prominence than it has received" (IMF, 1995b, p. 15).
From that moment on, far from being denounced as a singular Argentine eclecticism, convertibility became a part of the stabilization recipes recommended for struggling economies by international experts (Sgard, 2004). Top Argentinean officials and economists promoted the adoption of similar systems, especially in former socialist republics. For example, in 1996 Cavallo advised the government of Ecuador, in 1998 he traveled to Moscow to "export" his program, and another Argentine economist participated in a similar experience in Bulgaria. By 1997, Estonia, Lithuania, Bulgaria and Bosnia had implemented currency board regimes and other countries (e.g. Brazil with the Real Plan) advanced in hard pegs schemes (Rapoport et al., 2000, p. 978). In that context, even the IMF published a specific study on the advantages and disadvantages of convertibility, in which it acknowledged that currency boards "have undergone a revival", although it only recommended them to small and open economies, and those who lacked highly developed financial systems (Baliño y Enoch, 1997, p. 1).
Instead of defending the local sources of the success of convertibility, Argentinean experts attributed them to the alignment of local institutions with the "First World", which international organizations enthusiastically accepted. By 1995, the IMF accorded Argentina and its experts the status of "best students" in international recommendations. The 1995 agreement included a reformulation of the commitments made by Argentina to the IMF. Bembi and Nemiña (2007) show in their analyses of the IMF Letters of Intent that parity exchange ceased to be an "instrument" and became a "guiding principle" of economic policy in order to later symbolically absorb the structural reforms program under the "Convertibility Plan" label. The IMF support was not just symbolic; it also included the necessary funds to support the exchange parity at Argentina's disposal. Faced with the delegitimization due to its failed intervention in the financial crisis in Southeast Asia, Argentina appeared to the IMF as a successful case to offer the world.

Concluding remarks
This article analyzes the work performed by local and foreign economic authorities along Argentina's path to neoliberalism and stresses their crucial role as links between local and international forces, highlighting their actions in light of the opportunities and constraints they faced at both levels. The diffusion literature has identified structural reforms such as trade liberalization, the deregulation of markets, and privatization as the core components of worldwide neoliberal reforms, associating their widespread adoption to global promoters and international institutional mediators. Our research complements these findings by paying particular attention to the implementation of policies on the ground. The Argentine case illustrates that free-market reforms could be adopted because they were congruent with urgent local problems. Reforms were not articulated as self-justifying goals, but as the "only possible" means of achieving stabilization and growth.
Beyond the diffusion of neoliberalism: Embedded reasoning and policy innovation in the origins of the Argentinian currency board As other authors have already stressed (Biglaiser, 2009;Neiburg, 2006), the international components became a part of economic policy in Argentina through the training of high state officials and due to the extension of the expert circles involved in the decision-making process. But foreign components were not only present in economists' training and networks; opportunities available at the international level, inspiration from foreign models, imported price references, the interests of concerned parties, conditions imposed by institutional mediators, and the local effects of the fates of competing countries were all external ingredients of local policy. Nevertheless, in the implementation of convertibility, public officials acted in pragmatic and creative ways to develop policy solutions even against the original advice of their foreign partners.
Concretely, the participation of Argentine politicians and economists was reflected in three facts: first, the implementation of convertibility as an instrument for curbing inflation, despite the fact that it was considered an outdated exchange arrangement, and the rejection of the Washington Consensus' recommendation of adopting a floating exchange arrangement with a devalued real exchange rate. Second, Argentinian convertibility included several unorthodox features that allowed the Central Bank some room for maneuver on monetary and exchange rate policy. Finally, their immediate achievement enabled a change of tone in the IMF and the international community from prudence and skepticism to final approval and celebration of the convertibility regime from its initial adoption to the Mexican crisis. Argentine convertibility was far from being a handme-down recipe from the IMF master chef. It was the product of local innovation, which went through a dramatic arc of international discussion and questioning to close monitoring and, later, promoted to the status of a recommended international economic policy.
The position of foreign officials was far from being unified in favor of the Argentinian plan. In fact, in the early 1990s most members of the IMF board criticized the adoption of the Argentinian currency board. It was only after its stabilizing effects and in the light of the Mexican crises that they recognized and celebrated the virtues of Cavallo's strategy. The reasons were not only technical: countries represented in the IMF board, such as Spain and Italy, had economic motivations since they defended the high profitability in dollars of their companies based in Argentina. Nevertheless, even if they were in a minority position, critics remained active throughout the period, and foreign officials' arguments anticipated most of the limits of the Argentinian choice.
Argentinian innovation and its later international endorsement can hardly be considered a story of success.
While the global community celebrated the Argentinian case and encouraged other countries to follow its path, many indicators confirmed the early warnings. As the heterodox economists had foreseen, there was a severe increase of unemployment, poverty, and social inequality. As orthodox economists had alerted, public and private indebtedness grew dramatically and ended up weakening the confidence of financial markets. Trapped in its achievement, Argentina could only decide to abandon its currency board after the most dramatic and full-scale crises of its history.