Preliminary Draft 18/7/05 Lessons From the Argentine Case of Debt Accumulation, Crisis and Default by Mario Damill, Roberto Frenkel and Martín Rapetti* Paper prepared for UNCTAD * Resarchers at CEDES, Buenos Aires 1 Lessons From the Argentine Case of Debt Accumulation, Crisis and Default By Mario Damill, Roberto Frenkel and Martín Rapetti 1. Introduction This paper presents analyses of the Argentine foreign debt issue from various perspectives. Following this Introduction, in chapter 2 we analyze the processes that led to debt accumulation in the long run and the macroeconomic policies that contribute to explain it. Particular attention is put on the nineties’ period and on the events that immediately anteceded the debt crisis and the default. Then, chapter 3 presents an analysis of the macroeconomic performance before and after the recent crisis. One of the purposes of this chapter is to evaluate the role played by the default decision in the development of the crisis and in the following economic recovery. Chapter 4 is dedicated to the post-default period. It describes the restructuring proposal and presents the evolution of the public sector’s financial obligations after the default. Finally, chapter 5 examines the relationships between Argentina and the IMF after the default and the role played by the institution in the debt restructuring process. In order to draw lessons form the Argentine experience it seems worth pointing out in this Introduction some controversial issues tackled in our analyses. Firstly, we contend a conventional explanation of the Argentine recent crises and default that is vastly disseminated. An uncontrolled public spending is seeing as the main cause of de debt accumulation, the crisis and the default (Mussa, 2002) (IMF IEO,2004). We have examined this issue in previous works (Damill and Frenkel, 2005; and also Damill, Frenkel and Juvenal, 2003). The polemic leads us to a detailed examination of the fiscal accounts. We show that the cumulative effects of the interest rates’ rise, pushed by the increase in the country-risk premium after the Asian and Russian crises, were the main cause of the public debt dynamics in the last quarter of the nineties. The fiscal deficit increased despite a significant rise in the primary balance surplus. The climbing trend in the country-risk premium and the interest rate in the last quarter of the nineties could be associated with the situation of the fragile external accounts or, alternatively, with the evolution of public finances, or with both, as the investment funds analysts and the risk rating agencies actually did in their reports. However, even if the uncertainties regarding public debt sustainability weighted significantly in the investors’ assessments, this should not hide the original source of the rise in public deficits and debt in the late nineties. The main source was not an exogenous mistaken fiscal policy, but the compounded effects of external and financial fragility and contagion. Secondly, we underline the role of macroeconomic policies – particularly the exchange rate policy - in the generation of the unsustainable debt path. In this regard, the Argentine case is an extreme example of badly managed trajectories of integration into the financial globalization process, leading to debt traps of high interest rates, low growth and high vulnerability vis-à-vis contagion effects and capital flows volatility (Frenkel, 2003b). One lesson that can be drawn from these analyses is that 2 macroeconomic policies – particularly the exchange rate policy – are crucial in avoiding international financial integration trajectories leading to unsustainable debt paths. Macroeconomic policies – helped by capital account regulations, if needed – should be carefully instrumented in order to manage the evolution of the current account and the external indebtedness. Thirdly, we also challenge an extended opinion about the role of the default decision that considers it the main responsible factor for the Argentinean deep crisis and its high social cost. Our analysis shows that the abrupt contraction in the activity and employment levels came up to a great extent before the default, while the government submitted the country to big efforts to keep the debt services on track. Actually, the default turned out to be one of the conditions that allowed the recovery that took place soon after. This was not only due to the positive fiscal effect of the payments suspension, but also a consequence of having freed the economic policy from the need to continuously issue signals aimed at facilitating the roll over of the debt obligations. The lesson here is that when a country faces a crisis motivated by firm expectations of default, what is really costly is the postponement of the default and not the default itself. Fourthly, we underline that the debt restructuring took place in the context of a peculiar and conflictive relationship between the IMF and the country. The most unusual feature in this process is that the IMF did not participate in the design and management of the debt restructuring. Neither did the organism audit the government’s financial projections that justify the sustainability of the proposal. These circumstances have no precedent in the international financial system that has been developing from the seventies. Could the successful story of the Argentine debt restructuring be a lesson for other countries suffering from heavy external debt burdens? The Argentine case shows some very specific features that reduce the likelihood of consider it a valid antecedent in other cases. But, on the other hand, it also shows both the viability of an independent debt restructuring and the strength that the sovereign debtor enjoys in that context. 2. Financial opening and indebtedness in the recent phase of financial globalization 2.1. The Argentinean debt in the long term Before the recent financial globalization process – which initial steps can be dated in the early seventies - Argentina showed low and stable debt indicators. The foreign debt, public and private, was mostly owed to multilateral organizations and governments. It fluctuated in a range of 10% to 15% of GDP from the beginning of the sixties to the mid seventies, as it can be seen in graph 1. As from the mid seventies, the confluence of new international and local factors gave birth to a new stage markedly different from the previous one. In first place, after the oil shock in 1973 the strong expansion of the euro market opened for the country the easy access to international credit. Meanwhile, a deep liberalizing financial reform was 3 implemented in 1977 and was followed by the progressive dismantling of foreign exchange controls to capital account private flows in 1978-80. These changes would jointly operate to completely change the country links with the international financial markets. Graph 1 As can be seen in the graph, the foreign debt/output ratio showed a rising trend between 1976 and 2000. The ratio measured with the PPP exchange rate grew approximately 3 percentage points of GDP per year in this period. The curve is more volatile when the ratio is measured at current exchange rates, with sharp rises in the beginning and end of the eighties as well as in 2002, and a strong fall in 1990-93. These jumps are due to the real exchange rate instability experienced in the period, as can be seen in graph 2.1 The total foreign debt/exports ratio, another standard debt indicator shown in graph 3, complements the mentioned evidence. It rose abruptly since 1977, especially between 1977 and 1982, and never returned to the previous level. The 1976-2003 average much more than duplicates the level registered in the period ending in the mid seventies. Graph 2 Graph 3 2.2. Three stages Between the mid seventies and present times, three main stages can be distinguished in the evolution of the debt. In the first stage, between 1977 and 1982, Argentina went trough a phase of financial opening and accelerated indebtedness that ended up in massive capital flight, exchange rate crisis, devaluation and default. The second stage is a long period of international credit rationing, between 1982 -the Latin American’s debt crisis year- and 1990. The third stage comprises the 1991-2001 period. As the first stage, it was also characterized by financial opening and accelerated indebtedness and exhibited again many of its features. This is the convertibility period, which would also end up in capital flight, exchange and financial crises, devaluation and default. In what follows, we present the main stylized facts of the mentioned stages. 1 Note that the debt ratio is defined as: (d.P*E/y.P), where d is the debt measured in real dollars, P* is the international price level, E the nominal exchange rate, y the real GDP, and P the domestic price level. Therefore, this ratio is affected by variations of the real exchange rate (EP*/P). Ceteris paribus, the real depreciation increases the debt ratio and the real appreciation reduces it. 4 A first fact that deserves to be highlighted is the role played by the private sector in the generation of external financial obligations. In both stages of accelerated indebtedness this sector was initially the most dynamic one. As can be seen in graph 4, the government’s proportion in total obligations declines between 1978 and 1980. Something similar can be appreciated in the period starting in 1991, although in this case the process would be longer. Despite the strong public external debt rise in the nineties, its participation in total debt declined in more than 20 percentage points during that period. Graph 4 The second important fact is that fiscal policies have not been the main factor of the debt growth and the fiscal desequilibrium has not been the main cause of the crises and defaults that followed both phases of accelerated indebtedness. We will consider this issue in detail below in this chapter, when we focus the analysis on the nineties’ evolution. A third fact relevant for the understanding of the indebtness process is that Argentina entered both phases of accelerated indebtness in the context of stabilization programs based on the fixation of the nominal exchange rate. High inflation was experienced in the mid seventies and the same happened in the beginning of the nineties. The opening of the capital account was in both phases adopted together with the launching of antiinflationary programs (jointly with other liberalizing-reform measures in goods, and financial markets). In both cases, the key instrument of the stabilization policy was the fixation of the nominal exchange rate instrumented as an anchor for the stabilization of the prices.2 The stabilization programs based on the exchange rate anchoring and financial opening put in motion a cyclical macroeconomic process (Frenkel, 1983; Taylor, 1998; and Frenkel, 2003a). The exchange rate fixation encourages private capital inflows, induced by the difference between international and domestic interest rates. The aggregated demand expands while inflation declines, although the residual inflation causes the real exchange rate to appreciate. The current account worsens as a consequence of increasing net imports caused by both the exchange rate appreciation and the demand expansion. The external financial needs rise and debt accumulates. This implies that the vulnerability of the economy to negative external financial shocks progressively increases. The domestic financial fragility increases as well. Exogenous shocks may trigger the reversion of the expansionary trend. The change in the trend can also be caused endogenously by a domestic financial crisis. This happened, for instance, in Argentina at the beginning of the eighties. The failed stabilization attempt of the late seventies led to an internal financial crisis that started in early 1980 and developed 2 We are refering to “la tablita” , a program of prefixed devaluations implemented from the end of 1978, and the convertibility regime that established the free convertibility of the peso to the dollar at a 1 to 1 parity. 5 along that year. Finally, the program collapsed in early 1981 leaving a heavy burden of external financial obligations. The mentioned stylized facts help to understand the features of the debt evolution presented in what follows. Table 1 presents the changes in the debt/GDP ratio, and the factors explaining them: the changes in the amount of the debt in dollars and the variations in the real exchange rate and in the GDP. It can be seen that between 1975 and 1980 the debt ratio rose by more than 19 points of GDP, measured with the PPP exchange rate (it passed from 13,2% to 32,4%). The figures in the same table show that this result was hidden by the strong exchange rate appreciation, since the debt ratio calculated with the current exchange rate not only did not rise but fell in almost 4 points of GDP in that same period. Table 1 In 1981, the exchange rate anchor stabilization policy was abandoned, putting end to the first phase of accelerated indebtedness in an appreciated exchange rate context. A new phase followed, characterized by massive devaluations of the peso. These devaluations caused the foreign debt ratio measured in current dollars to reach a peak level close to 60% of GDP in 1982. The figures in table 1 show that the jump in the debt ratio at current prices between 1980 and 1982 (more than 44 GDP points) was to a great extent due to an increase of more than 200% in the real value of the dollar. Nonetheless, the debt in dollars rose 37% between 1980 and 1982. An important factor behind this increment was the rise in the international interest rates that resulted from the policy carried on by the Federal Reserve since 1979. An important jump in the public sector proportion in the country’s foreign debt can also be observed in those years (graph 4). In 1981-82 the public sector ended up absorbing a considerable proportion of the private foreign debt, with the approval of the international banks. This to a great extent explains the jump in the participation of the public sector in total debt. It should also be stressed that no haircut provided relieve to the public debt in the early eighties default situation (it would only come late and in homeopathic doses with the Brady agreement in 1992-93). In the following period of international private credit rationing the debt measured with the PPP exchange rate kept on increasing, though at a slow pace. It increased the equivalent of 10 GDP points between 1982 and 1990 (table 1). The external obligations in dollars continued rising despite the lack of access to the international market (although at a much lower speed than in previous stages). The stagnant output trend also contributes to explain the mentioned rise.3 3 Although the access to voluntary international funding was closed, part of the interest flows accrued in the eighties were accumulated as new debt, i.e. as bank credit involuntary funding, and would end up being recognized and instrumented in bonds with the Brady agreement. 6 Later on, in the nineties, the debt’s rate of growth accelerated again, especially from 1992. The Brady agreement did not provide significant relieve to the debt inherited from the mistaken polices of the late seventies. The achieved haircut was practically insignificant. The main favorable impact of the Brady agreement was on the banks’ portfolios, since they could transform into bonds the defaulted credits, including the past due interests.4 During the 1990-2001 period, the foreign debt/GDP ratio, measured with the PPP exchange rate, rose almost 30 GDP points (table 1). This jump was completely due to the increase in the debt in dollars, which surpassed the accumulated GDP growth. However, it can be seen that the debt ratio measured with the current exchange rate barely rose, as a consequence of the important real appreciation that took place in the period. 2.3. The public debt in the nineties We have seen that the total foreign debt, measured with the PPP exchange rate, increased in almost 30 points of GDP between 1990 and 2001. About 60% of that rise was generated by the private sector. The participation of the private sector was even more accentuated in the early nineties: it originated approximately 70% of the increase in the external financial obligations between 1990 and 1995. The public sector debt issuing was more significant in the second half of the decade, when the international financial conditions worsened. Besides, the placement of public debt in the domestic market started to play a more significant role in those years. The following graph illustrates the public debt evolution in the period. Graph 5 The series in the graph and the figures in tables 2 and 3 allow us to describe the main stylized facts of the Argentinean public sector indebtedness in the convertibility decade. Table 2 Table 3 The analysis of the fiscal accounts allows us to distinguish three periods in the nineties. In the first period, a sharp adjustment in the public accounts is observed. The average deficit, which in the eighties was about 7% of GDP, decreased to less than 1% of GDP in the 1991-94 period. As figures in table 2 show, this was mainly due to an 4 In 1992, before the agreement, the bonds in circulation were only 17% of total public debt, whereas in 1993 they had reached almost 65% of it. On the other hand, foreign currency denominated bonds represented less than 13% of the public debt in 1992, but approximately 57% in 1993. 7 improvement of 6 points of GDP in the national public sector balance result, from which 90% is explained by the primary balance result. The year 1994 was a breakpoint in last decade for three main reasons. In the first place, the social security reform that created the Private Pension Funds was then instrumented. One of the consequences of the reform was a considerable loss in the contributions to the public subsystem. In the second place, the expansion initiated in 1990 was then coming to an end: Argentina would go through the recession associated to the Tequila effect in 1995. In the third place, the government took several measures aimed at compensating for some of the negative effects of the combination of commercial opening and exchange rate appreciation. It did that by lowering the tax burden on the tradable goods production sectors. All of the mentioned factors negatively affected the public finances. Even though, in spite of these negative effects, between 1995 and 1997 the average fiscal deficit was only 2 points of GDP higher than the early nineties deficit. This figure is almost equivalent to the increase in the public social security subsystem desequilibrium caused by the reform. However, after 1997 the fiscal panorama would change significantly. The impact of the Russian and Brazilian crisis in 1998 resulted in a new jump in the country-risk premiums, which had already started rising since mid 1997, after the South East Asian crisis. This, on the one hand, negatively affected the internal demand and triggered a new recession trend. On the other hand, it increased the financial vulnerability of debtors, including the public sector as well as many private agents that were in a net debtor position. Before analyzing this stage in more detail, let us take a look at the association between the fiscal results and the public debt evolution using the figures in table 3. A first important observation refers to the discrepancy between the variation of public sector’s financial obligations and the accumulated fiscal deficit, which represents more than 30 billion dollars in the nineties. The figures show the main reason of this inconsistency: the verification of debts incurred in previous periods but not registered in the fiscal results balance, especially debt with the public sector purveyors and with the social security system’s beneficiaries (skeletons). There had been erroneous liquidations and payment delays, mainly during the 1989-90 period, when the economy experienced two short hyperinflationary episodes. The documentation of past debts was mostly concentrated in the initial stage, between 1991 and 1994. Even though, it should be noticed that the public debt ratio measured as percentage of GDP, was relatively stable up to 1994, in around 30% in the case of total debt and 25% in the case of foreign debt (graph 5). In comparison to the eighties, the 1991-94 phase was in synthesis characterized by a significant improvement in the public accounts and by the relatively ordered absorption of a considerable volume of debt mostly generated in previous periods, i.e. by the 8 regularization of liabilities, many of which were litigious. It is clear from these figures that the standard financial vulnerability indicators did not show evidence of fiscal sustainability problems towards 1994, when the economy was reached by the shock resulting from the Mexican crisis contagion. However, it is undeniable that the high debt burden inherited from the previous phase - a sort of an original fiscal sin of the nineties- is partially hidden by the real appreciation veil. Graph 5 shows the public debt/GDP ratio calculated with the PPP exchange rate. As it can be seen, the curve intersects the 50% line in 1993. The dollarization of the public debt establishes a direct link between the external fragility and the fiscal financial fragility, since taxes are paid in domestic currency. The relevance of this link is stressed by the exchange rate appreciation. Between 1995 and 1997 the public debt/GDP ratio increases, in part as a result of the 1995 recession, and also because of the significant financial aid package led by the IMF, amounting approximately 11 billion dollars. This support enabled the country to quickly recover from the crisis that followed the tequila effect. As can be seen in graph 5, in the expansionary phase that followed the crisis, the debt ratio tended to stabilize again between 35 and 40% of GDP, a relatively low level in comparison to international standards. Again, in spite of the rise in the current deficit and the desequilibrium in the social security system, the standard debt indicators did not suggest fiscal sustainability risk towards 1997, before the beginning of the depression. However, the debt ratio measured with the PPP exchange rate had already reached 60% of GDP. As pointed out above, the Argentina’s macroeconomic panorama would drastically change soon after, since the August 1998 Russian crisis. Table 4 helps us to understand some key features of the fiscal evolution in this stage. Public sector’s deficit took a significant rising path that would lead it to reach about 6 points of GDP in 2001, despite the many rounds of contractive fiscal policies instrumented to stop the trend. Table 4 In the table 4 we compare the average desequilibrium of the depression period to the deficit registered in 1994. In 1998-2001 the average accrued annual deficit (amounting 11.5 billion dollars) was 7.1 billion dollars higher than the deficit registered in 1994. Which were the sources of this increase? As it can be seen, it was chiefly due to the rise in the interest payments (+6.8 billions) and, in the second place, to the amplification of the social security system gap (+4.9 billions). Contrary to the standard interpretation, a relatively minor figure (+582 millions) is explained by the desequilibrium in the result of the provincial administrations, though it is true that it was on an increasing path. 9 The table also suggests that the procyclical fiscal policies implemented were not ineffective: they produced a substantial increase in the primary surplus of more than a 5 billion dollar annual average (without including the public social security results), though that was not sufficient to compensate for the rises in the interests item and in the social security system desequilibrium. The explosive trend in the public debt interests account is also observed in the following table. Table 5 The weight of interests on tax resources, which had slightly increased after 1994, takes a fast upward trend after 1996. In 2000, that ratio was nearly 19%, duplicating the ratio registered in the middle of the decade. This was in part due to the decrease in tax revenues caused by the recession, but it was fundamentally originated in the rise in the average interest rate paid by the public debt. The average interest rate of the total public debt went from 5,8% in 1996 to 9,4% in 2001. Considering that this is an average rate, it is easy to see that the marginal rate rise was extremely higher. The rising path of the interest rate is associated with the increasing trend in the country- risk premium (the two variables are narrowly correlated in the 1997-2001 period). These rising trends are the main factors behind both the consolidate deficit trajectory and the explosive path taken by the public debt. This is illustrated in graph 5. Between 1997 and 2001, in only four years, the public debt/GDP ratio increased by more than 20 percentage points. 3. The macroeconomic performance before and after the default 3.1. The nineties: from euphoria to depression The basic plot of the macroeconomic story of the late nineties was quite simple. To start with, the negative financial turnaround in the foreign environment experienced in 1997- 1998, after de South East Asian and Russian crises, found the Argentine economy with a significant and growing current account deficit, a considerably appreciated currency and a visible lack of policy instruments to deal with this problem, given the rigidities of the adopted macroeconomic rule. No surprise, in these conditions the country-risk premium jumped upwards and the access to foreign funds became more and more problematic. As explained in the previous chapter, the subsequently increased interest burden had a negative impact on all borrowers, including the public sector. Given that the government lacked of other policy instruments, restrictive fiscal policies had to bear with the main burden of the adjustment to the new situation. The official story used to say that fiscal discipline would entail stronger confidence, and consequently the risk premium would fall bringing interest rates down. Therefore, domestic expenditure would recover pushing the economic out of the recession. Lower 10