Principal Advisor, Research Department, Inter-American Development Bank; and ex. Chief Economist, Banco Central de la República Argentina
Writing on recent Argentine economic history is no easy task. There is much to discuss, and much has been said. The authors of this chapter are assisted by a clear terms-of-reference and a well-defined approach. Following closely the methodological stance of the project, they are able to condense fifty-seven turbulent years of Argentine economic history into just twenty or so pages. This is an incredible feat; the authors deserve hearty congratulations! The paper gives an excellent, succinct description of the main events and clearly illustrates the association between Argentina’s fiscal and monetary travails. In these brief comments, I will not attempt to be comprehensive but rather will limit myself to just three main takeaways from the chapter where I hope I can add value.
The first takeaway relates to the strong view that Argentina’s woes were essentially fiscal and that monetary instability and high inflation including the bouts of hyperinflation in the end boil down to fiscal impropriety. Overall, I have sympathy for this perspective, but there is surely a richer set of dynamics, and there is a danger of pushing the “it’s all fiscal” doctrine too far. The main equation defining the government’s budget constraint provides a clear link between the fiscal and the monetary but does not ascribe causality. The causal mechanism in mind seems to consider spending as set, but then if financing through taxes or debt is not available, residual financing through money creation drives inflation. But high (and uncertain) inflation may have an impact on activity and tax revenues, affecting the level of funding. In the case of Argentina, high and uncertain inflation, as well as the actions to lower inflation, surely had an impact on activity and revenues.
The authors place emphasis on the fact that the 1976–1990 period was the one with the highest fiscal deficits. It was also the period of lowest growth. The authors’ estimate of trend growth pre-1980 was more or less reestablished after that decade. Arguably, the 1980s account for why Argentina continues to lie below that trend today. I won’t repeat everything leading up to the lost decade, but the 1977 financial liberalization seems key. Coupled with weak regulation and supervision, severe financial vulnerabilities arose. At the end of the 1970s, commodity prices fell and then US interest rates rose, and the dollar appreciated. A banking crisis ensued. The first major bank to fail was in March 1980, followed quickly by the failure of three more banks, all before the 1981 devaluation—which then deepened the crisis. Decisions to resolve the crisis proved expensive. Exchange rate guarantees, full deposit insurance, and the liquidation of seventy-one banks increased fiscal woes.1 The above contributed to the 1982 default, but the causal links are not obvious. The fiscal problems surely played an important role in Argentina’s subsequent inflation trajectory, but like most airplane accidents, many things went wrong: weak banking oversight, poorly sequenced liberalization, exogenous shocks, decisions taken to resolve the crisis and default, all with growth, revenue, and availability-of-financing—in addition to spending—implications.
The second takeaway on which I wish to comment is the authors’ view that, because there was a structural break in inflation, as fiscal deficits fell and debt stabilized, the currency board “worked.”2 While this is one definition of “worked,” ultimately the project failed; I would argue that it did not produce the fiscal discipline needed to guarantee its survival, given a set of significant shocks. There has been much discussion of late regarding de facto dollarization and consequent vulnerabilities. I don’t want to rehash those discussions and would point the interested reader to meta-analyses on the origins of the crisis.3 Suffice to say, I think the authors are broadly right to focus on the fiscal. After the Tequila shock, Argentina grew at 5.5 percent in 1996 and over 8 percent in 1997 but ran a 1 percent primary deficit in the first year and just scraped a primary surplus (around 0.3 percent) in 1997.4 To withstand negative shocks that would surely emerge at some point, Argentina should have saved more, as the International Monetary Fund argued at the time.
Argentina was hit by strong negative shocks, such as the November 1997 attack on the Hong Kong currency board, the August 1998 Russian default, falling commodity prices, and the January 1999 Brazilian devaluation. One of the fascinating discussions, and which recently came back to the fore, was that regarding full dollarization in early 1999. It might be of interest what didn’t happen, as well as what did. President Menem asked the president of the central bank, Pedro Pou, to consider the options to dollarize. The advice provided by the central bank was A) against unilateral dollarization, B) that a monetary union with an Argentine seat on the FOMC would not be palatable to the US authorities (!), but C) that if Argentina wished to dollarize, the option to consider would be in the context of a US–Argentina monetary treaty. The treaty would state that the United States would not be a lender of last resort, that the US authorities would not regulate Argentine banks or markets, but the treaty would advance a seigniorage-sharing scheme.5 Warnings against dollarizing without wide support from Congress and the population were also included. Two missions to the United States ensued. Despite resistance at the Federal Reserve and mixed views at the US Treasury, Larry Summers (US Treasury secretary at the time) gave a speech with remarkable resonance to the Argentine technical position. However, Floridian Republican senator Connie Mack put forward hasty legislation in favor of seigniorage sharing. The bill was unacceptable to the US executive branch, and Democrats voted it down. There was also no widespread support in Argentina, and the idea was shelved.
What would have happened if Argentina had dollarized? Would the scheme have survived 2002? Would the cost of having had no monetary policy have exceeded the benefits of lower interest rates? Any answer would be a judgment call. More generally, the relative merits of dollarizing decrease if monetary shocks can be contained, if an alternative credible nominal anchor can be developed, and (hence) when an independent monetary policy can provide a floating exchange rate that stabilizes rather than destabilizes. The experience of the currency board is only partially informative regarding the possible success of dollarizing. While dollarizing may reduce interest rates, without fiscal discipline debt can spiral, requiring debt restructuring from time to time.6 And dollarizing does not prevent a new currency being introduced. If, as the authors suggest, the bottom line is fiscal profligacy, then, while dollarization is not orthogonal, it may not solve the underlying problem. Other countries in the region have successfully developed alternative credible nominal anchors and contained fiscal excesses to some degree.7 In the case of Argentina’s, there is clearly more work to be done to attain this objective.
Despite the negative shocks of the late 1990s and their impacts on economic activity, the required fiscal adjustment and the required adjustment in the current account to ensure that the currency arrangement was sustainable were not large.8 Even as late as the first quarter of 2001, there seemed to be no technical reason to prevent them from being attained. But policies tended to be reactive rather than proactive, and then they became counterproductive, and politics appeared to often get in the way.
This brings me to comment number three regarding multiple equilibria stories. The authors appear drawn to one particular possibility, following Calvo (1988) and Ayres et al. (2018). But they offer no real evidence in favor versus a simpler alternative of higher interest costs due to a rising risk of default. And there are also alternative multiple equilibria hypotheses to consider. Of particular importance was the trajectory of bank deposits (critical to rolling over the public debt) and the role of the IMF. The authors state that there was a continuous outflow of deposits through 2001, but on closer inspection there were four distinct runs, the first starting in 2000.9 The authors could have explained more regarding Argentina’s “systemic liquidity policy” consisting of high bank liquidity requirements, international reserves in excess of the two-thirds backing required by the Convertibility Law, and an innovative US$6bn plus Contingent Repo Facility signed with a group of international banks and supplemented by loans totaling US$1bn from the Inter-American Development Bank and the World Bank.10 Arguably, the strong liquidity position allowed the currency board to survive so long. The third run, in June–July 2001, was the most serious to that point, but was halted, temporarily, by the IMF agreement signed in August 2001.
The role of the IMF as a type of lender of last resort has come under close scrutiny. But in such circumstances, the IMF is placed in a tough spot. If the IMF agrees to support a country without reservation, then this may prevent a run, ruling out a no-rollover bad equilibrium. But there is a danger of “moral hazard”—in other words, the country may choose more risky strategies (or gamble for resurrection) that the IMF would disapprove of, and for a systemic country may risk international financial instability. On the other hand, if the IMF does not support without reservation, then the country may be vulnerable to the bad equilibrium and to avoid that outcome may be forced into a safe strategy, which the IMF should be happy to support. This simple logic suggests that in a one-shot game, there may be no equilibrium in pure strategies, and the only feasible equilibrium is in mixed strategies where the IMF may support with some probability (not without reservation) and there is little clarity about whether the country is playing safe.11 This is my interpretation of what was going on through much of 2001. Formally, Argentina had an IMF program. But there were doubts about whether each subsequent set of quantitative targets would be met, whether the program would continue, and whether the next disbursement would arrive. And Argentine policymaking became difficult to interpret.12 Then came the IMF agreement of August 2001. While this agreement did stop the bank run, temporarily, it also signaled that Argentina needed to renegotiate its debt, without providing advice nor sufficient financing to do so. The zero-deficit policy then failed, the IMF withdrew, a new run commenced, and the currency board fell.13
Let me make one comment on the resolution of the 2002 crisis that often seems to go unquestioned. Namely, I would suggest that there were many potential ways to exit the currency board, each with associated costs, benefits, and distributional consequences. The policy of a systemwide asymmetric pessification (where deposits were converted from dollars to pesos at 1.4 to 1, and dollar loans were converted to pesos at 1 to 1), coupled with the depreciation to over 4 pesos to 1 dollar from the currency board’s 1:1, was not the only way and the choice was not random. One alternative would have been to float with guidelines on how banks and firms should renegotiate loan contracts.14 The blanket rule inflicted losses on private banks that had significant dollar loans and deposits and protected public banks whose balance sheets were largely in pesos, such as the Banco de la Provincia de Buenos Aires, which had a weak balance sheet at that time. In contrast, the private banks had substantial capital. Indeed, I suspect if they had had more capital, then pessification would have been more asymmetric; the objective appeared to be to redistribute bank capital. Moreover, while Argentina’s public sector was short “external” dollars (public external debt exceeded reserves), the private sector was substantially long.15 Any owner of a business with dollar debt with the local financial system and money saved abroad received a huge transfer. And consumers with dollar local loans received transfers irrespective of their “needs.” The policy appeared designed to hit private banks and protect certain public ones, and local businesses received transfers. Following the money makes the political economy of the exit policy clearer. The result was a dismantling of the private financial system, likely deepening the collapse. It did, however, allow the economy to recover quickly from the crisis with very little bank credit.
Given the various banking crises in Argentina (1982, 1995, and 2002) and in other countries, follow-up work on banking could nicely complement this project. Not all banking crises have fiscal origins, so this would provide nuances to the “it’s all fiscal” doctrine. Moreover, the region has come a long way in managing financial system risk. First, the quality of regulation and supervision to control risks ex ante has improved. Second, after interesting experiments, most countries now have some type of deposit insurance in place. In the case of Argentina, the contagion within the system during the Tequila crisis amply illustrated the risks of its elimination. Finally, bank resolution techniques have improved. Argentina’s novel system implemented after the Tequila crisis is a case in point. The authors’ preferred measure of banking sector contingent liabilities is M2/GDP, but according to this measure, the 2015 contingent liabilities of banks in the European Union was about 24 trillion euros, or some 146 percent of EU GDP, dwarfing those of Argentina, which at most were a modest 20 percent of GDP. This measure essentially values all bank assets at zero. If ex ante risks can be contained and resolution techniques can minimize losses, this appears too crude a measure. A wider project might be entitled “The Fiscal, Monetary, and Banking History of Latin America” and could focus on better measures of financial sector risks, and how they were and can be managed.
To conclude, this is a fascinating story told in a succinct way in part owing to the benefit of a strong methodological choice. A tremendous amount of information is summarized, and yet it leaves the reader wanting more. It is akin to an award-winning short documentary with a strong message from the director(s). And despite the issues I have raised, if countries had actually followed the advice implicit in this chapter and the other country chapters, I have no doubt that economic stability would have been enhanced and the region would be considerably more prosperous today. The project has been an extremely valuable endeavor, and I have greatly enjoyed acting as a commentator.
1. See especially Diaz Alejandro (1985), Baliño (1987), and D’Amato and Katz (2018) on the lead-up to the 1980 banking crisis and its resolution.
2. It’s worth detailing that the “currency board” was not a “pure currency board.” International reserves only had to be 80 percent of the monetary base in 1991, falling to two-thirds, although there was also a restriction on the growth in Argentine dollar bonds on the central bank’s balance sheet that made up the residual backing. Also, at the start, the currency could fluctuate in a very fine band (this was shut down after the Tequila shock), and several currencies could compete for transaction purposes within the country.
3. See Cline (2003), International Monetary Fund (2004), and Powell (2002) for an evaluation of several theories about the crisis.
4. Data from the International Monetary Fund’s World Economic Outlook, October 2018.
5. For further discussion on dollarization and the idea of a monetary treaty, see Guidotti and Powell (2002).
6. See Powell and Sturzenegger (2000) for a discussion of the impact of dollarization on interest rates.
7. See Mariscal, Powell and Tavella (2018) on how the “credibility” of Latin America’s inflation targets developed.
8. See the estimates in Powell (2003).
9. In Powell (2002), I discuss the four runs.
10. See Calomiris and Powell (2002) for a description of various elements of banking oversight introduced in the post-Tequila period. I normally like to describe these measures as seat belts, very useful in 30 mph crashes. But such rules will not protect passengers if the truck is driven off a cliff!
11. See Powell and Arozamena (2003) for a game theoretic model that encapsulates these ideas.
12. Such measures included the use of “monetary policy” under the currency board (i.e., the relaxation of liquidity requirements to promote bank credit), a move from the dollar to a basket as the exchange rate reference currency, a system of export subsidies and import taxes to simulate a devaluation, and the zero-deficit law.
13. The policy of announcing a zero planned deficit turned out to be counterproductive. One argument was that while many knew Argentina was in a tough spot, this was not common knowledge. When Argentina announced the policy, then every investor knew all others were refusing to provide any additional financing. See Mussa (2002) for a critical examination of the 2001 IMF agreement.
14. For example, in Ecuador after a maxi depreciation (followed by dollarization), small loans were restructured by extending loan maturities and gradually introducing increasing payment schedules, and there was a largely voluntary scheme for large borrower workouts avoiding large transfers or bailouts (see International Monetary Fund 2000, para. 49).
15. On public versus private sector balance sheets, see Powell (2002).
Ayres, João, Gaston Navarro, Juan Pablo Nicolini, and Pedro Teles. 2018. “Sovereign Default: The Role of Expectations.” Journal of Economic Theory 175 (C): 803–812.
Baliño, Thomas J. T. 1987. “The Argentine Banking Crisis of 1980.” Working Paper WP/87/77, International Monetary Fund.
Calomiris, Charles W., and Andrew Powell. 2000. “Can Emerging Market Bank Regulators Establish Credible Discipline? The Case of Argentina.” In Prudential Supervision: What Works and What Doesn’t, edited by Frederic Mishkin. Chicago: University of Chicago Press.
Calvo, Guillermo A. 1988. “Servicing the Public Debt: The Role of Expectations.” American Economic Review 78 (4): 647–661.
Cline, William R. 2003. “Restoring Economic Growth in Argentina.” Policy Research Working Paper wps3158, World Bank.
D’Amato, Laura, and Sebastian Katz. 2018. “Una constante en la evolución macroeconómica Argentina: dinero, deuda y crisis (1945–2015).” In La Nueva historia económica de la Argentina: temas, problemas, autores. El último medio siglo. Ensayos de Historiografía Económica. Desde 1810 a 2016, edited by Roberto Cortés Conde and Gerardo Della Paolera. Buenos Aires: Edhasa, Ciudad Autónoma de Buenos Aires.
Diaz Alejandro, Carlos. 1985. “Good-bye Financial Repression, Hello Financial Crash.” Journal of Development Economics 19 (1–2): 1–24.
Guidotti, Pablo, and Andrew Powell. 2002. “The Dollarization Debate in Argentina and Latin America.” In International Financial Markets: The Challenge of Globalization, edited by Leonardo Auernheimer. Chicago: University of Chicago Press.
International Monetary Fund. 2000. “Ecuador: Selected Issues and Statistical Annex.” Staff Country Report 00/125, International Monetary Fund.
International Monetary Fund. 2004. Report on the Evaluation of the Role of the IMF in Argentina, 1991–2001.” Independent Evaluation Office (IEO) of the IMF, IEO Publications.
Mariscal, Rodrigo, Andrew Powell, and Pilar Tavella. 2018. “On the Credibility of Inflation-Targeting Regimes in Latin America.” Economia 18 (2): 1–24.
Mussa, Michael. 2002. Argentina and the Fund: From Triumph to Tragedy. Policy Analyses in International Economics (Book 67). Washington, DC: Institute for International Economics.
Powell, Andrew. 2002. “Argentina’s Avoidable Crisis: Bad Luck, Bad Economics, Bad Politics, Bad Advice.” Brookings Trade Forum 2002: 1–58.
Powell, Andrew, and Leandro Arozamena. 2003. “Liquidity Protection versus Moral Hazard: The Role of the IMF.” Journal of International Money and Finance 22 (7): 1041–1063.
Powell, Andrew, and Federico Sturzenegger. 2000. “Dollarization: The Link between Devaluation and Default Risk.” Unpublished manuscript, Business School, Universidad Torcuato Di Tella.