Last week, the Central Bank of Argentina let the nominal exchange rate rise from 6.7 pesos per dollar to 8. This devaluation mounted on top of another 10% devaluation that had occurred at a lower pace since early December 2013. Can we consider this as a currency or balance of payments crisis? Because the very notion of an economic crisis —whether it is fiscal, financial or a balance of payments crisis— is always blurry, let me start by providing what I believe is a shared definition of a balance of payments crisis. This happens when the Central Bank unsuccessfully tries to keep the nominal exchange rate around certain value and is forced —after loosing a substantial amount of foreign exchange (FX) reserves— to implement a discrete devaluation or to let the domestic currency depreciate.
Has this happened in Argentina recently? These are the facts. By July 2011, the stock of FX reserves was around U$ 52 billions. The exchange rate was 4.1 pesos per dollar, only 7% higher than in July 2009 (3.85 pesos per dollar). In August 2011 —three months before the presidential election— the Central Bank started loosing reserves. In November 2011, just a few weeks after Cristina Kirchner was re-elected, FX reserves were U$ 46 billions. Since the depletion of FX reserves showed no sign of stopping, the authorities started to implement a series of measures to limit the demand for foreign currency. The controls triggered the blossoming of black markets. FX reserves kept falling, softly during 2012 and very rapidly since early 2013. In November 2013, after a poor mid-term election, the president fired the governor of the Central Bank and put in charge a new economic team. FX reserves were U$ 33 billions and the exchange rate had reached 6 pesos per dollar. As I mentioned above, a week ago, the price of US dollar jumped to 8 pesos. FX reserves are now close to U$ 28 billions. The Central Bank has managed to keep the exchange rate at 8 pesos at the cost of loosing U$ 150/200 millions of reserves per day. There are still widespread expectations of further devaluation and few believe that authorities can sustain the exchange rate at the new level, especially because the rate of inflation has accelerated above 30% annually. In short, FX reserves have so far fallen by 46% and the exchange rate has risen by 95%. If it looks like a dog, walks like a dog and barks like a dog, then… it's probably a balance of payments crisis.
Some analysts still reject the idea that Argentina is going through a balance of payments crisis. Matías Vernengo´s recent piece is a good example. He argues that “this is not a balance of payments crisis” and supports his claim by pointing out the fact that Argentina has a small current account deficit and that the real exchange rate (RER) is not as overvalued as in Brazil. He is right: the current account deficit was only 0.5% of GDP in 2013 (although it would higher without the import controls) and the RER is certainly not as overvalued as in Brazil (which, by the way, is very overvalued). But these facts do not substantiate the claim that Argentina is not experiencing a balance of payments crisis. Matías seems to miss the important point that as long as expected depreciation at the exchange rate that the Central Bank is defending is higher than the yield of domestic assets, there would be an excess demand for foreign currency that would eventually lead to the depletion of FX reserves and the collapse of the domestic currency. Although it would be a rare event, this could even happen in a country with a current account surplus.
This kind of inconsistency was at the root of current Argentine crisis. Starting in 2007, inflation has been in the range of 25-30% per year. Since 2010, the central bank began to use the exchange rate as the main nominal anchor for inflation. During the two years before the 2011-election, domestic prices rose 54% whereas the nominal exchange rate only 12%. As a result the RER appreciated significantly. Nominal interest rate during the same period was about 10% (negative 15% in real terms). A policy like this is clearly unsustainable and at some point expectations of devaluation/depreciation would start to emerge. The 2011-election served as an expectation-coordination device. Since governments do not take unpopular measures before elections, but they typically do after them, people began to anticipate a devaluation in mid-2011. This was why the run against FX reserves started just before the election.
Almost a year ago, I began warning about these inconsistencies and to argue that the controls imposed since late 2011 would not solve the problem but would only exacerbate it. I argued that a RER correction would be unavoidable and that the government should take care of the inconsistences as soon as possible. I was aware that a real devaluation would have a contractionary effect. My fear was (and still is) is that a RER correction done through a balance of payments crisis could lead to an exchange rate overshooting and a devaluation-inflation spiral beyond the government’s control. Such a scenario would imply a significant fall in real wages and employment. This should be avoided.
Formulations like mine seem to be the source of another confusion in Matias’ analysis. He argues that people like Luiz Carlos Bresser Pereira, Roberto Frenkel and me were advocating for a devaluation because we support the idea that maintaining a competitive real exchange rate (CRER) is good for growth. I do support a CRER strategy for development, but if you read my articles (here in English; here and here in Spanish), you will see that my claim was based on the inconsistencies of macroeconomic policy and not on my frustration about the abandonment of the competitive RER strategy that Argentina carried out between 2002 and circa 2008.
The source of Matias’ confusion is, in my view, clearly exposed in this piece. He claims that growth is wage-led, thus real exchange rate devaluations affect growth negatively (as in the Krugrman-Taylor paper) and not positively (as in the Frenkel-Taylor paper). Are the two papers by the same author contradictory? No, the problem is that Matías confuses an important distinction between short-run and long-run effects of the real exchange rate on economic performance. In Krugman-Taylor, a real devaluation (a change in the RER) has a negative effect on output and employment in the short run; in Frenkel-Taylor, a competitive RER level has a positive effect on long-run growth. These are theoretical arguments that should be evaluated empirically in each historical circumstance (I’ve done some work on the latter here and here), but we can’t properly do that if we misunderstand the economic reasoning behind them.