Argentina s Crisis and Recovery: A Demand Side Story Alberto Martin November 2013

Transcription

1 Argentina s Crisis and Recovery: A Demand Side Story by Ariel Burstein and Ivan Werning Alberto Martin November 2013

2 Overview Revisit argentine experience : Prolonged recession: 5.4% contraction between 1998 and 2001 (real GDP/capita) Collapse: 10.5% contraction in 2002 Role of fixed exchange rate Overvaluation and competitiveness problems Speculation and credibility problems

3 What the paper (slides) does Overview of main facts ( ) Contraction in GDP, rise in unemployment (9% fall in employment) Gradual decline in prices and wages (approx. 2.5% overall) Rise of real exchange rate (approx. 14%) Calibrate sticky price / fixed exchange rate model: contribution of shocks Foreign borrowing: Russian crisis, sudden stop Real exchange rate: Brazilian devaluation Terms of trade shock Bottom line: large effects under sticky prices / fixed exchange rate

4 The Model SOE with sticky prices Continuum of households Consume: Nontraded goods Traded goods: imported and home good Two types of households: Optimizing: subject to debt limit Hand to Mouth (HTM) Domestic production combines intermediate goods Intermediates produced by monopolistically competitive firms Price and/or wage rigidity (price plans)

5 Old nail with a new hammer Friedman (1953) With nominal rigidities, flexible exchange rates smooth adjustment to real shocks Recently revisited in the context of European crisis Schmitt Grohe and Uribe (2010, 2011, 2012), Farhi and Werning (2012), Fornaro (2013) Stress costs of fixed exchange rates in presence of nominal rigidities Emphasis on demand

6 Main results Interest rate shock / sudden stop (temporary) Appreciation shock (permanent) Negative terms of trade shock (permanent) Fiscal shock (temporary) Corralito (expected redistribution) shock Expected redistribution from optimizing to HTM agents Fall in consumption: under fixed prices, significant contraction

7 General reaction Very interesting exercise (and close to my heart!) Still preliminary Nothing major to disagree with Qualitatively, most of the results as expected Quantitatively, large effects of price (exchange rate) rigidity Incidentally, great data to explore!

8 Comments Specific: shocks are unanticipated Makes comparison between flexible and fixed prices a bit unfair (but ok) Seems especially important to capture speculative, credibility aspects General 1: Peg is costly, but so is abandoning it Foreign denominated debt: defaultable Balance sheet effects and the financial system (Cook 2004, Gertler, Gilchrist, Natalucci 2006) These costs (and their anticipation) might have been important factors in the crisis General 2: Lessons for Europe Some similarities, but also important differences with Argentine experience

9 Defaultable debt Argentina accumulated large debt in foreign currency Made possible by the peg Made it harder to abandon the peg Important part of the story: Ex ante: peg reduces cost of borrowing increased credibility, lower inflation Ex post: abandoning the peg raises likelihood of default loss of reputation loss of output: collateral damage to domestic banking system Shocks affect the benefits but also the costs of abandoning peg Foreign financing shocks: reduce output directly but also lower costs of default (e.g. Arellano 2008)

10

11 Defaultable debt Argentina accumulated large debt in foreign currency Made possible by the peg Made it harder to abandon the peg Important part of the story: Ex ante: peg reduced cost of borrowing increased credibility, lower inflation Ex post: abandoning the peg implied a default loss of market access (reputation), trade, etc.. loss of output: collateral damage to domestic financial system Implications: Speculative or self fulfilling aspects Shocks affect the benefits but also the costs of abandoning peg Foreign financing shocks reduce costs of default (e.g. Arellano 2008)

12 More on the financial system Argentine devaluation entailed large redistributions Dollar loans were pesified (at the fixed exchange rate) and indexed by CPI Dollar deposits were pesified at a favorable exchange rate (1.4 times the peg) and indexed by CPI Government compensated banks through long term bonds (+ 10% of GDP) Nonetheless, sustained collapse of credit (still to recover...) Similar to corralito shock? Agents understand that devaluation becomes increasingly likey There will be (random?) redistributions when that happens

13 Source: World Bank

14 More on the financial system Argentine devaluation entailed large redistributions Dollar loans were pesified (at the fixed exchange rate) and indexed by CPI Dollar deposits were pesified at a favorable exchange rate (1.4 times the peg) and indexed by CPI Government compensated banks through long term bonds (+ 10% of GDP) Nonetheless, sustained collapse of credit (still to recover...) Similar to corralito shock? Agents understand that devaluation becomes increasingly likely (check deposits) There will be (random?) redistributions when that happens

15 Lessons for Europe Characterize similarities between Argentina, Greece, Spain Most notably, sluggish adjustment of RER despite the slowdown Even within Europe, heterogeneity Price / wage flexibility of southern Europe relative to Argentina Clearly, also important differences with Argentina

16 Default Argentina, debt substantially lower than in GIIPS Mostly in the hands of foreigners In Europe, it is mostly (and increasingly!) in hands of domestic residents Potential costs of default might be higher (benefits lower)

17 Debt Holdings: Residents vs. Non Residents Spain Bps /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/2012 Residents Non residents spread (RHS) Bps 01/03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/ /03/ /09/2012 Greece Residents Non residents spread (RHS) Source: Broner, Erce, Martin and Ventura (2013)

18 Financial system As in Argentina, abandoning the peg is bound to generate important redistributions within financial system Creditors to debtors? Uncertainty But completely different order of magnitude!

19 Source: World Bank

20 Sudden stops within the Eurozone Argentine sudden stop : reduction of CA deficit by 4% of GDP between 1998 and 2001 What about Europe?

21 Total and private capital inflows, Spain Greece Total inflows Private inflows Total inflows Private inflows Q1 2002Q3 2003Q1 2003Q3 2004Q1 2004Q3 2005Q1 2005Q3 2006Q1 2006Q3 2007Q1 2007Q3 2008Q1 2008Q3 2009Q1 2009Q3 2010Q1 2010Q3 2011Q1 2011Q3 2012Q1 2012Q3 2013Q1 2002Q1 2002Q3 2003Q1 2003Q3 2004Q1 2004Q3 2005Q1 2005Q3 2006Q1 2006Q3 2007Q1 2007Q3 2008Q1 2008Q3 2009Q1 2009Q3 2010Q1 2010Q3 2011Q1 2011Q3 2012Q1 2012Q3 2013Q1 Inflows expressed as a percentage of 2007 GDP (source: Bruegel)

22 Eurosystem finance: TARGET2 net balances

23 Summing up. Interesting, promising project Fixed exchange rate (currency union) might be ill advised But once it is in place, costly to abandon Anticipation of these costs potentially significant in Argentine crisis Some similarities, but also major differences between Argentina and GIIPS