Continuum of exchange rate regimes:From flexible to rigidIMF classificationSchemes for de facto classificationAdjusted de jure classification schemesDe facto classification schemesThe de facto schemes do not agreeCorrelations Among Regime Classification SchemesThe MEA Dept. now has its own “de facto classification” -- still close to official IMF one.Bénassy-Quéré et al (2004): correDo exchange rate regimes matter “really”?Most important finding of last 10 yearsAdvantages of fixed rates, cont.Which dominate: advantages of fixing or advantages of floating?Performance by category is inconclusive.Exhaustive survey of classificaiton schemes by Tavlas, et al (2006) findsNew popularity in 1990s ofinstitutionally-fixed cornerAre countries moving to the corners?The Corners Hypothesis has ebbedPossible rationales for the corners hypothesis, cont.Bottom line on corners hypothesis:Level of financial development Aghion, Bacchetta, Ranciere & Rogoff (2005), “Productivity Growth and the Exchange Rate RegimLevel of financial development, cont. Husain, Mody & Rogoff, “Exchange Rate Regime Durability and Performance in DevelopingAn old conventional wisdomTerms-of-trade variability returnsHow would it work operationally, say, for oil-exporter Iraq?Inflation targeting is the new reigning orthodoxyWhy is PEP better than CPI-targetingfor countries with volatile terms of trade?More moderate versions of PEPAppendicesAppendix I5 reasons China should let RMB appreciate, in its own interestChina’s sterilization of reserve inflow 2003-2006Longer-run perspective:Balassa-Samuelson relationshipEstimation of B-S relationship for 2000(118 countries, PWT)Does B-S relationship have predictive power?How should changes in real exchange rate, when necessary, be achieved?What would new regime be?What about the currency reform announced in July 2005?The post-July-2005 de facto regime for the RMBValue of the Yuan in terms of other major currencies(1) RMB’s implicit weights on major currencies – Frankel & Wei (2007)(2) Implicit RMB weights on minor currenciesand estimate of gradual trend away from the dollarWhat about the new announced changeof May 2007?Appendix IIAppendix IIIProfessor Jeffrey Frankel – Kennedy School of Government – Harvard UniversityExchange Rate Regimes:Current Issues in Research and PolicyIMF Institute, May 25, 2007Jeffrey FrankelHarpel Chair, Harvard UniversityProfessor Jeffrey Frankel – Kennedy School of Government – Harvard UniversityTopics to be covered1. Country classification, de jure vs. de facto 2. Advantages of fixed rates3. Advantages of floating rates4. Which regime dominates?1. Tests of economic performance 2. Conceptual frameworks for evaluation: • Traditional OCA criteria• 1990s emerging markets criteria5. Intermediate regimes and the corners hypothesis6. Financial development7. Terms-of-trade shocks return.• New proposal: Peg the Export Price (PEP)8. Addendum: The case of the RMBProfessor Jeffrey Frankel – Kennedy School of Government – Harvard UniversityContinuum of exchange rate regimes:From flexible to rigidFLEXIBLE CORNER1) Free float 2) Managed floatINTERMEDIATE REGIMES3) Target zone/band 4) Basket peg5) Crawling peg 6) Adjustable pegFIXED CORNER7) Currency board 8) Dollarization9) Monetary unionProfessor Jeffrey Frankel – Kennedy School of Government – Harvard UniversityDe jure regime ≠ de facto1. Most “fixed” aren’t: Countries declaring a peg, often abandon it.Only 6 major open economies had kept a peg > 5 yrs.-- Obstfeld & Rogoff (1995).Mean duration of pegs in Western. Hem.: 10 months -- Klein & Marion (1997).2. Most “floating” aren’t: “floaters” Var.of E (vs. Res) not > fixers’≡ “Fear of Floating”-- Calvo & Reinhart (2002).3. Most basket pegs aren’t. It takes more than 100 observations for an observer to distinguish a true basket peg statistically --Frankel, Schmukler & Serven, “Verifiability & the Vanishing Intermediate Ex.R. Regime” (2000)4. Contrary to popular assertion, the East Asian crisis countries had not declared $ pegs pre-1997.5. IFS abandoned its de jure classification tables after 1999.Professor Jeffrey Frankel – Kennedy School of Government – Harvard UniversityIMF classificationLast de jure (end-1999). Of 185 Fund members,• Have given up own currencies: 45– EMU: 11– CFA Franc Zone: 14 – E.Caribbean CA 6– “dollarized” 8• Currency boards: 8 (minus Argentina = 7)• Intermediate regimes: 89– pegs to a single currency 30– pegs to a composite 13– crawling pegs 5– horizontal bands 7– crawling bands 7– managed floats 26• “independent floaters”: 51“De facto” (end-2004)4112 (plus Slovenia in 07)1469 (plus Montenegro in 06)710433 (incl. China)865 (minus Slovenia in 06)15135Professor Jeffrey Frankel – Kennedy School of Government – Harvard UniversitySchemes for de facto classification• have themselves been surveyed by G.Tavlas, H.Dellas & A. Stockman, “The Classification and Performance of Alternate Exchange-Rate Systems,”March 2006,• and divided into two classifications, viewed as:– “mixed de jure-de facto classifications, because the self-declared regimes are adjusted by the devisers for anomalies.”– Vs. “pure de facto classifications because…assignment of regimes is based solely on statistical algorithms and/or econometric estimation.”Professor Jeffrey Frankel – Kennedy School of Government – Harvard UniversityAdjusted de jure classification schemes1. Ghosh, Gulde, Ostry & Wolf (1995) identify “peggers” who in fact devalue often.2. Bubula & Otker-Robe (2002): New tables of IMF MEA Dept. adjust by consulting IMF economists, and movements in reserves & exchange rates.3. Reinhart & Rogoff (2003, 04): add category of “free falling”, use parallel rates in place of official, and look at longer-term frequencies of classification.4. Eichengreen & Leblang (2003): using historical sources, extend back to 1880.Professor Jeffrey Frankel – Kennedy School of Government – Harvard UniversityDe facto classification schemes1. DeGrauwe & Schnab (2005): “z-score” measure of exchange rate variability.2. Ghosh, Gulde & Wolf (2002): consensus intersection of de jure and a “z-score” measure of ex.rate variability.3. Levy-Yeyati &