Argentina's Experience with Parallel Exchange Markets: 1981-1990

Board of Governors of the Federal Reserve System International Finance Discussion Papers Number 407

August 1991

ARGENTINA'S EXPERIENCE WITH PARALLEL EXCHANGE MARKETS: 1981-1990

Steven B. Kamin

NOTE: International Finance Discussion Papers are preliminary materials circulated to stimulate discussion and critical comment. References in publications to International Finance Discussion Papers (other than an acknowledgment that the writer has had access to unpublished material) should be cleared with the author or authors.

ABSTRACT

This paper surveys the development and operation of the parallel exchange market in Argentina during the 1980s, and evaluates its impact upon macroeconomic performance and policy. The historical evolution of Argentina's exchange market policies is reviewed in order to understand the government's motives for imposing exchange controls. The parallel exchange market engendered by these controls is then analyzed, and econometric methods are used to evaluate the behavior of the parallel exchange rate and its impact upon the balance of payments.

The main conclusion of the paper is that exchange controls were never effective enough in Argentina to allow the authorities to set the commercial exchange rate independently of the parallel market rate. Attempts to set the commercial exchange rate at too appreciated a level consistently prompted widespread evasion of exchange controls that undermined the government’s international reserve position. Econometric evidence supports the hypothesis that important components of the balance of payments were negatively correlated with the parallel market premium during the 1980s. The evidence also confirms that the parallel market premium was influential in the determination of the commercial exchange

rate.

Argentina’s Experience with Parallel Exchange Markets: 1981-1990 Steven B. Kamin? I. Introduction Argentina’s history of exchange controls and parallel exchange

markets dates from the beginning of the 1930s and has persisted, with occasional episodes of liberalization, nearly to the present. The most recent unbroken period of exchange controls in Argentina lasted from early 1982 to the end of 1989, and coincided with one of the most turbulent periods of macroeconomic crisis in Argentina’s history. Faced with balance-of-payments problems, accelerating inflation, and severe private sector indebtedness, the authorities imposed exchange controls as an alternative to a mix of real devaluation and fiscal adjustment that would have secured long-run stability, but at the cost of short-term contraction. In the absence of fundamental reforms, the macroeconomic crisis deepened, leading to the hyperinflations of mid-1989 and the first quarter of 1990. This hyperinflation, in turn, generated balance-ofpayments pressures so strong as to force the abandonment of exchange

controls and the floating of the currency in December 1989.

V/ This paper contributes to a World Bank project on "Macroeconomic Implications of Multiple Exchange Markets in Developing Countries". The author is a staff economist in the International Finance Division. This paper represents the views of the author and should not be interpreted as reflecting those of the Board of Governors of the Federal Reserve System or other members of its staff. I am indebted to Richard Agenor, Dale Henderson, David Howard, Deborah Lindner, Miguel Kiguel, Helen Popper, Andrew Rose, Enrique Szewach, and the staff at FIEL for their helpful comments and suggestions. Special thanks go to Neil Ericsson for his econometrics advice, Will Melick for programing various econometric procedures, Saul Lizondo for his careful comments on an earlier draft, and Stephen Thompson for providing a constant flow of news and data from Buenos Aires. Daniel Kelley provided excellent research assistance.

This paper reviews Argentina's experience with exchange controls and parallel exchange markets during the 1980s, and evaluates the impact of these institutions upon macroeconomic performance and policy. Section II provides an historical overview of this period, highlighting both the factors influencing the government’s exchange market policies and the impact of those policies on the macroeconomic situation. Section III describes the structure of the parallel market during the 1980s, focusing upon the sources of supply and demand for foreign exchange, as well as the efficiency of the parallel market. In Section IV, the behavior of the parallel exchange rate is analyzed, and empirical research aimed at identifying the key determinants of that rate is presented. Section V gauges the influence of the parallel market premium over the official balance of payments. Finally, Section IV presents an evaluation of Argentina’s exchange-control policy as an instrument to achieve macroeconomic stability.

Fundamentally, the rationale for exchange controls was to permit the authorities to set the commercial exchange rate at a level that would have been unsustainable without controls in order to stabilize the economy. The main conclusion of this paper is that, in the absence of fiscal discipline, exchange controls were never effective enough in Argentina to allow the authorities to set the official exchange rate independently of the "market-clearing" rate. An active parallel market in foreign exchange developed immediately in response to exchange controls. Depreciations of the parallel market rate, by causing diversions of export receipts to the parallel market and reductions in short-term capital inflows, consistently undermined the official balance

of payments and led to shortfalls in the government's international

reserves. These shortfalls, in turn, inevitably led to devaluations of the official exchange rate. Efforts to contain the parallel market premium through contractionary monetary policy proved to be too costly to be effective. Hence, exchange controls served only to delay necessary adjustments of the official exchange rate, not to postpone them permanently.

Exchange controls were not merely ineffectual, but also imposed significant costs upon Argentina’s economic performance. First, because they were able to offset the effects of a misaligned exchange rate, if only temporarily, exchange controls encouraged the use of short-run stabilization strategies built around fixed exchange rates to the neglect of fundamental adjustment policies. Second, because exchange controls created strong incentives for evasion, they furthered the extension of the underground economy and the breakdown of compliance with economic

regulations, including tax laws.

‘BuTJUdsaAy S| US SOTquey op [Oaquoy [Tq ‘Apnas 6861 IdId @ Wory uMeIp sem potaed sty} Zutanp sjorzquoo eSueyoxe ynoqe uoTJeWAOJUT eYy2 Jo YonW /z

[eutezxXe UO sToIQUOD ‘peTjJtTun JoyTeW eSueyoxe syQ pue peqQeUTUTTe @I9M STOIQUOD asZuRYyoOXe UeYyM ‘GCG PU EEG UPEMI2EG “CCST Pur ZEST BuTAnp @QeP1 [eTOTJJO ey AeAO Queored QF ATYZnor Jo umquerd e Ae BuTperzq eqer eZueyoxe YoyreM YoRTG ey? YITM ‘pedoTerasep uoos AdueAAND AOZ QoxAPU YoeTG e ‘aouenbesuod UI “TEST 19qG0900 UT sUOTQORPSURIQ QUeTANO UO sUOTIOTAASeA pue sjTorquoo [eqtdeo pesodut 31 ‘smoTzq3no Teqtdeo ZutnutTquoo yAIM peor; anq ‘eqei eSueyoxe vy Butn{eaep Aq pepuodsexz ATTeTATUT JueuuAeAC? oY] -A[TQUPOTJTUSTS Apore 03 uOoTATSOd squeuked-jo-sourTeq pue oTWOUCde010FU eTqeloaesy S,euTQUeSIY pesned ATOTJep TePOSTJ ey Fo uotTsuedxe ue pure epejz Jo sw10Q oY. UT UOTIPAOTASeIEp B ‘SOEST AYA FO ButTuuTZeq ey 3V ‘s[TOIQUOD AUeUUAeAOZ TeUOCTITppe AnoYyATM AouerANSO ey Jo enTea ey 3A10ddns 02 pedtey oste uotsseideg ey. 03 AoTAd saeeh ayQ uT euTqQues1y Aq pefofue epeirz FO su19 STqeAOABFZ syL “SN[NWTAS [eOSTJ eATSSEeOKe YIM poe zetToosse BTAQITInbestp ofmouocdeort0eU YesSFJO 09 STOTQUOD eduUeYoOKXSe AOF poeoeu eTIITT sem o1ey ‘Awouooe ey} UT eTOr QuemUIeAOC’ pepuedxe ue A0zZ AdUENQTISUOD Teotatqtod ey Fo sseuyeem oyQ Jo esneoeg ‘“[eAeT eATITIedwWoouNn ue Qe QuewUTeAOS oyQ Aq oQeA1 eZuURPYOXE aYQ Jo BuTAQes sReAEqTTEp 249 pepnytoeid ‘soqattod ut pue Awouose sy ut yqoq ‘Azqsnput 31odxe ey Fo aoueqrodmT aeATReTAA eyL ‘eWwTQ AyeYyQ 3e AWouOde TeOTATTOd s,euTIUEZAy Fo SOTISTIEQOPABYO OTSeq oy? YATM QUeQSTsUOD sem oUTZeA STYL ‘eR2BAI eBUPYOXeE ey3 Jo 3e0TZF AQATp e& pue pAepuUeAS PTO? e usemjeq BuTQeUAEQTe ‘peTFtTun pure eely SEM PUTQUEZIY UT JoYAeM sZuUPYOKXE SY ‘OEGT PUP OOSGT UeEMQeEg

29L6T-Te6T TSJOyACH OsuUeYOXY Tend YIM SoouetTioedxgq SNoTASIg [ “II

MopTAASAQ [BOTAOISTH Il

transactions were alternately tightened and loosened, depending upon the pressure on the balance of payments posed by external shocks and internal disequilibria.

Following the end of World War II, the strongest pressures derived from the escalation of aggregate demand that accompanied Argentina’s industrialization under Peron. The black market premium reached a high of 400 percent in 1951 compared with an average of only 10 percent during the 1941-45 period, notwithstanding very favorable terms of trade in that year. This escalation was symptomatic of fundamental inconsistencies among the. government's macroeconomic policies which were reflected in a decline in growth and increase in inflation relative to the immediate post-war years. These problems provoked a shift in policy toward economic liberalization which culminated in the elimination of exchange controls by 1959, accompanied by the lowest fiscal deficits since 1945.

By the early 1960s, the combination of resumed deficit spending plus an external. debt-financed investment boom led to the threat of a new balance-of-payments crisis. This inspired a return to exchange controls and the parallel exchange markets they engendered during the 1964-67 period. However, the exchange market was unified again during 1967- 1971.

Following the highly successful stabilization program launched in 1967, the fiscal and external balances began to deteriorate in 1970, leading to substantial capital flight and a balance-of-payments crisis in 1971. In response, and as in previous periods of crisis, the government re-imposed exchange controls. These were to remain in effect through

November 1976, by which time the fiscal deficit exploded to over 15

‘a8en [eel ey UT suUOT}ONper [TeTQUeASqns (ATQUeQAOdUT ez0m)

pue uoT{oONpser ATOTJep YyBnorzyQ uoT_AeTJUT VsoNper 09 SQAojJje BuTpueqsy¢tTaqou

‘8/61 Ut queorzed 9‘c/T 3e Apeaqs poufeuwer ‘//6][ UT Queored

T°'9LT 92 9461 UT Queored yyy Jo yeed s2]T wWorZ peuT[Oep peYy YOTYM ‘eje2

UOTIBTJUT ay eSNeoeq UABDUOD ABTNOTAAed Jo sem sty, ‘eseq AreQeUOU

eya FO ToAQUOD s,yUe_ TeAQUeD 9YyQ peueqQeetTyA Jey? sMoTjJuT Teatdeo

03 Zatpesay, ‘ATTeTqueqsqns:esort seqexr ysereqUT ‘//67 ABW UT soqez qserzeqUT

atsedep Jo UuoTIeZTTeASqTT eyQ 02 eNq ‘TeatTdeos Jo smoTJuT [tTeqans

02 se os ATMOTS SLOW PeZTTeAeqTT e7eM STOAQUOD TeqtTdey ‘peztT[ereqTT

ATeZAeT YsuUOTQOPSURAQ QUNODDe QUeTAND AOF seTes edueyoxs Jo TorzQUOD ayQ

pue petjtun o10M sqzoyzem oSueyoxe 942 ‘9/61 AeqMeAON UL ‘AMouode ayQ Fo

S1OJD8S SNOTAPA AZAO STOAQUOD eTIUPUSTP 02 peAoW ZOH ep ZeUuTAAeW AeAsTUTW Awouody ‘9/61 YOAPH UF BPOTFJO Jo uoTyAdumsse sty BurTMoTI[og

TSOT-9L6T (20H Op ZOUTIAEW ASpUN UOTIEZTTEISqTI © IT

“UOTIEZT[TPASGTT AOF soeqseq AeZu0rzAs YATM suoTAeAAQsTUTUpe

mou Aq yseTAoeq e& peyoaord ‘setotT{Tod AsTuoTqUeAreqQUT AeyQO Aq

petueduosse etem ATTeLeues YOTYM ‘SToAQUOD esoYy YITM VvoUeTAedxe pepueseqzxe

ue zeqye ATUO 3nq ‘peptsqns sTsTAo ey Aeqze AToReTpeUMIT peAoWeT Jou 270M

S[OI}UOD vseyq ‘puodsesg ‘sesTIo squewhed-jo-soueTeq Aq peqeaTjow skenTe

@1emM STOAQUOD eZUFYOKe Jo suoTAeoT[dde TeTaTUT ‘asATg ‘“sTorQqUOD eZueyoxe jo AzojsTy s,euTqQuesazy 09 eTqeot{dde azeedde suotjezt[erieues omy

‘AoT[Tod oTwouoose s,euTquesizy

ut eseyd 4xeU eYyQ peqAeTATUT pue QUOMAZACZ UPT[TATO S,uoteg Toqes]

peoetder dnoo AzeQAT [Iw e usUM ‘9/61 YOAeW uT Quesoried OF, AesA0 peyorer

untweid JeyzeMW YoeTq oyR {mor Be UT APeA PUOdES ay AOJ peuTloep ddqoy [eer

pue ‘(sTseq [Tenuue) queoied gov peddoq ejexr uoTIeTJUT aya ‘qay Jo queorzed

In December 1978, the capital account was liberalized completely while the exchange rate was depreciated, according to a pre-announced scheduled, at a rate that was to gradually decline over time. It was the hope of the architects of the "tablita" ("little table", in reference to the table of pre-anounced exchange rates) program that eventually the rate of price growth would converge to the declining rate of exchange rate depreciation, and that by pre-announcing the exchange rate, the reduction in inflationary expectations would be accelerated. In fact, the rate of inflation declined much more slowly than anticipated, registering 159.9 percent in 1979 and 100.8 percent in 1980, and exceeding the rate of exchange rate depreciation under the tablita.

Consequently, the real exchange rate, shown in Chart 1, appreciated substantially.> Coupled with continued large fiscal deficits, this led to a turnaround in the current account balance from a $2.1 billion surplus in 1978 to a $4.8 billion deficit in 1980 and a $4.7 billion deficit in 1981. These deficits were financed by a surge in capital inflows motivated by a combination of high domestic interest rates and the low rate of expected exchange rate depreciation guaranteed

by the tablita.

3/ Kiguel and Liviatan (1988) attribute this appreciation to a combinaticn of inflationary inertia and insufficient fiscal adjustment. Other analysts have argued that the real exchange rate appreciation was the result of heavy capital inflows, discussed below, and a strong balance of payments; according to this explanation, the high inflation rate was not the ultimate cause of the real appreciation, but rather another reflection of the capital inflows. (Various versions of this argument are discussed in Calvo 1986, Sjaastad 1989, and Rodriguez and Sjaastad 1979.) Connolly, Rodriguez, and Tyler (1991) embrace both explanations, arguing that an initial equilibrium appreciation reflecting capital inflows was followed by further, unsustainable appreciation induced by excessive domestic credit creation.

OG6l 886l 9861 vs6l c86l O86t 8Z6l

00¢

oOv

009

008

(d/.d)d3 JoHe2e@g- — — — (d/.d)3 JeluewWwog

OOOl

(AjyJUOW)) seoipuy eyey eHueyoxy jelouswwoy pur jayjeueY jeoy

} HeYD

-e/-

By 1980, however, the growing overvaluation of the real exchange rate led to expectations of a corrective devaluation, motivating the private sector to reverse its prior stance by reducing external borrowing and sending capital abroad; gross international reserves less gold declined by $6.2 billion in 1980 and 1981 after rising approximately the same amount during 1978-79. Continued heavy borrowing during 1980-81 reflected government borrowing to maintain the overvalued exchange rate, as well as distress borrowing by firms in the tradeables sector that were adversely affected by the real appreciation of the exchange rate.

Between end-1978 and end-1981, the total external debt rose from $12.5 billion to $35.7 billion.

Due to the deterioration in the balance of payments, in February 1981 a 10 percent devaluation was announced, accompanied by a statement that the tablita would thereafter be adhered to for an additional seven months. In April 1981, however, the new Economy Minister, Lorenzo Sigaut, announced a 23 percent devaluation and the replacement of the tablita by a series of smaller, more frequent exchange rate adjustments. Capital flight abated only temporarily, leading to another maxidevaluation of 23 percent in June 1981, shortly before the return to exchange controls.

II.3 Initial Dual Exchange Market: 6/1981-12/1981 By mid-1981, the authorities faced at least six key economic

problems remaining as the legacy of the preceeding period:

1.) A highly overvalued exchange rate, still 37 percent more appreciated than its December 1977 value.

2.) Severe capital outflows reflecting amortizations on recently acquired external debt and private capital flight motivated by expectations of continuing devaluation.

Tl® 02 pet{dde sem e4ei [eToOTeMMIOD pexTZ & ‘[86T eUNGC eqeT UL ‘AWOoUODAe ey. JO Aser ayQ 09 AanfuT AnoyyATm qunoose Teqtdeo eyq uo seanssead eyQ qiosqe 02 93e1 eBueYyoxS BuTAeoTJ e YIM sUOTQOeSURTQ epetQ AOFZ paesn ojer aZueyoxe ey. Jo uoTAelToeidep [eet AJemoTS e BuTUTqUOD JexAeM eZUeYOXEe [eNp e 103 poeqdo weeq oTwouo;e s,qne3Isg ‘squTeAQsUOD eseyi YIM posoey ‘ZoQ0eS aqeatad pejqeput ATYZTY eyQ Jo AoueATOS 9yQ ezTpaedoal AEeYyAINI p[Nnom setot{tod yons yeyW peAeTTeq sem AT Ang ‘AUSTTJF Teatdeo eyy pesrzener oaey 3Yystw AottTod Azeqouow AYSTQ pue squeuqsn[pe TeosTjJ ereaes jo werZ0id Azequewetduos y “AQTTIqeqsuT OTWoUOD.e0IDeM pur UOTIeTOeAdep ajei e8ueyoxe eangnj Jo suotjedtotque Zuorzqs Aq usaTIp sem YoTym ‘quZITJ Teatdeo eyQ younejs of paeodTzsgns eaey Jou ptnom ATqeqoad eje1 asueyoxe SutTourleq-junoooe 4uerzinod Ao ,uWnMTAqT[Inbe, ue of UoTIeNTeAEp eB ‘teaoet0Q =‘ AQT[TqQeysuT TeToueuTyZ pue Aoqdnayueq peerdseptm jnoqge 3yZnoiq eAeY PTNOM osTe AT ‘Aqep TeUreQKXe Jo enTea AoUeTINO oOTIseWOp aya Butseerzout Aq pue ‘uoTAeTjuT peseerouT AT[eTIUeAsqns seAeY pP[NomM suOTe uoTIeNTeAep YZnorzyA wetqord squewtked-jJo-soueTeq ey. BuT.oeI1109 ‘uzaou02 Jo seere T9eYyAO 9YyQ TOF vere STYR UT UeXe] seansPou Jo saouenbesuod esreApe 942 ATWTT 02 BuTqdweqqe eTTYym ‘sTstT1io squewted -Jo-so0urTeq 9yQ ‘weTqord Zutssead qsouw ey. sseippe 02 sem seTITAoYINe eu Aq uasoyo ABeqeaqs oyQ ‘ATATOTTdWI ‘eaoqe peysTT swetqoad ay3 Jo T]1® peqoezI09 say p[Noos eZeyoed Aot{od eTBZuts ou ‘ATSnoTaqg ‘ZoQ00S [TeTOUPUTJ-UOU eQeATAd ey pur weqsAs TeTOUPUTJF 942 YIOq UT ‘TeurEyXe pur ITQSeWlOop YyIOq ‘sseUpeqqepuT Jo eerZep snorzesuep Vy ('9

"Z861 UT queoxzed 7°G pue [g6T UT Queorzed 7°9 JO {qD) [eel UT seuT[oep 09 BuTpeeT AQTATIOe OTWOUODS UT UOTIONper VW ('¢

‘1861 ounce uT Queored 4°6 02 OB6T ASNBny UT (sTseq ATYQUON) queorzed +'¢ Jo MOT & WOAF UOoTAeTJUT soTAd AJSswuMsUOD Jo UoTIeASsTVONR UY ("+

“T86T UT ddd) Jo queorzed /*9T 02 6/6T UT ddd Fo quested ¢°QT JO MOT QUeDeT B WOAF ATOTJeEp TeosTF ey Jo uotsuedxs uy (‘¢

——— ((Ep/E)-1)*100

1978

J

1980

-9a-

Chart 2

The Parallel Market Premium

1982

(Monthly)

1984

1986

1988

1990

175

150

125

100

lor

25

UL BUTZTTTqeqIs erojzeq ATqerepTsuos peqetoeradde ATQuenbesqns y4nq ‘eje1 TeToueulTy snotaeid eyQ Jo TeasT syQ 09 ATAPeU pejeToeidep oje1 asBueyoxe petsrun oy. ‘peqeoTZ Buteq uodn *e4e1 eBurYyoKe oy pe JeoTy pue ‘werz3oad sequerensd ojer1 oBueyoxe oyq pepuedsns ‘stTozquoo eZueyoxe [[e peAower oy ‘yquou ey. Jo pus 94. AY ‘UUeUIEeTY OATeqoYy peputu-[ereqt{, Aq Awouosy Fo AaASTUT_ se queusoeTdezr s,qne8ts uT peja[Nset sSTSTIO oTWouoZe ZuTAFTsuequt ayy Aq peozoZ sSuUOTRIeIAASTUTUpe UT ssUFYO ey ‘Tg6T AequiIeDeq UT ‘STSTI9 2ey 09 asuodsez eTqeqatTaouT ue jou azem Aaya ‘sog6l ATAee eYyA Jo STsTz9

OTWMOUOD.=<0TOeU 9YyQ Aq PezPATIOW 219M STOATQUOD sBuURYOXS eTTYM ZEST/TL-TS6T/CL a ONTEN SSUYORT Tend puooss syy pus WoTIPoTFTU IT ‘Z QABYUD UT poqeoTpuT se ‘[g8EI AeqUIeAON UT JUsdzed Z/ Fo YSTY e peyorer untTweid ey. feqeA [eToOremMIOD aYyQ sAOqe qUsodAed gt 03 9Qe1 eBUeYoOxeE [eToueUTF BuTQeoTF syQ FO uoTAeToordep ojelpemT eyQ UT peAoeTjoer sem eainsseid sty} ‘seaztesel [euOTIeUTeAUT JO eTes 9yW YsnorzyA AYsTTF [eatdeo BuTAepouooce peddojs pey mou quemureACS oy. esnedeg ‘sjzesse USTeT0J oQUT SOT[OJatod AFTYS 03 SeATQUeOUT ZUO0IAS peptaord AQTTTIeTOA DTWOUOD<OZDeU pue ‘uotjetoeidep ejei esueyoxe ‘uot ,e[jUuT ButstTzA Jo suoTqeqoodxy "WUSTTJ TeqtTdeo z0F seaTjom euQ AT[eTIUeASqns soNper jou pPTp

weiZoid sequeien3 ey ‘Iqep TeureqQKe FO HASTA 9yQ ButTonper eT Tym (1861 ‘Vdod) ‘vot IeToerzdep quanbesqns ysutTese jqep TeureqyxXe YSTY YITM SsWATF yOoeQOAId 09 Se [TEM sF ‘yueg [TerqUeD eyQ UO seiansseid squouted-jJo-soueTeq eraeT[et 09 sem esodaind SII ‘qep [eUureRXe ATeUQ eTNpeYyoser 02 BuTT[TIA seTaTque eqeatTad Aoz weaZoid sequeiend ojei eSueyoxe ue peeTATUT seTATAoyAne ey ‘uOTITppe ur ‘oqei eBurPyoXe [eToueUTJF BuTQeOTF oy Je soeTd exe 02 STEM SUOTIOPSUPIQ ZuTUTeWeT [ITV ‘2qep [eUreqXe peyoRrqUOD ATSnoTAerzd Jo squeuhedaer

pue ‘sqiodxe ,pejoword, Fo queorzed 06 ‘sqaodxe [euoTATperq pue sqrodut

-OT-

-ll-

real terms. This strengthening of the peso was accompanied by, and to some degree induced by, a tightening of domestic credit growth and a slackening of inflation.

Alemann's experiment with liberalization was cut short in April 1982 with the eruption of war between Argentina and the United Kingdom in the South Atlantic. The government’s suspension of payments to British banks motivated a suspension of financing to Argentina by the international financial community, as well as the imposition of trade restrictions on Argentina by the European Community. This pressure on Argentina's balance of payments was compounded by increased capital flight motivated by the surge in uncertainty concerning Argentina's general. situation. At that time, the Central Bank would have liked to have contained the ensuing depreciation of the free market peso through a tightening of monetary policy. However, the crisis had led to the threat of a run on the banking system, forcing the Central Bank instead to lower reserve requirements and raise its supply of loans to domestic banks.

To limit the depreciation of the peso, exchange controls were re-imposed in April 1982; the exchange rate, while nominally still floating, was de facto controlled by the government. All foreign exchange sales not authorized by the Central Bank were suspended, while quantitative limits on different import categories were set, depending upon how essential they were considered to be. The repatriation of profits and investment income was also suspended, while the government began to pay some of its external debts with BONEX, 10-year dollar denominated government bonds. With the coming of exchange controls, foreign exchange trading on the parallel market immediately resumed. The

parallel market premium averaged 54 percent in the second quarter of

pue sueoT yj0oq uodn s8uT[Te0 eQet YseTeqUT TeUTWOU pesoduUT seTITAOYy ne eud Z861T AlN UT ‘ere Zoy ep zeuTIAeW eYyQ BuTanp ATTeTIuUeIsqns umoiZ osTe pey Yorum ‘ueping Aqep oTASeWop ey soONperT Oo] ‘eoTAIES Aqep TeUreAXe $,10}0eS aqeATId ey JO 4Ysod ey peZTptsqns se1Zep edie, e 02 ‘7861 Ane UT peqeTaATUt-e1 ‘omeyos sequerzen3 ejei eZueyoxe eu (066T ‘olqed ep pue yosnquiog osTe e985 ‘786 ‘VNOd) ‘Usepang Aqep s,10jDeS ajeatad [eToueuTF-uou ey Fo uoTIONper ey Anq ‘squeuked jo soueTeq ayy Qou SBM Z86I-PTW UT WesQ oTWOUODe aYyQ Jo AQTAoTAd AsaYysTY eUL ‘squeuked Jo souertTeq eyq Jo eQeQs 9YyQ uodn BuTpuedep ‘peueRYy3TI pue peuesoo,T ATeQeUrEQTe eq PTNOM suUOTIOTAASeA eseyQ ‘ATQUeNnbesqns {peqeUTWT [Te Jou 4nq peZzTTeAeqTT e7eM AGep TeUrEeRXe ey} UO JsSeTEeRUT pue ‘squeuked yserequT pue ‘uotqeztqiowe ‘sqiodmy tojz eSueyoxe ustTerzo0j Jo eT¥S dy} UO sUOTIOTAQSEA EYL “[TeM se pesZed Buteq e10jeq syoem Moz P AOF pe,eoTF WeyQ suoTAoPsueAR ASeyRO AOF ojReA [eTOURPUTF e pue suUOoTIOBSUPIQ [eToreuuos AOF Quoewuzeac3 oyQ Aq pesded ejer [eToOAeMMIOD e OQUT atids ujtege sem oqer eSueyoxe ustTez0y TeSeT eyd ‘7g6T Ane UI ‘setdroutad oTwou0[e [ereqT[ 02 peaqtuMoo sseT ‘ej0qseg outuseg asor ‘Xkwouooq Jo ASeASTUTP MoU eB YITA UUeWETY AsTWAOJeT eYyQ peoeTder ABM BY PEeMOTTOF 2eYQ squewureAcd uT eZueyo ey ‘ATTeuOTITppy ‘“SuTanjew MOU 919M [RST Tequeceq UT JUewUreACZ |yQ Aq OUT peteque sdems eSueyoxe UT UOTTTIW 00S$ 3noqe pue ‘peTputmp pey AQTunuMOD TeToUPUTZ TeUOTIeUTEAUT aya Aq 2a0ddns ‘si0jTper19 Teure}xe 09 sAzverTe UT ATTeTQUEeISqns Se@M JUeWTEeAOS |YyQ ‘TeSu0rAs orem AYZTTJ Teatdeo AJoFJ seansserig “APM aya Jo Jesqno oy 3e peTTeaeid pey uey. QueueotTpeid zedeep e uy euTquesiy 3F°T 786T SUNL UT OTQUBTIY YANOS ey UT JOTTJUOS ey FO pus SUL ‘syquow esey 3uTAnp eqer eSueyoxe (TeToOTFFO)

[eToxsemuod ay Jo uoTQeToerdep [eer TeTIUeIsqns ZuTpueqsyATMjou ‘7g6T

-é@L-

-13-

deposits at levels below the rate of inflation. This had the effect of substantially reducing the real value of the private sector’s domestic debt and thereby reducing its domestic debt burden. However, it also led to an acceleration of the flight out of local currency deposits. In consequence, inflation jumped from 7.9 percent in June 1982 to 16.3 percent in July; the parallel market premium soared from 78 percent in June to a peak of 126 percent in August.

The Central Bank’s annual report for 1982 argues that the dual market regime was necessitated by the ceiling on domestic interest rates; otherwise, the subsequent flight into dollars would have depreciated the commercial exchange rate with both unwanted and unnecessary effects upon inflation and trade. However, the report went on to note that during this period, substantial international reserve losses occurred as a result of export delays motivated by expectations of future exchange rate depreciation, leakages of export receipts to the financial market, and intervention by the Central Bank in support of the financial market exchange rate. (BCRA, 1982)

In response, the authorities began to unify the two legal markets in September 1982, both by devaluing the commercial rate while keeping the financial rate fixed and by increasing the proportion of transactions conducted in the financial market, completing the process in November 1982. Because private capital flows continued to be controlled, however, the gap between the parallel market rate and the new, unified rate continued to exceed 40 percent through the end of 1982, although it

had declined considerably from its mid-year level.

‘ATOTJOp [eosTy 9yQ UT uoTAonper eB YBnorzyA peazanoes eq 02 sem AQTTIqeas esotad wrzeq-Bu0yT (6861 ‘Ten3Ty ees) ‘“suoTjeqoedxe AXeUOTASTJUT pue uoTAeT FUT UT UOTIONper AFTMS & eAeTYOR OF Aepszo UT UeZOTZ SPM 93PT eBZUPYOXE TeTOTJJO ey Fo oenyTea TeuTWoU sy ‘sezeeI;Z aotad pue e8em YATA UoTIeUTQqUOD UL ‘“UOTIeTJUT Jo uoTAeAeTeooR ZuTNUT QUOD oy 04 esuodser UT Gg6T euNr UT pequeueTdWT sem YoTYyM ‘ueTg TPAQsny ay. UT ATOTTdxe epeu sem ,zZoyoue TeuTWOU, ATBUCTIETJUISTIp e se 9je1 aZueyoxe ey FO osn 9UL “UoTIeTJUT eonper 02 pTq [NJsseoonsun ue uT oer UOTIETJUT oYyQ UPYQ eoRd JeMOTS eB Ae peqetToerdep ATeIeASEqTTEp sem 931 asueyoxe oy. ‘7861 FO Ysow BuTAng ‘qQunouezed seuweoeq AZeqQeAAs UOTIeTFJUT -TQue s,jUeWUeA0S 9YyQ UT o3e1 eZUPYOXe ey FO eTOI sYyQ ‘suOoTIeTEepTsUOD

aseuy Jo sourqiodut eaTAeTeAT eYyQ UT UOTIONper ey YITM ‘untTweid JYexrzPU Te TTeazed ayy pIp se ‘Zg6l A99Fe peuT[oOep AUSTTZ Te3tdeo jo seanseow snoTaea ‘$¢g6T Aq pereTdwoos ATe8reT ueeq eary Aew ssaoo0id sty. ‘10Q0es aqeatad eyq Aq sjesse usTeAOF OQUT AFTYS OTTOFRA0d eaTssew e pedeANooUs SO86T 2U2 FO STSTAD OTWOUOCDSOTDEU ey. VY QUEeQXe 9Y2 02 ‘ATEATQETNOAdS azow pue ‘ATTeUTY ‘“3qGep uUsTer0F aqeatad eyQ Fo uoTRIeZT[eUOTIeU TeNQUeAe ayy pue ‘wearZ0id sequeren3 oje1 eB8ueyoxe ey. ‘7861-PTW JO seToTTod ejKV2 QsereqUT ey YZnorzyQ pe_QeUTWTTe ueeq pey ATesreT ‘[eurEQXe pur OTISeUOp y20q ‘ssoupeqqeput 30j0es eqeatad Jo wetTqord ey, ‘puooes ‘“Queutiedxe UOTIBZTTPASEGTT eyQ Fo ButTuut3eq oyy ‘9/61 UT 2eY2 OF eTqeaeduod [eaeT B 02 pen[eAep useq prey o3e1 eduUPYOKe [eer ey ‘3SAITY ‘peqeTdwoo ueeq pey potsed zoy ep zeuTIAeW eY4I YITM peqeytoosse eTaqt{Inbestp eyy

02 squouqsnf{pe queqiodut jo zequmu e ‘eget Jo ButTuuTZeq ey Ag

S8Ol/L-CSol/1L -oAIeW SsUPYOXY ToTTered SyL C*Il

-T-

~14a- Chart 3

Consumer Price Inflation per Month (Monthly)

Percent 200

190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30

20

1978 1980 1982 1984 1986 1988 1990

*potied styuq 3utanp eje1 e8ueyoxe s,euTquesizy Jo sseueaTqTjeduoo aya oe AsieAO ATOATT YSou soxepuyt TereqeTTaqtnwu oy? ‘/g6T pue 986T UT eperq JO suz9q s,euUTQUSZAY UT UOTIeAOTASQep eAeAGeS ay JO aesnedeg ‘“CR8ET Jeqye seTouerzino ArQuUNOd-TeTAasnpuTt Jofew 1eyQO 9eYyQ ASUTeZe TeT[Op eyQ jo Tleyz aya Aq A0F poqunoooe st soueressIp ey, ‘“Apnas sTyQ ut pesn st yoTym ‘s0q9eaS peatug eya YQIM 94e1 eSuFYyoxe TereQeTTq Teer eyQ Aq pe qeoTpuT uoTqetoeidde [ver oy. YAIM SqseazqUOD sTUL ‘“UeTg TerAsny ey. BuTMOTTOF uoTQeToeidep penutjuoo moys AT[Tereuesd ‘szeuqied ZuTperzq sqt jo oeS8erzeae peaystem-epeiq eB jsuTeSe eqer eSueyoxe [ver Ss,PuUTJUeSIY Jo QueweAoW ayy MOYs YOTYUM ‘aqezr eSueyoxe TeAEQeTTI[NU Teer s,PuTQUeZAY Jo seanseey /4

“O861T UT qQUedTEd ZIT 02 HE86T UT qUeodzed gE worZ BuTUTT OEP AOAFS (BGT UT QUeored gz peseroeae YOTYM ‘umtuerd JexAeW YORTQ ey UT perozrAtu sen y otPt aZueyoxe TeToOTJFO Teor oyQ Jo uotqepToordde [enpesz aya YQTA SuoTte ‘[orQUOD oTWOUOD=<0ADeU UTeQUTeEW 02 esANTTey ey, ‘*‘seanssead ATEUCTILTJUT UTeQUOD YoU PTpP 23e1 VBUeYOKe TeTOTJJO ayQ Jo uoTIeTosrdep 249 UT SUMOPMOTS pue sezeery eotad quenbesqnsg ‘“¢ AreYO UT QuepTAs se ‘UOTIeTJUT UT seseeTOUT AEYAANZ 02 peT YoOTYUM ‘986T TTadv ut BuTqazeqs aqel aZueyoxe sy. Fo suoTAenTeasep ITpotied pue sjTorquo. eotTad Jo Butsee ue peoroy seansseid ssey_ ‘seansseid AXeUCTIeTJUT eqeUuTWTTe 03 ATQUeTOTFRNS poonper Jou sem AT ‘986T UT Jueorzed » pue Cg6T UT Queorzed g of HgZET UT da) JO queorzed oT ATYBnor WoAZF peuT[oOep ATOTJep [eosTF aya eTTUM “AT TeTQueAsqns seqer YseTeQuT pester aeya 9gET AIM pue ‘CcgEl AequeroNn ‘sg6t ATne ut Aot[od Azeqouow Jo sZutTueqystq Aq pezeqstoq sem joaezjo sTy2 {smo[TjutT Teqideo Buo1qs peonput pue sjesse uUSTetoF OF BATIeTOT Sqesse OTJsSewlop uo uANQeA Jo 9Qe1T 9YyQ peseeroUT ATdzeYS ‘eaQerT eBuPYyoXe TeTOTJzZo ay. FO BuTXTF ou. yATM poutquoo ‘werzZoid ey? YATM peqeToosse uoTIETJUT UT uoTIONper dzeys seul °7 3AeYD Aq peouUeptas se ‘umtwerd QeyTeU YoeTq eyQ ButTonperz ut pue ‘werzZorzd eyq Jo syquow xTs 4SATF aya ZoF YQuow ted queorzed ¢ qnoqe ATuo peSerzeae YyoTYyM ‘uoTAeTJUT Jo ojV1

ay ZuTonper ut yIog [NJsseoons sem AT[eTITUT ueTd TerIsny eYyL

-S—T-

-16-

These problems, combined with a severe decline in Argentina’s terms of trade during 1986 and 1987, led to a ballooning of the current account deficit in 1987 to over $4 billion, its highest level since 1981. In October 1987, the commercial exchange rate was devalued substantially, while a new floating financial exchange rate was authorized for travel, tourism, direct foreign investment, and non-guaranteed private borrowing. The new regime largely replicated that extant during the second half of 1981 and the third quarter of 1982, but did little to change the underlying structure of the exchange markets; the new financial rate quickly depreciated to the level of the parallel market rate, reaching a premium of 42 percent over the commercial rate in January 1988 before gradually declining subsequently.

II.6 The Plan Primavera and Temporary Unification: 8/1988-6/1989

Following continued acceleration of inflation, the Plan Primavera ("Spring Plan") initiated in August 1988 was intended to keep inflation low through the presidential elections scheduled for May 1989 without fiscal austerity. The linch-pin of the program was an agreement with private sector leaders to keep the growth of public prices, private prices, and the official exchange rate within 4 percent per month. This was accompanied by the transfer to the free, financial market of all imports and 50 percent of industrial exports. The Central Bank was to auction foreign exchange 2 keep the floating financial rate within 25 percent of the official rate; because purchases of foreign exchange from exporters were to be at a more appreciated rate than the auction price, it was hoped that this multiple rate regime would generate substantial fiscal benefits. Finally, monetary policy was tightened to further

support the value of the austral in the free market.

A[snonuTt{uoo enTeaep 09 AT Buyporozy ‘seareser TeuoTIeUAEAUT s,yUeG [erqUSD aud pip se ‘peSuntd yueg Tearquep eyq 09 sqdteoer qaodxe Jo Aepuerzans ey] ‘(eqeX [TeTOTemUod oyA AeA0) uMTwerd JeyTeU TeTOURPUTF eYW pue ‘UOT eT FUT aotad Jo aqezr oyq ‘ajer eSueYyoOXe TeTOUPUTJF 94} JO UoTQeTOeAdep ey uT efans e 09 SuTpeeT ‘[Teazqsne ey uo UNA pa zteMme BuoT eyQ JFO peyonoz STUL ‘e981 TeTOURPUTJF OYA BuTAeOTF snyQ ‘VoyxAeW eerZ OY UT UOTIUSAT9QUT peseeo yueg TeAqUeD oy ‘YOOJS eATESEA TeUOTIeUAEUT eATQJUSe SIT FO SSOT aud UTM peoey ‘6861 AZeNAqeg Jo ButuuTZeq ey. ay ‘eBuer eTqeqdeooer ue uTUITM unTwerd JYeyreW TeTOUeUTJF ey deey 09 Aspro ut Aouenberzz BZuyseerout UQTIM euUsATEAUT 09 YUe_ [PAQUSD 9Y} Ped1OF sAOQORZ eseUL ‘and00 p[Nom [eAqsne 243 uo uni & ‘WeUeW soTIeD eqepTpurS ASTUOTeg Jo ATOQOTA TeI0JOeTe peqoedxe ay} 09 JoTAd ewTjewos WeYQ peAeTTeq sem ATTeA9UeS QT ‘ATTeula (6861 ‘886 ‘Yormezs 92S) ‘uoTSsTUe Azejeuow yBnorzyA peoueuty 9q 0F sAey PInom sjsoo r9UusTY eseyq ‘JeqeT AO AeUOOS ‘JqGep OTISeUIOp Ss ,qUeUUIeA03 auq BuToTAres Jo sj3sod AeYySTY 072 peT ojer eZueyoxe oyq Aaoddns 0} pepeesu SoqeA YSeTSAUT ASUSTY eUL ‘UOTIeNATS TeOSTF BuTIeAoTAeQep s ,quewureA03 aya Aq perteqstoq e1em suotTqeqoedxe esey] ‘“uoTeNTeAep Juenbesqns eB Jo suotTjeqo0edxe 09 ButTpeeT ‘reek ay Fo pue ey Aq sw109q Teer UT AUTetTqueqasqns poqepooridde aqear TeTOTFJO ey ‘uoTJeToerdep eje1 esueyoxe [eLOTFJO JO o2eA oy} 02 SUT[OSEP Jou pPTp 23eA UOTIeTJUT sy esneoeq ‘Zeaemoy ‘Aequeaon Aq Queorzed 9 02 gg6T AsNBny uT Queorzed gz wor; PeUT[OSePp UOTIeTJUT STTYM ‘JoeXTeW serF OYQ UT UOTQUeATeQUT YyUeG [eIQUeD Zoy pesu oj peonper uoTAeToerdep ejei oZueyoXe MoT peinsse puer seje1r Qsa1eqQUT oOTQsewop YZTY Aq peIeaATQoW sMoTJUT [eqITdeD “[NJsseoons sem

ATTeTAIUT ereAeUTAgG UeTg ey. ‘IT eATOFeq URTG TezIsny 947 eATT

-/T-

-18-

and shift increasing proportions of export transactions to the floating financial rate.

In mid-April, unable to peg the exchange rate any longer, the authorities floated the commercial exchange rate, temporarily unifying the exchange market. When the exchange rate was again pegged by the government at the end of May, following the presidential elections, the black market premium jumped to 75 percent in the absence of any government strategy to resolve the economic crisis. Monthly inflation peaked at 198 percent in July 1989, due to both the collapse in money demand precipitated by the developments in the exchange market and the corrective adjustments taken by the new administration.

IIl.7 The End of Exchange Controls in Argentina: 7/89-present

Following President Menem’s inauguration in July 1989, the new economic team announced a stabilization program broadly similar to the Plan Austral: a freeze on public prices, private prices, and the exchange rate, coupled with long-range plans to reduce the fiscal deficit. As in the aftermath of the Plan Primavera, inflation fell dramatically, precipitating a return of flight capital that reduced the black market premium essentially to zero.

By October, however, wage pressures had become more pronounced and the pace of fiscal reforms slowed while continued inflation had caused substantial real appreciation of the exchange rate. As expectations of a corrective devaluation developed, the parallel market premium re-emerged, leading to a marked decline in the Central Bank's international reserves; this decline was exacerbated by Central Bank intervention in the parallel market in an unsuccessful attempt to support

the austral. As in the Plan Primavera, the Central Bank also tightened

‘ouTZel QoyreU eSueyoxe [enp eB 09 BuTUAN}JeZ Jo uoTSsnNoSsTp

SNOTA9S OU ST ETSY ‘BuTATAM STYyQ JO sy ‘UOTIeTJUTSTp peutejsns

eB 09 A8y BYQ eq [ITA oWTZeT oQer eZueYyoKe Jo sdTOYO ay uPYy. ey AeA

Butue,y3tq Teosty/Arzeqouow QeYyQ ‘TeAemoy ‘SezTUZ00eT JI = ‘“YyoreW 22eT

UT ‘STOTQUOD eZuPYyoOXS AnoYITIM ‘e3er eZuPYoOKe eYyQ XTJ 09 QueuUTEACZ 23

pejdmoid [661 ATAee UT UOTQeTJUT Jo eZans meu e ‘AeAeMOH ‘YGG6T AEquedeq

Aq ueoszed ¢ moteq 03 uoTAeT[JUy peonperz sesn{dans Zutjerzedo Quewuzea03

fq e[qtssod epeu uoftsstwe ATeQeuow UT UoTIoONper e ‘ATQUenbesqns *Q66T

yore Aq queored ¢'¢6 09 peqereTeooR UOTIeTJUT ‘AeuOM AOF puewep 943

UT SeUT[Oep pexrtew of enp ATesze] ‘wey peuestom sary OF JOU smaTqord

oTWOUDDs0TDeU S,eUTQUeZIY PeATOS eAeY 09 JeYQTEeU sivedde sjzeyxIeU eZueYyoxe ay. Jo uoTIeoTFTUN pue oqer oZueyoxe eyQ Jo ZuTIeoTF sy]

‘6861 Tequeceq eqeT UT sqexreM ey AZIuN pue 9jer aZueYyoXe TeTOATeMMIOD

aya AeOTF 02 YuegG [eAQueD eyQ pedzoF ‘uoTATsod sareser TeuUOTIeUrEQUT

s,yueg [eAQUeD 9<Y4Q JO UOTIPAOTAeJep penuT juoo eYyQ pue ‘ejVA [TeTOURPUTF

aya Jo uotTjetToeradep penutquoo ey, ~“Aemiepun sem [eatds AreuotjzeTzUtT

Mou eB 4eYQ SUOTRIeQOedKe eoTOJUTEeA 03 peAtes ATUO ‘ueZOAF ToOrQUOD

S21 Jepun sen,Tea TeUTWoU 9yQ desy 03 QuemqtTuMOD snofaerd s, }uUemUIEA0?

ey worz BuTqaedep Aq ‘setotj{od eseyy ‘potaed zoy ep zeuT ABW ey. soUTS

qeyrzeuw eSueyoxe [enp [eZeT yAInoJZ ey BuTQeer7D snyQ ‘JeyAeW TeTTered ey

peztrTeseT osTe al ‘aeeXk-pTwW je un3eq pey AT weaZoad uot IezT[Tqeqs eyy

aZeartes 02 ZutTdoy ‘seinseew [eosT} [eUOTITppe pue ejer eZueyoxe TeTOTJJO ayq Jo UoTIeNTeAep e peoUNoUUe QuewUAeACS ey} ‘Aequeoeq ATAve UT

‘lequeseq UT uoTsstTwe AZTeQeuoM UT esPeroUT dieys

eB asneo Aqerz0yq pue ‘sesso, ZutTqeredo sqt 4sooq ‘Aqep s,yUeG [eIQUeD

ayQ ZuToTATes JO Asoo oyA esTeA 02 ‘QgET UT SB ‘SEM STYI JO Joazze

ayL ‘untwerd jeyzew TeTTezed eyQ uTeQUOD 03 AJepio ut AdT Tod AxeQouow

“617

-20-

III. The Structure of the Parallel Market in Argentina III.1 Institutional Structure of the Market

Compared with many developing countries, Argentina has a well developed financial system. Foreign exchange transactions are engaged in by the Central Bank, official banks operating as agents of the Central Bank, private commercial banks, and authorized entities specializing in foreign exchange transactions. In addition to a comparatively deep spot market, a shallower market in forward transactions usually has operated, though forward transactions generally have not had maturities longer than one month, and activity in this market has dwindled when macroeconomic uncertainty has been high.

During periods of exchange controls, restrictions have been circumvented through a number of means. First, many of the same banks and exchange houses providing legal exchange services also have provided illegal services "under the counter". The authorities have been notoriously lax in their enforcement of exchange control regulations, and generally have been content to allow the parallel market to operate without interference. Second, an active market in foreign exchange has developed in neighboring Uruguay, which has been extremely accessable to many Argentine residents. Third, the secondary market for dollardenominated government bonds (BONEX) has provided a legal mechanism for otherwise restricted capital account transactions. Generally, it has been legal to buy and sell BONEX for both domestic currency and dollars; the relative prices of BONEX in the two currencies have therefore defined an implicit black market exchange rate.

During periods when dual exchange markets have been legal, three

exchange rates have been in effect: the legal commercial rate, the legal

uT ‘ZeAemMoy ‘9°gG ATGEL UT peqeoTpuT sy ‘jexAeW TeTTezed eyQ ut aTjoad e qe eSueyoxe usTer0F sseoxe ATEYI T1esS 03 Jepio ut peseyoind ATTenQoe ueryy Ssjiodwt elow erleTIep 02 eATQOW Be sAeY SXeqaodmT ‘YysTY sT umTWerd JoeyxreU TeTTered eyQ useym {B3utotoauTzeao jrodwy useq evry AYsTU JeYxTeW YorTq ay out A{ddns eZueyoxe usTe10J Jo eodAnos puodves e ‘ATTeTIUeIO”g ‘szeek aATJ esoy. UT sqaodxe estTpueyoreu peqiodeiz ut uot{{Tq 6¢€$ ATYsno1z aya Jo Queorzed QT Anoge Ao ‘uOoTTTTG #$ ATYSNor peTeq0R BuToToAuTAepun qzodxe (peqetnoTes reek 4se7eT E42) 98-7861 BZUTANp Jey. saqeoTpUT ar ‘Aaqunoo Zutqrodwt ey. Jo saTqTAoyane eyQ 09 peqiodez euTqQuesazy wWory SsqioduT 02 seTATAoYyAne euTQuesizy 07 pejtoderz sqiodxe eizeduos 2eu. (6861 ‘IHId) SUoTQeTNOTeO squeseiad g'g eTqeL ‘SuToToAuTirepun Qaodxe y8norzy 2exATeW TeTTered peotad-zeystTYy ey 02 ‘sqdtTeoer ATeYI Zepuerins 03 perztnbez useq evaey steqiodxe erzeym ‘JeyAeWM TeTOTeMMOD ey worgz sqydte0er qaodxe Jo uoTSsAeATp ey} UsEq Sey eT0YI ‘ASATY “JeyxrzeU 9y oqUT eSueyoxe uStTez10g jo settddns [euoTqIIppe Jo seoinos [eTIuejod ee143 AseeT Je useeq eAPYy e10Yy ‘poTied [orQUOO-eZueYyoOXS ey} BuTAng SITE Tend Syy OF SsUeyoRY UsTeroy Jo ATAdNS jo seomog ¢ ITI ‘aqel [eToOTemuoo TeZeT syQ 02 eaTQeUIEQTe Aex ay se uodn pesnooz ST o2e1 OQeI TeTTeAzed oy ‘Apnas sTyQ UT “Zuo, AJoZ AeyVO 9YyQ pepeesedxe ATQuUeysTsuoD eQeI ASeYyATeU ‘6g6T TTAdV pue (861 129qGo290 UseeMjeq Inq ‘9qel [eToueUTy ey (UeYQ peqeToOerdep ero sem ‘ST Jey) pepeeoxe ATTeNSN ae [eT Tered eyQ ‘potized [gg1 Aeqmeseq-eunr ey BuTing ‘eje1 eZueyoxe peqetToeidep elo ey} ueSeq eaey p[noys eqe1 TeTTeIed [TeseTT]T ey To ajeI [eToueUuTy TeSeT ayy AeyQeyM snoTaqo jou st Ay ‘FAOFAd VY ‘eqesz TerTezed ay} 02 eSOTO ATOA STOASET 2e peperzy AT TeAEeUEd sey AT ‘spoTred Ggé6T Ttady -/861 29q0290 pur [861 Tequeoeq-eunr eYR UT SB ‘JeOTF 02 peMoTTe useq sey

aqel [eTOUPUTF eA UsyM ‘“e7er TET TeAed TeSeTTT eya pue ‘ejer TeTOURPUTZ

-1C-

-22-

Argentina importers have been more likely to underinvoice than overinvoice their transactions, probably because the tariffs levied on imports more than offset the benefits to overinvoicing deriving from the parallel market premium.” (By contrast, the presence of export taxes on Argentine exports reinforces the incentive to underinvoice exports.)

Finally, the Central Bank on occasion has intervened in the parallel market to support the local currency, thereby representing a third source of supply into the market. This role was most important during the Plan Primavera, but also has been undertaken during other periods when the parallel rate was considered an important indicator of program credibility, such as during late 1989. III.3 Sources of Demand for Foreign Exchange in the Dual Market

In Argentina, it is likely that portfolio demands for foreign exchange have been more important in sustaining the parallel market than purchases of foreign exchange for purposes of importation. Sales of foreign exchange for merchandise imports have been much less tightly restricted in Argentina than, for example, in many African countries with exchange restrictions. At the end of 1981, when the premium of the financial rate over the commercial rate exceeded 60 percent, there was no

rationing of foreign exchange in the commercial exchange market .° With

>/ These data mirror findings by Bhagwati, Krueger, and Wibulswasdi (1974), who applied the same methodology to a large sample of developing countries and found strong evidence of export underinvoicing but no evidence, on average, of substantial import overinvoicing.

6/ There was in effect a regulation requiring importers to postpone import payments by 180 days after the purchase of merchandise. (IMF, 1981) Given the availability of supplier financing, presumably this requirement did not substantially increase the demand for foreign exchange in the dual market.

‘uTege pousqy3tq e1emM suUOTIOTIAQSeA

qaodut ‘sog6, ATrzee eyA Fo sesTAo squoeuhed-jJo-soueTeq seyQ 03 Auenbesqns

‘potied zoy op zeuTqzeQ ey. Butanp peztTerzeqtyT ATTetqaazed ATuo 920M

eseyQ pue ‘sqzodut 02 szeTAIeq JFTAeQ-uoU pue FFTAeQ YSTY AIT9A padoTeaep

pey euTquesiay ‘AZeqeaqs uoT3¢naTAsqns-jazoduy sat Fo azed sy ‘owmt3er

Qaodut [ezeqTT & YIIM pesnyuood oq You pTNoys euTQUeSIy UT sQaoduUT 10J @qe1 TRTOTFFO oyQ Qe oBueyoxs ustez0F FO AAT[IGeTICEAS VATIPTOT oY JL

02 pueQ ‘zeYyWO YORE YATM peyAeTertt0; ATAood Texr9UEes UT eTTYM ‘euTQUeZIy

wots AUZTTF Teatdeo yo seianseou querezssFtIp ‘ATTCUCTATPpy ‘“sueqtT

22@40T2-31q AeyAO pue ‘eqeqse [eet ‘seTTqowoqne jo sseyoind ey. uT sae T{op

jo asn 944 09 szeTTeqez Aq sze[[op ut seotad Jo Butjonb oy worz ButTBuerz

‘epuTQqUeZIY UT UOTIeZTAPT [Op Jo souepTAS peeidsepTM sT ereyy *PUTIUeZIy

UT esse OTTOFAIA0d e se ‘szeTTop ATAe[NoT_zAed ‘eBueyoxe usTer0F

yo sourqiodut ButseerouT ey 03 QuTod suoTjereptsuos snoTiea ‘sqizodut SUTOUPUTJ UT JeyZeW TeTTezed ey. Fo eToI AouTU ey. YATA pereduoyg

"qeyzeu TeTTezed ey. ut

aZueyoxe AOJ puewlep MOTF OQ 02 peanqTazAQUOD sAeY PTNOM oOSsTe eseyQ ‘JeXxITeU

TetToremmos ey. YysnorzyA peoueuTF eq 09 peqaQtTuAed Jou eTeM saodTAres Jey O

pue wWstinoj useym spotied Butang ‘sxzeTrAreq eATATAUeNb AO szZTAeQ Qaoduyt

apeASe 02 SeTITAoYANe sYyQ wWorZ UeppTYy ATeIeAEqTTep sqioduT BuToueuTZ

SPM JoyreU TeTTered ey ‘ATqeqorlg ‘“eyxTeW [eT [ered eyA uT peoueUuTZ Buteq

@10M suOTIOBSUeAA QiodWT oUlos YseeT Ae QeYQ AseB3ns seop ‘Aeaemoy ‘Z°III uOTIOeS UT paessnostp ZutToToauTAepun Aeqrodut Fo souepTae ey

*sqaiodwy 09 szeTrAzeq JJTAPQ-uoU

ZO JJTAPI AOYATS UT sUOTIeTIeA 02 eATSUOdSeAT Jou seM ajeT JoXTeEM TeTTeIed

eur 3aeya saseB8ns pI uot 9eg UT pessnosTp eoueptae ‘AT [eUTY y Srt0dut

esTpueyotem 4Ysow Jo ButToueutTF Jo eoinos AzAofew e Jou sem AT [e198Ued

qeyreU TeTTered ayy AeYyA sqyse88ns soUepTAs [eQOpoeue yjnq ‘juesuUTIAs

aToW suedeq sUOTIOTAASeA AAOdUT ‘oTQUeTIY YANOS ey UT AoOTTJUOo ayy

-€S-

-24-

agree that during 1980, 1981, and 1982, private residents exchanged domestic assets for over $15 billion in foreign assets. (See Appendix Table B.6) Estimates of the stock of flight capital outstanding by the end of the 1980s usually exceed $20 billion. Most of these assets are held overseas, but it is generally believed that about $5 billion in U.S. currency is being held in Argentina "beneath the mattress". By comparison, domestic monetary aggregates Ml and M2 (measured using the official exchange rate) totalled only $4.5 billion and $7.7 billion, respectively, during their mid-decade peak in December 1985. Iil.4 The Efficiency of the Parallel Market for Foreign Exchange

The anecdotal evidence suggests that the authorities did not interfere significantly in the operation of the parallel exchange market. Given the importance of foreign assets in domestic portfolios, a wellfunctioning parallel market would have equated average rates of return on dollar assets to average rates of return on austral-denominated assets, assuming these assets were considered to be substitutable. To evaluate whether this was the case in Argentina, the following equation was

estimated:

EP /Ee a+ pl(l+ ipsa + it di +e, (1)

EP: parallel exchange rate (local currency per dollar)

i: domestic interest rate - either deposit rate or free loan rate - (monthly basis)

i: U.S. 1-month Treasury Bill rate - (monthly basis) e : error term

t : time subscript - months

Sv

Ov

SE

oe

Se

02

Sk

Ol

+O!

OL

St

02

Se

oe

Se

Ov

Sv

OS

SS

09

OL

SZ

08

0661

8861 9861

wi re I aN

veel c861 O86l 8261

ajey ueo]- — — — ayey ysodaq

(Ajyuoyy)

seyey ueo7 pue ysodaq pejsnipy-uoijemadeq

pueuo

-25-

The estimation sample extended from May 1982 to July 1988, the longest period of (roughly) unchanged exchange controls. Estimates are made using both the 30-day deposit rate, which was frequently controlled by the government, and the freely determined loan rate. A_ priori, the deposit rate represents the more appropriate candidate for a rate-ofreturn variable; because it was controlled, however, the loan rate might be better correlated with the rate of return to which the major players in the financial market had access. Chart 4 indicates both rates to have been closely correlated during most of the sample period.

Estimates of this equation using developed-country data generally have rejected the three implications of exchange market efficiency: a 8 equal to one, an a equal to zero, and a non-serially correlated error term. (Baldwin, 1990) By contrast, the estimation results presented in Table 1 indicate that the Argentine parallel market performed relatively well in arbitraging rates of return between domestic and foreign assets.

The Durbin-Watson statistics in both regressions are low, indicating first-order serial correlation of the errors; this was confirmed by additional tests on the residuals. As noted above, however, this problem also characterizes equations applied to markets with no exchange controls. On the other hand, the estimate of 8 using the deposit rate is statistically indistinguishable from unity, and the intercept term in that regression is essentially zero. The results using the loan rate are less consistent with the efficient markets hypothesis, but still confirm the significant linkage between interest rate differentials and the depreciation of the parallel market rate in

Argentina.

(sesoyqueied ut sotjstqeqas - q)

OCT Le°T “A'd

ST" LT" ua

(08° €) (93eY TTqQ-L °S*n+T)

19° /(2381 T ueOT 2079+T) (€0° +) (928Y ITq-L °S*n+T) C6" /(a3e1 T 3ysodeq +7)

(08°T) (se)

9€° 60° que jsuo9

(Z) (T)

da/(1+) ga): @Tqeyze, Quepuedeq

(ATYQuUop) Z°eB6T-S C86 -SUO;IUNDY AZ TAvg ISoiojUy porsaosuy “T STqeL

-26-

IV. The Parallel Market Premium

Chart 1 indicates a number of important features of the parallel market rate’s behavior during the 1980s. First, the real parallel market rate has exhibited strong medium-term swings, depreciating strongly at the beginning of the exchange control period, appreciating gradually through much of the remainder of the decade, and then depreciating to record levels during the hyperinflation of 1989. Second, the parallel market rate has fluctuated sharply around these trends, exhibiting much greater volatility than the commercial exchange rate.

This behavior is consistent with a view of exchange rate dynamics that distinguishes between long-run "fundamental" determinants and short-run "portfolio" determinants of exchange rate behavior. (See Dornbusch, 1983, and Kamin, 1988, for applications of this framework to black markets for foreign exchange.) In the following pages, we derive and estimate relationships for both the long- and short-run determination of the parallel market rate.

In addition to shedding light on the parallel exchange rate's behavior, these estimations identify various means by which the government could, and did, affect the parallel market premium. In the short run, contractionary monetary policy that raised the domestic interest rate would have reduced the demand for foreign assets and appreciated the parallel rate. A more sustained reduction in the premium would have been achieved by devaluing the commercial exchange rate and/or reducing export taxes.

IV.1 A Simple Theoretical Model Long-run behavior Equation (2) presents the steady-state equilibrium

condition for the parallel exchange market. In equilibrium, the stock of

°(Z) uotqenbe

UT peUTJEep QUNOdoSe QuezANd eQeaTad eyQ YZnozyQ ANdD.0 ATUO ued sjesse uStTei0y ssoiZ Jo yoo3s oyQ uT esueyo Aue ‘ertojsert0yL “1[g6l 799Fe ONAQ

Ayazeutxordde sem styq {setqTque usTeroy pue squeptser oTQseuop eqeaTtad usemjeq soe[d ayeQ suoTJORSURIQ QUNODODe TeQTdeo ou ey peuMsse ST QI /8

ay. seztieuums moteq (¢) uoTJenby ‘ozez ATJOeKe eq 02 BOUPTeq JUNODOe quezino eqeatad eyQ esneo [TM , ga 9981 JoyTeU TetTezed umtaqt{tnbe enbtun e -- setTotT{od eperzq pue ‘eperq Jo swizeq eyQ ‘yndqno [ee1 ‘Teas, votad eyQ ‘eqer eSueyoxe TeToremmod 9yQ -- soUReTeq JUNODDR QuerzANd aqeatad ayQ BuTQoesjJe ,S[Tequewepunz, SNOTAPA FO STeAeT 9Y4} USAT "queqiodwt useq sary osTe Kew Qeyxrew [eTTeazed oyA uT poser pue JeyreU [eTOTemMMo0D ey. UT peseYyoind SIBT[OPp YyBnoyz Te ‘sadteoer qiodxe peotToauTiepun worz ATUTeW peATIep sary 0} peumsse st joxreW TeTTeAed oy oqUT saeT[op jo Atddns oy, “szztreq jo @ATSN[OUT eSueyoxe ustetoz Jo soTad eaTQeTeA ey. uodn Zutpuedep ‘sjeyrzeU [enp Jo [eToreumod eyQ YysnoryQ sjioduT ATeYy_ soUPUTZ 09 AeyQEeYyM sesooyos 03 eTqe e10M sreqiodwtT ‘queqKXe ewos 03 Jey poeuMsse ST QT ‘sazeTazreq qZeyo pue sJjTzeq ptoae 03 (pe [Z8nus 10) peotToauTiepun Jo Qoeyxrew sdueyoxe [eToLemod ay UT eTqeToueUTJ Jou JeyITe (WSTInoZ BuTpNToUT) seTITpoumos yo sqiodut worj puewep peATiep e ST SAeT [OP OFZ puewlep MOTF 3yL squeptser eqeatad kq peumo (sie[[op ut) sjesse uSTer0z ssoiZ jo 4009s 7 qeyreu TeTTeAzed UT sAzeT[[oOp AoF pueuwlep moTF :d qoeyrzeu TeTTered oqut sazeq{{[op Jo Atddns moTZ :s

(2) Oo=-a- Ss = dP g 2eatem Tetterzed eyq oqUT saeT[oOp

jo settddns moTZ Tenbe ysnu saseyoind Qunodoe quetino AoF SAeT[Op OF

spueuiep MOT {UeQSUOD eq Ysnu JoJOeS ajeATAd syq Aq pouMO sjesse usTe10F

-1@-

-28-

effect of these various fundamentals on the equilibrium parallel market rate. The direction of these effects, shown in parentheses above their

respective arguments, are derived from a more complete model described in

Appendix A. (4) G4) (2) 0) GH) OG) (|) * * Br = fC E,P,Y, PY, Py, thy. ty, ary) (3)

E: commercial exchange rate (local currency per dollar)

P: domestic price level Y: real GDP

* , *

Py: dollar export price Pu: dollar import price tiy: import tariff rate ty: export tariff rate

Wy: level of non-tariff import barriers

The relationships indicated in equation (3) generally have straightforward intuitions. Focusing on key policy variables, a depreciation of the commercial exchange rate is likely to reduce the parallel market premium, reduce underinvoicing, lower the flow of dollars to the parallel market, and hence boost the parallel market rate. Conversely, an increase in export taxes will divert export receipts to the parallel market, thereby appreciating the parallel market rate. Increases in both import tariffs and QRs are likely to shift the demand for dollars from the commercial to the parallel market, leading to a depreciation of the parallel market rate.

Short-run behavior In practice, the actual parallel market rate EP is

likely to deviate occasionally from the equilibrium rate EP’ described in Equation 3. Out of equilibrium, it is presumed that the parallel market rate moves to equate the portfolio demand for foreign assets with the

stock of these assets, which is predetermined at any point in time by the

aver eZueyoxe Tenjoe oy} YIM BZuoTe ‘¢ uoTjenbe uT peTfTqUepT seTqeTzeA AzojeuelTdxe eyy ‘sdtysuotjetez esey. SuTTepou 03 yoroidde uotjo0e1100 -1OTIe/UOTAILAZEAUTOD (@eSO0O0T) B aYeI OM ‘eaoge peqTiAosep JoOTAPYaq 9qe1 edueyoxe uni- 1o0ys pue unI-ZuoT 9yQ usemjeq YsTNSuTASIp of sz[nsey Teotatdig ¢ “AT *‘pezoqsez st umtaqt{iInbe [1qun eqQetoerzdde 02 e9e2 TeTTePAed ey BuTsneo ‘(7) uoTJenbe ut e3e1 TeTTeAzed ey UO QoeFFO SATQREZOU SRT YBnozy? ‘pue oqerTnumooer 09 ¥ sjesse uSTeIT0J ButTsneo ‘snidans qunoooe quezano e& 02 pret [TIM STUL ", qa 0982 wunTiqt[ Tnbe ey? eA 4A eqe4 aZueyoxe TenqQoe ayQ JO UOTIeTASP seATIATSOd e& ZeptsuOD ‘gd sqesse ustez0j Jo queuwqsnfpe eyq ysnorzyQ 1qd eel uNTAIQTTInbe ayy piemoz pueq p[noys jy 29e41 TetTeazed Tenqoe eyqQ ‘AT TPoTIeAOSUL "qd eqer T°1lerzed snosureroduequoo aya JO enTeA sy SouTUAEIep (7) UoTIeENbe ‘uoTQeTOeAdep eAe1 QoYyAeW TeTTe1zed peqoedxe pue ‘seqei jsezequT ‘qandqno ‘sjesse uStTet0g ‘y00385

Asuow ey. JO STeAeT pouTWAeQepetd osTMrey}O AO peT[orqUOD-jQUeMUIeACS AOg

q 28 poeqoedxe

[+9 22 uoTQeTOeAdap aqer sBueyoxe [eT Tesed wa « 1433 ‘ _ (7) Cx qi +,F © Fm = (aga /W

(+) (-) (+)

:;Aquow oF puewep suoTOPSUPIQ 9YQ SeUTMUATeJep YOTYM ‘AQTATQO" OTWOUODe Fo TeAeT ayy pue ‘sjesse usTet10F uo uAnjer Jo eer oY. ‘eQ"1 YSeT9QUT OTQsSeWOp sy Jo sTseq eyQ Uo pezeTNUAOF ST s}esse USTeTOZ 02 sATIAeTeA AdUOW OTQsSoWOp A0J pueuep su (dg) SqJesse uSTez0J Jo enjTea AouerANS oTQsewop ey pue (H)

aqedeiz3e ATeJouoW ZuTAPeq-JsetEeqQUT prorg eB UsEeMQeq YQ[FEM ATeYyQ sQeOOTTE

03 peumsead ele squeZe eqeaTIg ‘“QUNODOe QuezZAIND ey Fo AAOQSTY snotaesead

-67-

-30-

EP | are hypothesized to form a cointegrating vector; the errors from this equation, while not necessarily serially independent, should be stationary. The variables identified in equation (4), the portfoliobased short-run equation, should form a second cointegrating vector. To the extent that this equation held at all times, its errors presumably would be serially independent as well as stationary.” Long-run behavior Table 2 presents results for the estimation of the long-run cointegrating vector. The equations were estimated, using both monthly and quarterly data, over the period from May 1982 through July 1988, the longest period of continuous, "normal" operation of the parallel exchange market. Monthly data on import prices, export prices, trade policy variables, and real output were not available. For monthly estimation, the U.S. wholesale price level, combined with a measure of the U.S. real exchange rate vis-a-vis its major trading partners, was used as a proxy for foreign prices; implicitly, this formulation constrains the terms of trade to be constant.

In the monthly equation, the residuals were highly serially correlated, making interpretations of confidence intervals around the

coefficient estimates misleading. However, as indicated by the augmented

9/ The Johansen procedure was applied to a subset of the variables of interest -- the parallel exchange rate, the commercial exchange rate, the domestic price level, and the money supply -- to determine the appropriateness of our modelling approach. The tests supported the hypothesis that two cointegrating vectors linked the variables in question. They also failed to reject the joint hypothesis that (1) the elasticities of the parallel rate with respect to the commercial rate and the domestic price level summed to unity (homogeneity), and (2) there was a unit elasticity of the parallel rate with respect to the money supply. These findings bolster the view that distinct "fundamental" and "portfolio" relationships act to influence the parallel rate. See Melick (1990) for an almost comprehensible description of the Johansen procedure.

‘(seseyqueied ut sotqastaeqs - 4)

9S €- 89°€- 91° 9- sjTenptsey jo “A'a'v 90°¢ L6°T 8L° “Md 00'T 00°T 66° zu L0° 10° OT’O uotsseisey

jo 70171q paepuejs

ee

(97° -) et’ - Tob (07 ?-) (88°T-) x Le°T- €€°1- 3 (70°) w 60° ja (ST°€-) (68°7Z-) GeI- LE'T- (A) Bot (81°T) (17°) W py? LG’ Cd 307 (S0°Z) (26° T) x et" Ly" CD 3o_T (L6°7-) 82" - (ond) 307 (17°47) LTE (ond) 201 (17 €) (Z1°€) (€z2°€) OL’ €L° €y" (d) 307 (07° T) (96°) (1L°€) val Ze os’ (gq) 307 (€) (2) (1) ATaA0gAeNH AT qjuoW

8861 AIne-Zg61 APH (qa) ZoJT :eTqetTire, Quepuedeg FysuoTIeTSoy , [ej usuepuny, UNny-suc]T +suoTjenby o}ey ssuelyoxy [olT[eied °c O1deL

-31-

Dickey-Fuller statistic for the residuals from this equation, the errors are stationary, suggesting that the variables do indeed form a cointegrating vector. In the quarterly equations, the errors are not only stationary, but apparently are serially independent as well.

Turning to the coefficient estimates themselves, note that in both the monthly and quarterly equations, the coefficients on the logs of the commercial exchange rate and the domestic price level sum to close to unity; formal F-tests failed to reject this restriction. This implies that equal growth rates of the price level and the commercial exchange rate will generate proportionate depreciations of the parallel rate, which is an appealing homogeneity result. A further implication is that an increase in either the price level or the commercial exchange rate, by itself, will generate a less than proportionate increase in the parallel market rate. Hence, real devaluations of the official exchange rate will generate less than proportionate devaluations of the parallel rate, thus reducing the premium.

The quarterly regressions enable us to evaluate the impact of different trade policy instruments. The export tax rate apparently had a significant negative impact on the parallel market rate; presumably, and consistently with our predictions, increases in the export tax rate increased export underinvoicing, diverting additional flows of dollars into the parallel market and driving down their price. Conversely, neither import tariffs nor QRs on imports significantly affected the parallel market rate. This is consistent with other evidence that the flow demand for foreign exchange was not an important factor in the

parallel market.

(9) *e((9t1)/T) + CxSotc(a+t)/¥) - (3a) 80 ((g41)/9) +

*5C(g41)/2) - (7a)80T((4T)/T) - Cw)80T((9F1)/T) = Ca) 80T

:enyTeA peqoedxe sqt 1z0z Axord

e se ia JO enTea Tenjoe ey. esn 09 uoTjdumsse suoTqjeqoedxe TeuoTtqeI1

aya BuTyoaut ‘epts puey-3usTz eyQ uo savedde oqeir qoyrzeu ToeTTezed

aud FO eNnTeA sANQnjJ oyQ ATUO AeYQ OS sUZeq YOSTTOO om ‘uoTQenbe ayQ

wolz ‘Wopeezy Jo seerzZep saresuod o9 ‘peddorzp st aqel YseraQUT UsTerOF ayy ‘aqel 4YsezTeqUyT oTQsewop eyQ Aq AAT[TqQeTAePA Pue eZTS YIOq UT pezAeMp

ST 21 osneoeg ‘W192 AOAIAe sSTOU-aeqTYUM B aq Of poeUMsSse sqT 5 ‘er0yH

(s) 4a 4 (A*x)B0TY -

[ (ga) 80T- (Taq?) Sot] 7" + 720 - (a)80T - Cw)B0T = (3a)30T

:eaoge (17) uoTQenbe ezTAPeUTT-SOT ASATJ em ‘ae QoYAeU

TelTlTerzed eyq Jo AoTAeYeq UNA-AAOCYS ey SeqeUTISe OL AOTAPYSq UNnA-FJA0YS “SOTQeTAPA |sSayQ Usemjeq dTysucTjeztez

uni-Zuo,T eniq ey} TeeAerT 09 Zuo, ATAUEeTOTJJNS Jou sem [PAIEQUT UOTIeUTISE aud 2eY42 eTQTSsod osTe ST AI ‘SOS6T eu BUTANP seTqeTAeA OTQSeUOp TeuTMOU 09 SATQETEA seTqeyzea sotad usTezo0F oy Fo UOTIeTAPA FO TOAST aqnutW ey. AoeTyer Ae seTouezsTsuooUuT eseuy ‘UZTS Buorm 9yQ sey sjiodxe yo eotad eyq uo QueToOTZjze0o oy. eTTUM ‘4seq qe ‘queoTsTusts ATT euTsiew ATuo eq 03 savedde ootad qaodut eyq erzeym ‘suotssez3ea1 ATAeqAenb 9yQ

UT JUepTAe JSOW ST STULL ‘0381 QOYAePW TeTTeAed eYyQ UO AoRdWT pejoTpead

ATeyQ eaey jou op seTqeTseA VoTAd usTer0y eyQ ‘ATSuUTIASOUOOSTG

-¢@C-

-33-

No data are available on F, the foreign assets held by the private sector. However, the residuals from the long-run cointegrating regressions described above and presented in Table 2, which we denote Ups provide an estimate of [log (EP) - log(EP')], the deviation between the actual and equilibrium exchange rates. We posit in equation (7) below a

log-linear approximation of the relationship between the private current

account surplus and this deviation:

log(F.) - log(F,_4) = n{log(Eb) - log(ER’)] = nu, (7)

n : elasticity of current account balance with respect to deviation of exchange rate from equilibrium

Hence, u’, the cumulative sum of these residuals, should be proportional

to the level of private foreign assets outstanding:

lM 8

= P - P, eos) 7 Blog (ER ;)-Loe cee!)

= A/nQELC Log (Fy 4) -L0B(Fy 5 4] (8) i=

= (1/n)log(F,) - (1/n)log(Fo)

We substitute ULL for log(F) in the actual equation estimated: 1°

10/ The lagged value of u’' is used, since the contemporaneous value will include in its computation the dependent variable itself, and hence lead to simultanaeity bias. In a loose sense, u’ plays the same role in this equation as the residual u, from the cointegrating regression would play in a more conventional error-correction equation; it links short-run dynamics to long-run behavior, forcing the dependent variable to move toward its long-run equilibrium value.

‘asTou eqaTym aq 02 eNnuUTQUOD T[TM uoTQeNbe sTYyQ Wor} sAOAAe ey. os ‘QuUeQsUOD sYyQ Aq dn peyotd oq [ttm w1e3 (97) 80 (41)/T ayuq ‘uotTjenbe poejewtjse ey ul /Tt

Aqatazed e4eit AsertequT eyq UT wieq eQe1 QTsodep ey} Uo QUeTOTFJ909 ATun ayy yaIM QUueASsTsuOD sT sTseyjodky sTyQ ‘spueuep Jesse eATIJeTeA Jo JueUTWAEJep Aoy oyQ ST Ter UetessIp O9eA YsoerequT ey. aeYyQ BuTATdwT ‘eqer eSueyoxs PT ey. Uo 4eYA FO eATQeZOU OYA 9q OF PaeROTAASeA ST eRe ASeTEQUT ay. UO QUeTOTZJZOOD oyQ ‘AT[PUCTITPpe ‘Z# TePOW peaoTAIsey UT °(¢) uotjenbe ut petgtoeds ATTetatut Atddns Aseuow ey 09 Qoedsez yRTM aqe2 aZgueyoxe ey Fo AQTOTASeTS ATuN 9YyQ sJoOeTJer sTyQ ‘eaoqe (6) UoTQIenbe Aq petitdur se ‘ejei e8ueyoxe pet ey uo 4eYyQ snuTw euo Tenbe 09 pejoTaqsel st A[ddns Aeuow eyQ Fo BoT eyQ uo QueTOTZJe0o ey ‘TH [PPOW peqoTaqsoy

UI ‘peQeWTAse elem [epow oTseq ey} FO suoTAOTAASeA OM] ‘suotjenbe 944 o4UT peqertodioouT orem wieq ,N vy. Fo sBeT snoTzea ‘uoTsserZe1 BuTAeAZequtToo unz-ZuoT sy. worzz S[TeNpTsert oy. 02 S3UTp[oY Jesse UZSTeOJ souSeYy pure QUNODOe QuUatInd JexAeU YORTQ ey Fo osuodser pesZeT & JO PpoouTTEATT ey. AoeTJer 02 ‘ATTPUTY ‘TBATEAUT UOTIeUTISSe eYyQ Jo Aaed A[UO BuTAnp peTToOrQUOD-JQUeUTIIeAOS Sem YyoTyM ‘aqer ysortequT eyq Jo AQToueZopue eyj wWorZ se [eM se ‘suoT Ie Qo.edxe aqei e8ueyoxe rz0z Axoad ay. UT JoAZe Quewernseeu WoIT BuTATNsel setq Ao0F 30eII0D O01 pasn sem UoTIeUTASSe (AI) SOTQETAPA TeQUeUMAQSUT ‘UOTIeUTISS SIO 03 UoTaTppe ut ‘ATeaTqoedser ‘eqep ATAeqAenb pue ATYQuow Butsn (6)

uotTjenbe jo suoTIeWwTAse ATOZ sarnser ey queseiad gE pure VE seTqeL

pO BOTUHHT/T) - Fa((g+T)/1) 22a

(6) 2a + (x) B0t((d+1)/X) - (Ga) S0T((941)/9) +

*2((d4t)/0) - VInuC(g+1)/1) - Cw8ot((d+1)/1) = (Ga)80T

-E-

-34a-

Table 3A; Parallel Rate Equations: Short-Run "Portfolio" Relationship

Dependent Variable:

Monthly Data:

Unrestricted Model

(OLS) (Iv)? Log (M2) 12 -.03 (2.32) (-.37) Deposit Rate i -.59 -1.03 (-2.93) (-3.92) Log (EP) .88 1.04 (15.18) (13.51) uy .70 .69 (6.08) (5.67) u'» - 87 - 84 (-4.52) (-4.13) ul 3 .28 .29

(2.50) (2.47)

Standard Error of

Regression .06 .07 a)

R 1.00 1.00 D.W. 2.12 2.14

A.D.F. of Residuals -8.79 -8.96

a Instruments:

Log (EP)

August 1982 to July 1988

Restricted

Model #1 (OLS) (Iv)?

1-.88 1-.92 (--) (--)

-.59 -.72 (-3.53) (-3.84)

. 88 .92 (24.58) (23.48)

.70 .65 (6.22) (5.77)

-.87 - .82 (-4.57) (-4.26)

.28 .25 (2.57) (2.31)

.06 .06 94 - 94 2.12 2.12 -8.79 -8.94

A log (P), A log (P_,), Log (EP_,), Log (EP 4),

(EP),

(t - statistics in parentheses)

Restricted

_Model #2_ (OLS) (iv)?

(29.10) (28.65)

.67 .64 (5.89) (5.54)

- .84 - .80 (-4.33) (-4.07)

. 26 .24 (2.36) (2.15)

.06 .07 .94 .94 2.04 2.04 ~8.6 -8.77

Log (M2), Log (M2_4); Log (M24), iy: iio: iis, wy u' wis, Log (EP 4), Log

Log (eP .), Log (E), Log (Ey); Log (E_5); constant

886l L864 986} S861 PRE; eest c86h o8t

Ole

Ore

OL2e

00€

O0ce

i Nw y \ r\

09€

O6E

~_— _— woe

——— ee oe ee ee meee ae me —,

Ocr

a ~

OSP

O8r

OLS

Ovs

OZS UNY-HOYUS > = — = — uny-6Buo7- —— — rensy

009 001 = 1864 AeW

(AjujuO;\) senjep peli4 uNyY YWOUS 9g Bu07 “sa jenpy :YF Jal/Bed |eoY

Bs UeYD ~~

-35-

equation estimated in Section III. F-tests failed to reject the restrictions underlying either of these models.

The monthly estimation results shown in Table 3A strongly affirm the importance of portfolio factors and expectations in the determination of the parallel market rate. The standard error of these regression are considerably lower than that of the "fundamentals" equation shown in Table 2; anaiysis of the residuals of these equations indicated them to be serially independent as well as stationary. Chart 5a c.mpares the fitted values from the IV 2stimate of Restricted Model #2, transformed into real exchange rates, ‘to those from the fundamentals equation presented in Table 2 and to the historical data. It is evident that while the fundamentals explain most of the broad swings in the real parallel rate, portfolio variables capture the short-run dynamics more accurately.

Turning to the coefficient estimates themselves, note that the coefficient on the led exchange rate is estimated at .95 in the IV estimation of Restricted Model #2. Therefore, the elasticity of the black market rate with respect to the differential between the deposit rate and the expected rate of depreciation, the original f (and therefore a) in equation (5), is calculated 19.00. This is an extremely hizh elasticity, and implies that in the short run, the government could 2nfluence the parallel market rate considerably through its influence on monetary conditions and the domestic interest rate. It also implies that the parallel market rate was extremely sensitive to expectations about the future rate, as most anectdotal evidence will confirm.

The estimation results using quarterly data, shown in Table 3B,

broadly resemble those based on monthly data. Given the relatively few

‘qg 32eUD UT peqeotpuT se ‘uoTjenba ButTqerzZequtoo

uni-SuoT ay. 02 eATJeTeET ‘uoTJenbe ayA Jo ATF rzood AToeaTAeTeA aYyQ A0Z qunoooe osTe p[nos uoTAe[erz00 Azood sty, ‘uotjenbe ATrzeqazenb ey. wor; queToTsjooo ey. UT seTq pxzeMuMOp e pue JOIIe JUsueAnseeM UT ZuUTATNse2 ‘suoTqeqoedxe aqez eSueyoxe Tenqoe pue eqer eZueyoxs ATAeqAenb peT ey usemqeq uoTeTeAI0OO Jood e yoeTJer Aew eqep ATAeqazenb pue ATYyQuow Zutsn si[nsez eyq ueemjeq AouederzostTp ey, ‘uotTsserZex ATYyQuow oy. Aq pet {duty qeyd MoTeq [TEM ‘¢E'T ATUO Jo TeTQUeTeFFTP e7eAI AseTeqUT sey 07 Joedser YaTA eqear eZueyoxe ToeTTezed ey Jo AAToTAseyTe-Twes e seT[dmyT sty */¢° ATuo st Zf# Tepow peqotazasey Jo eqeMy se AT ey. UT eqer oBueyoxe peT ey uo QUeTOTJJe00 ayy ‘eZuer Tensn ey. UTYITM TIeM ‘7° T IJnoqe Jo jandjno 03 qoodseaz yqTM puewep Aeuow jo AqToOTAseTe ue seT{Tdwy qqd Tver Jo ZoT eyq uo QUeTOTJJO0D SUL, ‘SsuOTQeUTASe ATYQUOU eYyQ UeYA eTqeTI[Eer SseT petepTsuos

eq plInoys fey. ‘reaemoy ‘stseq ATAeqAenb e UO eTqeTTeAe suoTIPATESqO

-9€-

-36a-

Table 3B: Parallel Rate Equations: Short-Run "Portfolio" Relationshi Dependent Variable: Log (EP)

Quarterly Data: 1983 I to 1988 II

Unrestricted Restricted Restricted Model a Model #1 a Model #2 a

(OLS) (IV) (OLS) (IV) (OLS) (IV)

Log (Y) -1.79 -1.83 -1.90 -1.86 -1.89 -1.90 (-3.60) (-3.56) (-3.84) (-3.73) (-4.22) (-4.16)

Log (M2) 97 46 1-.58 1-.59 1-.59 1-.57 (3.52) (2.60) (--) (--) (--) (--)

Deposit Rate i .12 -.47 -.54 -.72 -.59 -.57 ¢.15) (-.50) (-.80) (-1.05) (--) (--)

Log (EP, ,) 41 53 58 59 59 57 (2.36) (2.77) (5.26) (5.36) (9.40) (9.29)

wa -.04 -.06 .17 -,02 .18 -.05 (-.07) (-.12) (.37) (-.04) (.42) (-.11)

u's - .83 - .89 -.75 -,88 -.76 - .84

(-1.90) (-1.98) (-1.71) (-1.99) (-1.90) (-2.11)

Neen ee eee Un nIE EE EET EIS ag Sese ERED enadne mneRRemnenenanedenaenaneainnnnEnnnan enn en tieianeicn ae

Standard Error of °

Regression .09 .09 .09 .09 .09 .09 R? 1.00 1.00 86 87 .87 88 D.W. 1.90 1.85 2.22 1.95 2.23 1.92 A.D.F. of Residuals -3.17 -3.13 -3.78 -3.34 -3.79 -3.32

a . . * “s ’ ' ‘ Instruments: Log (M2), Log (M2 4), Log (M25); ip iia: iz, uy, wig ¥ -3, A log (P), A log (P_4), Log (EP 4), Log (EP 4), Log (EP 4), Log (EP), Log (EP 4), Log (E), Log (E_4): Log (E54), constant.

(t - statistics in parentheses)

8861 286} 9861 S86} ys6! es6h c86} 00¢

0ce Ove 092 082 o00€e OcE Ove ogee O8e 0oy Ocr Ove

09r

O8r

00s

0¢S UNY-HOYS » = ae oe uny-6u07 - — — —

lenoy

ovs 00! = 1861 AeW

(Apeuend) senjep peyi4 uNY YWOUS g Buo7 ‘sa jenoy :yq je}/e1eq [2OY

qg ueYd

-37-

V. The Parallel Market Premium and its Impact on the Balance of Payments

In this section, we evaluate the hypothesis that increases in the parallel market premium led to significant deteriorations in the balance of payments in Argentina. Equation (10) summarizes Argentina’s

official balance of payments:

dR = OX - OI + PXF + PMF + NSBAL + KA (10)

R: gross international reserves held by Central Bank OX: official exports OL: official imports PXF: pre-export financing PMF: post-import financing NSBAL: net services balance

KA: capital account balance

Based (in part) on the discussion in Section III, there have been three main points of linkage between the official balance-ofpayments and the parallel market in Argentina: diversion of export receipts from the commercial to the parallel market, diversion of import financing from the commercial to the parallel market, and conversion of profits earned on pre-export financing through the parallel market (see below). The following sections describe attempts to determine whether those linkages can be discerned in the time series data. To summarize the results of the analysis, increases in the parallel market premium

appear to have been associated with strong reductions in both official

setTazqunoos qogo0 J° ddd : x

sqjiodxe [eToTzyo ‘teea go Atddns : ( )xo

(ZT) C AC AS (83-1) 9/49 ‘a/Xa(%a-1)a)x0 = ¥a/X0 (+) (+) (-) (+)

:(ZT) uotTqenbe ut umoys se pouTwre ep aq 02 poumsse ete sqiodxe [eToTFJO [eer ‘ezojyerteyL ‘umtuead Tetteaed poasn[pe-xeqQ ey 07 AjTeatQeZeu puodsez prnoys ZutoToautirepun jzodxe 3nq ‘aqex1 [TeTOLeuMIOD peasn[{pe-xeq eyq pue ejer 2exr1eU TetTTerzed eyA yIoqG OF Ayeatatsod puodser prnoys sqaodxs [eRI0O] xeput eotad q10dxe Ae Top Ma oTqe1 BZutoToautT-azepun :¢ sjiodxe esTpueyoreu Fo euMypoOA :X

sIe[[op ut sydteoer qazodxe [TeToOTsJO XO

(11) ¥ax(#-1) = xo

:ZuTOTOAUT-JepuN Jo ee1Zep ey} uO ATaaTqeZeu pue sqzodxe [#07 Fo eunTos

aya uo ATeaTatsod puedep -- Joyx7eW aZueyoxe TeTOTemMOD ey} UT perTepuerANns asoyq -- sqdqeoer jiodxe TeToTFzo 3eUA soqeoTput (TT) uotaenby

SUTOTOAUTASpuy], teqATOGXY TA

‘sqzodut [TeTOTJFJO Ut

SUOTIETIPA YITM peqzeTetIOD Jou e7eM ANd ‘ZutoueuTy qiodxe-eid pue sqiodxe

-8E-

Constant

Log (E(1-t,)P'*/P) Log (EP/E(1-t,)) Log (Y)

Log (Y")

DUMQ2

DUMQ3

DUMQ4

Standard Error of Regression

Ry

D.W.

-38a-

Table 4; Merchandise Export Equation 1978/IV-1988/II

(Quarterly Data):

* Dependent Variable: LOG (OX/P, )

OLS

-6. (-1.

1.

06 16)

.28 . 60)

.27 .99)

.38 .65)

-o7 (1.

67)

.27 (5.

68)

.20 .85)

.08 .30)

.10

74

91

(iv)?

-1. .17)

(5

1.

10

.23 74)

.39 .06)

24 -46)

.78 (1.

97)

.29 . 66)

-22 . 84)

.02 .25)

. 10

-/2

88

“Instruments: Log (EP"/P), Log (E (1-tx)P"S Pp), Log (EP/E) 1, Log (Y),

* * * ; ty. tf qr Log (Y ), Log (OI/P ) 4: Log (OX/P) a> i

DUMQ2, DUMQ3, DUMQ4, constant.

(t - statistics in parentheses)

-l’

(€1) Mal (9+T) = 10

:ZuTOTOAUTAsAO Jo vserAZep ey pur

sqiodwy esTpueyorem TeSeT Fo eumtoa ey yIoq uodn ATeaTQTSod puedep o} sjiodwy pezeystZe1 AT[eZeT Fo Qunowe eyQ seqeoTpuT (ET) UoTIeNby

SUPOTOAUT-ISAQ TSITOMUT ZA

‘spotied esoyy ut sqdteoer qaodxe [eToTJJo Jo Queorzed

4Z BuTpeeoxe sesso, ydteoer qiodxe BuyjeotTpuT ‘6g6T ATAzve pue ‘7g6T-PTU

‘T9861 e2eT UT Queozed 09 eaoge esor untTuerd oY. ‘peyAeM eTOW Useq SAY

PInom o0ezJe sty ‘STsT10 Jo spoyred Buying ‘potized yey teao Queorzed

eT ATUBnor Jo YoyTeW TeToOTemM0D eyQ UT sjdteoer Yaodxe jo ssoT eBvireae

ue seT[dmy stud ‘SOg6él 242 UT STOrIQUOD eZuUeYyoxe Jo potred ey Butinp

queorzed ze peSereae umquerd [eT [ered eyA yey. BuTAeptTsuod ‘uoTAe[NITeo

adoTeaua-ey3-jJo-yoeq yZnor & sy ‘“TeTRUeISqns eQTNb st YyoTYA ‘tH

qnoqe jo umqwezad yeyzew ToTTezed eyq 03 Qoedser yITM sqrodxe [eTOTF FO

jo AQPOTAsSeTe eATQeZou we SaQeOTPUT UOTSSeTZeT AT PUL “FIP AT 1032enb Zutsn uoTQenbe sty AJ sA[Nser uoTIeWTAse ey squeseid y eTqQeL

‘uoTzenbe eyQ eQeUITISe 02 SIO 07 UOCTITppe UT pesn

SEM UOTICUTIS® AT ‘eTQeTAIPA snoueZopue ue FTesIzT ST umpuerd eyQ aesneoeq

‘AyTeulg ‘setqetiea Auump [euosees ATAe31enb Jo Yes B soqeirodiz00ut

os[e uoTjenbe pe_eutzse eyL “¢ eTqGeL UT peqrodexz sz [Nser uoTIeUTISE

ay UT peeqsuyT pesn sem -- a/ gpd 3-T)F -- ainsew sty} ‘eansreu

aotad ja0dxe ey roy Axord & se IdM “S’°N 2843 Zutsn sinseou & UPY TEM

SSeT BIep ey ATF ‘seoTad qaodxe AeT [Op euTqQUeZAzy Jo einseeu atTottdxe

ue sesn yoTym ‘(4T) uotaenbe ut umoys ajer1 eZuPyoxe [eTOTFFO peasn[(pe-xeq

‘Teer ey. 2eYQ peuTMJeIep sem AT ‘uoTIeNbe eyQ BuTdoTerep UT

-6€-

-39a-

Table 5: Merchandise Import Equation (Quarterly Data): 1978/IV-1988/II

Dependent Variable: LOG(O1/P™)

OLS (avy? Constant -4.28 -7.73 (-.87) (-1.14) Log (EP_/P) -.48 - 40 (-3.40) (-2.35) Log (EP/E) 04 01 (.31) (.09) Log (Y) 1.07 1.36 (2.33) (2.16) tf .1L5 .02 n (.36) (.05) qr -.13 -.12 " (-1.27) (-1.26) LOG(OI/P*) | 42 46 an (3.96) (3.93) DUMQ2 .13 .12 (2.90) (2.74) DUMQ3 19 .19 (4.19) (4.02) DUMQ4 -.10 .08 (1.90) (1.34) Standard Error of Regression .09 .09 Ry .92 .92 D.W. 2.12 2.20 a . * us P Instruments: Log (EP /P), Log (E (1-tx)P /P), Log (E /E) 4) Log (Y),

* * * : t, tf qr Log (Y ), Log (OI/P) 4 Log (OX/P) 4) i

x -1?

DUMQ2, DUMQ3, DUMQ4, constant.

(t - statistics in parentheses)

“ATOATT ere szzytazeq qaodwt

pue eqei eZueyoxe TeToreuMIod ay 03 sieqaodwt jo sesuodsez oTAzQewdse

pue ‘suotqoesueiq e3ueyoxe ustez0z worz ATeqerzedes paqoeT[oo e10M SJFTAeq qaodwt ‘1eQ10dxe eyQ Aq eB8ueyoxe usTez10J Jo Aepuetzans Jo Autod ey

2e peleqsTuTMpe sem YOTYM ‘xeQ Qaodxe ey exT[UQ ‘UoTIenbe eyQ UT a3e1 aZueyoxe TeToOreumM0d 9yQ worz AToQVeAedes pereqUe ere sJjTreq Aaraoduy /tt

ey. ‘or1ezZ WOIZ YUEeTEeTFTp ATQUBOTJTUSTS pue UZTS peqoedxe ey} Fo are qq) [eer pue eqer eBueyoxe TeTOTemMMIOD [ver 9Yy YIOq UO SjUeTOTFFI09 eyud eTTyM ‘uoTRenbe 41z0dxe eyQ Jo uoTIeUTISe 9YyQ 02 snoZoTeuP ASUUeU

@ UT (T) UuoTIenbe But QeMTAse WoIF sq[Nser ey. squesead ¢ eTqey] sqiodut estTpueyorem [eToTJjo ‘Teer : ( )TO

(v1) (Mab « Mya‘ x * a/ga ‘a/¥aa)t0 - Ka/10 (-) ¢) G @ (-)

zy (7) uoyjenbe se yons dtysuotqzejter e yse’3ns suotjereptsuos esoyL “FOYASW [eTOAeWMMOD ey UT ZupToueuTF oF Yoeq YOITMS 0} SJJTIe9 pToae 02 Jepio ut Aeyxrew TeTTered ey YyZnorzy4R ButToueUTFZ s19qio0duT aoNpuT p[Noo osTe pue ‘jZeyxTeW TeTTered ey uo eZueYyoxe oy [Tes 07 ZuTOTOAUTIeAO AAOdUT UT eSveTOUT Ue 02 peeT pTNoo untuead JoyreW TeTTezed ey. UT eseeroUT ue ‘ATesreAUOD ‘SAeTAAeGq FFTAeQ-uou pue JFzTreQ yIoq UT

SOSPOTOUT 02 SB [eM se ‘eqeI eZueYyoXe TeTOeMMIOD [veT ey? 02 ATeATIesoU

puodsez ptnoys sqiodut perzeqsTZe1 ‘TeZseT Jo eumToa eyL

XepuT sotad azeT[[op AAodut Wy

*

oT3e1 BuTOTOAUT-AeA0 QAoduUT aUM[OA AAoduT esTpueyorem [eZeT TenQoe JT

SIBT[OP UT sqAodwmy estTpuryorem pezeqsTZe1 ATTeTOTJZFO :10

-Qv-

-41-

coefficient on the premium is small and not significantly different from zero. To some degree, this is consistent with the evidence on the lack of import overinvoicing discussed in Section III. It suggests that official sales of foreign exchange were not an important source of supply to the parallel market.

V.3 Pre-export Financing

The pre-export financing facility has represented an important source of short-term capital inflows for Argentina. The services provided by the facility have often been changed during the period of study, and generally have included the provision of subsidized credit by the Central Bank to exporters of specific merchandise categories? In addition, users of this facility have been allowed to borrow foreign exchange from external sources, convert them at the official exchange rate up to 540 days before the time of export, use the local currency proceeds to pay export expenses, and repay the external loan with the proceeds of the exports rather than surrender them to the Central Bank or its agents.

During periods when the domestic interest rate has been high relative to the expected rate of exchange rate depreciation, the preexport financing facility has been an attractive method of legally repatriating capital with the assurance that it can legally be withdrawn again through sale of exports abroad. Of course, any profits earned through domestic financial investment of the pre-export inflows would

have had to be re-converted into dollars through the parallel market.

13/ These subsidies have been estimated at over $5 billion for the 1980- 1989 period, or roughly 0.5 percent of GDP annually. (See Maia, 1990)

Azeututtead Aq pewaxtyuoo sem sTyQ pue ‘eouetezztp eTqetoeidde ue exeu 0} ‘a2eI YSeTEeAUT OFQSeWOp ayQ OF eATQeTEI ‘BuTAATBAUN pur TTeUS 003 ST aqea uStTer0y eyQ ‘eotqoead ut ‘SutoueutTy yrodxe-e1d uo Aoe;zyze eATQeZeU FP qiexe p[Noys yoTym F ‘sqoesse usTerTOJ uo urAN}eT Jo aQe1 9YyQ 02 pereduOD

aq p[noys eaoge peuTzep uznjer Fo ejer oy. ‘ATTPOTIeTOeUL

14,3

q out? 42 waa yo uoTqeqoedxe aa

(punomun QueweZueize JX¥d) 310dxe Jo elep :1+9 poqetatur Zutoueuty qazodxe-aird yoTYyM qe eqep 33

ZutoueuytzZ 310dxe-erd 02 UAnjer Jo eRer :AxXdwd

(ret) (hSa7 9a) Ga/*a) =

(4a) /(E + 17a = axawe

(ST)

9q9e1 YoXTeU

Tetreazed ey Jo uoTAeToeIrdep peqoedxe ey} JemoT ey pue ‘umTwerd JoxreW TeTrerzed eyq AemoT ayQ ‘oeQ"r JQsereRUT OTASeMOp ey zeysty eyq ‘eaTQ0P1NNe azow Zutoueuty 3todxe-e1d putz TT}M AJoJseauT eyR APY JUepTASe ST AT

‘aqear [eTTezed eyq Aq (¢T) uoTIenbe ut uoTsseidxe AsaTjy eyQ ButATdT3arnu pue SuIpTATq ‘3eyx7ewW TeTTered ea ysnorzyq peoiqe peuinjer pur ‘sqtsodoep AnuezAnd-ofAseuop uy peqsaauy ‘aejer esuPyoKe TeTOTemMOD ey 42 euTquesiy oqUT 3YZno0rq sT sqUeMeATNbexr ZuTouPUTZ-Arodxe Ten}oe Jo sseoxe UT ABTTOP eB qeyuq ‘AqtTottduts tof ‘peumsse st 31 ‘quoweSuezie SutToueutTyZ yrzodxe-ead

eB UT ZuTZeBue 09 UANQeA Fo e3eA AeTTOP sey. SeUuTFep (SI) uotjzenby

-Cv-

-43-

regression analysis. Finally, increases in the real commercial exchange rate, by increasing the profitability of future exportation, are likely to lead to increases in the amount of pre-export financing drawn into the country for "legitimate" or commercial reasons. Based on these considerations, and transforming the variables slightly, a reduced form

for the quantity of pre-export financing is derived:

(+) (-) (+) (-) PXF = PXF( EP’/P , EP/E, ai, EP Py (16)

t t41/ Ft

Table 6 presents estimation results for the pre-export financing equation. Because the equation is expressed in logs, the raw quantity of pre-export financing cannot be used, since it varies between being positive and negative over the length of the sample. Therefore, the quantity of pre-export financing is added to merchandise exports and then expressed as a ratio to those exports before being converted to logs. 1+ Three of the four explanatory variables are endogenous -- the parallel premium, the domestic interest rate, and the expected depreciation of the

parallel market rate! -- so IV estimation was used in addition to OLS to

estimate the equation.

14/ Considering that asset demands usually are modelled as demands for stocks of assets, it may appear inappropriate to model the flow of preexport financing. However, much of this financing was self-liquidating, since export receipts were used to cancel the original pre-export loans. In consequence, maintaining a given stock of austral assets required repeated pre-export financing over time, generating a correlation between the flow of pre-export financing and the relative expected rate of return,

13/ To use the depreciation rate most consistent with the 30-day deposit rate, this was calculated as the average of the three led monthly rates of exchange rate depreciation.

‘(2°a) Bot Vy

(sesoyquerzed ut sotTastjeqas - 3) "queqsuoo ‘yoWwnd ‘EedNnd ‘zdWnd

4) 301 v ‘(a) 30L v ‘% (x0/(axa+xo)) 307

‘T’ (xo/(axa+x0)) Bor tor 8p er Ty 8 Cg ag) B07

‘O° (a/ya) B01 ‘' (a/ga) Bor ‘(8 4a) Bor (2° 4a) B07

ra at e¢L7 x ‘ x . ('gh) Bot SC’ (d/ gd C 9-1) 3) BOT ‘(ad/ gd 9-1) a) SOT :squeumajsul,

Lo"

oc”

v1"

(€8° 90°

(22° 9T°

(15° 70°

(S9° 09°

(IL° oT"

(18° T°

(68° 6T°

(S8 19

é

“T-) “Te

CAT)

(X0/ (IXd+X0) ) D0T

II/8861-AI/8/61

LO? ce"

yT°

(@2°T-) 80°-

(sS*Z-) LT’ -

(9L°-) So°-

(OL*T-) 9S" -

(99°T) 86°

(12°Z-) 8e°-

(80°2) 6T°

(Z0°2-) 9S°T-

S10

:(eqeq AT1091eNdD) uofjJenby supoueuyy Jazodxy-sxzqg 79 3s [qeL

“MA’d

Cy

uoTsseiZey jo AOIIG przepueys

yoONnd eonwna

zownd

(ga/Tia) 307 jo ae8ereay $preazend

T o2ey 321Tsodeq (a/ga) 301

(a/ pnd 3-1 a) 301

quejsuoy

:eTqeTieA Juepusdaeg

-44-

The estimations indicate that variations in the parallel market premium had a substantial and significant (at the 5 percent level using OLS, at the 10 percent level using 1V) effect upon the volume of preexport financing. The interpretation of the coefficient on the parallel market premium, using either equation, is that an increase of 1 percentage point in the parallel market premium generated a reduction in pre-export financing equal to roughly .4 percent of merchandise exports in that quarter. Given that the premium frequently changed by as much as 20 percent from quarter to quarter, its movements most likely contributed

to the substantial volatility of pre-export financing.

ayy Jo soueurozized azood ay 03 peanqTazquoo ‘sexeq qaodxe atTotTtTdxe yqIM

peutTquoo ‘xeq AToTTdwT sty, ‘sexeq uoTqadumsuoo to swoouT peseq-peord

qoaeT[oo 02 QuewureAcZ eyA Jo ssousuTTTTAun ey 02 enp ‘szeqaodxe uodn

xeQ ATOTTdwT ue pesodwy weqshs eel Tenp eu] ‘awTZel [TorquUOD eZuryoxe aya Aztasnf€ x Tesat Aq qou seop ‘zeAemoy ‘ITFoeueq TeOSTF STUL

‘potaed eyq 3utanp

dq) Jo queozed oT pue queored ¢ useemjeq ZurZuel AToTjJep TeosTF e YITA

perzeduoo ‘ATTenuue gq9 jo queorzed Z/T T noe oF yenbe squewked eotares

Aqep (Aouezano TeooT) [euCTITppe JueeM eAPY PTNOM sToOrIUOS asueyoxe

Jo eouesqe oy} UsyL ‘eqeI JoXATeM TET TeIed ZuT{TTeAead eyq og Tenbe

qnoqe 20 ‘TeAeT TeNQoe sayz ueYyI peqeToerdep ezow queosed o¢ useq eAey

pInom ejex eSueyoxe TeToTemmod ey. ‘STOTQUOD agueyoxe AnoyAIM ey sumsse

‘qTJoueq TROSTF 9yQ JO epnqtusew oeyQ eqQeTNoTeo OL “SOsET 4 gutanp

aeok zed gqy Jo queorzed ¢ AsAo 09 pe junoUe quewuzeaocS euTquesazy ey2

Jo suoT{eST[qo eoTAres Aqep [eure RxXe eTeyM ‘euTquesiy 03 set{dde quoum3ie

STUL ‘SUOTQeZTTGo USTeTOZ SAT Fo 3sSod AouaerInNoD oTAsewop sy sonpe1t

TITM TeaeT eqetadoidde sqqT moTeq e317 aSueyoxs ey BSutsseiddns ‘seornos

anusaez AouerIno usTe10J pesoxe ey. suoT IesTTqQo AouerINd-usTer0J

sey QuewureAcS eyQ JT :QuewUAeACS oyI 07 JTFoueq TPOSTF B ASF sqoyreu

aZueyoxe Tenp 2eyd (8s6T ‘O2UTd) penzie ueeq sey Al SqTyoueq [eosty

-goetndod eyq Aq eoueT{dwoo [eZeT 100d pue quewuzeac3 ey Aq

Zutyewhot Tod tood yQoq peZeinoous pue “TenjoeFJout ATeBIeT e7eMm ‘sqQTFoueq

Mey peTeFjZo sToAQuoo ey ‘ezTreuUMS OL “SO86T PU Zutinp eutTquesay ut

‘pezepuesue AoyQ eSueyoxe ustTor0jy AoZ JoOYTeU TeTTezed eyQ pue ‘sTozquoo aZueyoxs Jo sqyyTyoueq pur szsoo ay} a jeNnTeAe seSed BuTMOTTOF sul

Buljuesay UPL sporquoD esuEyoxy Fo UoTjeniTeaAq uv 1A

-Cv7-

-46-

export sector in Argentina during the past decade, and also helped postpone the government's fundamental need to increase explicit tax collections. Protection of firms with external debt One of the chief concerns of the economic team at the beginning of the exchange control period was the effect of devaluation on private sector firms with high external liabilities. These firms may have been aided to some extent by the combination of lower devaluation rates coupled with exchange controls, but not by much. Notwithstanding the exchange controls, substantial real devaluation occurred during 1981 and 1982. Moreover, the government eventually assumed the private sector's foreign currency exposure anyway, first through the exchange rate guarantee programs of 1981 and 1982, and second, when the government could not finance up-coming maturities under the program, by the takeover of the private sector's guaranteed external debt at the end of 1982. Reduction of capital flight There are two basic rationales for reducing capital flight: first, to protect international reserves, and second, to prevent a leakage of capital abroad that otherwise would have been invested productively within the country. As noted in Section III, most measures of capital flight show a substantial decline in capital flight after 1982, when the government stopped selling dollars without restriction to the private sector. In this sense, the former objective was certainly achieved; if the government had maintained the identical exchange rate policy without exchange controls, reserves would have disappeared.

However, there is no evidence that exchange controls helped to

maintain private investment levels. During 1981, when capital flight was

“@ATIOSFJFO ussq Qou sey ‘AAT[TqeTA squowked-jo-souelTeq uTeQUTeU 03 STOITQUOD edueyoxe Aq peqizoddns ‘xzoyoue TeuTwou e se oer eBueyoxe ay} jo asn 94} ‘Wi9eq Zuo, pue umTpew eyq ut ‘esoueq ‘TeATds AXeUOCTIeTJUT ey Fo

STemeuerl pure ‘werzgoid ,i2oyoue TeuTWOU, e243 Jo squeWUOpUeqe ‘aje1 aeSueyoxe TeToreumos ey} FO suoTjen[eAsep quenbesqns of pet ‘uinq uT ‘szXoqQoRz osouy “s~TTejsqazoys squewked-jo-sourteq pue ‘umqweid qeyzew [eT tered ey Jo SuTUepTA TenqueAe ue ‘eqer esueYyoKS TeTOTeWMMIOD [Teer ay Jo uoT eToOeAdde ue ut peqtnserz queuqjsnf{pe TeostTzy Arequowetduos Jo yoeyT oyQ pue eTjAeUT AzeuoTze[JuT Buoizqs Jo soussead oyQ ‘AeAemoy ‘seseo oe1yI [Te UT ‘uoT{eTJUT Fo o2e1l Oy BuToNper ut [nJsseoons oqtnb erzem ATTeIITUT [Te pue ‘sezeerzy ootTad pue oje1 eSueyoxe Jo uoTReuTquOD e puNore ATING [Te e19M 6861 AINE UT (Wee. OTUOUODS ASATJ S,WoeUeW QUepTsezg AoJ Teuuosrzed Aey ey peptaoiad yotym ‘uzog A e8ung Jo wATjF ouTQUeSsay ay Aeaqze PeTTR2-OS) gg UeTd O49 pue ‘gg6l ASNEny uT eAOAeUTAg UPTg eYy2 ‘SEG eUNe UT TeAIsny UeTd PUL ‘“UOTIeTJUT Jo QuewuTeqUOD 9YyQ sem ‘sTorQUOD eZuPYyoxe 4q peqaoddns ‘exe ay. APeTO Jou pTp 3eYy e4eI eZuUeYOxe Ue Jo BuTjIeS eyi A0F eTeucTIer Ay Oy ‘SOBST CYR BuTing UWOTIeTFUT Fo UoTjonpey ‘QUeUTQSeAUT OTJSeUIOp AOF sSeATAUSOUT ey. ester jou pip anq ‘yueg TeAQueD eyA Aq AUSTTF Teatdeo Jo Buyoueutys 9y pepue sTorjqUOD eSuPyoxe Jo uoTjatsodut seul ‘erTdwes AatToey. ut setTazqunoo Zey.O ey. JO Ysow uT AOU ‘euTAUeZIY UT AYSTTJ Teatdeos pue juewqseaut ueeM{eq UOTIFTS1IOD OU PUTF OSTe (88s) eUTAST pur UOpIOD ‘seTAjUNOD SutdoTeaep UT 3UuSTTZ Teatdeo Jo Apnqs ateyy uy ‘gqqd Jo jueorzed / 09 peddoip pey ean3Tz styq ‘(smoTjuT [eatdeo etem er0yQ ‘st 2ey9) eat eZou BUTeq se peqeWTjse sT AYUZTTJF Teatdeo useym ‘gg6l UT ‘gaqy Jo jueorzed

G°Il SW“ QuewqseAUuT pexTy eqeatad ‘ainseeu Aue Aq poanseow se y3ty

-{[t-

-48-

V1.2 Exchange Rate Unsustainability

Fundamentally, the rationale for exchange controls was to permit the authorities to set the commercial exchange rate at a level that would have been unsustainable without controls in order to stabilize the economy. However, as pointed out above, in the absence of fiscal balance, exchange controls could provide only temporary support for misaligned exchange rates. Eventually, increases in the parallel market premium would undermine the balance of payments and reduce the level of foreign reserves.

As noted in Section IV, the government faced with this situation had two options. First, it could tighten monetary policy to raise the domestic interest rate, appreciate the parallel market rate and reduce the premium, as it did in late 1988 and late 1989. However, in both cases this led to increased costs of servicing the government's domestic debt, a widening of the fiscal deficit, and subsequently, higher rates of monetary emission and parallel exchange rate depreciation than would have been the case without the initial monetary contraction. Unsuccessful attempts to use monetary policy to reduce the premium consistently led the government to exercise its second option, a devaluation of the commercial exchange rate. Hence, exchange controls served only to delay official devaluations, not to postpone them indefinitely.

The parallel market’s influence over the commercial exchange rate is supported by the results of a simple regression shown in Table 7. The authorities are assumed to devalue the commercial exchange rate to achieve a particular target level of the real exchange rate; hence, a high level of the real exchange rate should lead to reduced exchange rate

depreciation, while high domestic price growth should increase the rate

(sesoyquerzed ut sotastqeqs - 3)

6L°T 022 "Ad 9° 0s" oa cL’ 90° uoTSSaIZ0y JO AOAAA paepueyAs (Z2°€) (9€°2) I-. 1- 8S" a (" a/qg 3) 3oT (62°S) (61° L) l-.1- S38" 68° (gnd/° dd) SOT V (20° €) (S0'€) tee 9L°- TZ" - (© d/gyd A) 307 (10° €) (70° €) £8°9 68°T queqsuop AU TSR AeNH ATWauoHy

(A)80T VO :eTqetszea Auepuedeq gg6r Aine - 7861 AP WOTISUTUISEIEg e7eY SsdUetoxW [eporommoy -7 OTGeL

-49-

of devaluation. Additionally, a high parallel market premium is hypothesized to induce the authorities to step up the rate of commercial devaluation in order to reduce that gap. (Lagged values of these explanatory variables are used to avoid simultaneity problems.) The estimation results support all three hypotheses. They suggest that the parallel market premium, which averaged 32 percent during the exchange control period, induced an additional 3 to 6 percent of commercial exchange rate depreciation per month, depending upon whether estimates

from the monthly or quarterly regressions are used.

VI.3 Costs of the Exchange Control Regime

Distortion of economic policy Exchange controls were not merely

ineffectual. By allowing the commercial exchange rate to be set independently of economic realities, albeit only temporarily, exchange controls encouraged policymakers with short time horizons to adopt "nominal anchor" strategies -- such as fixed exchange rates -- at the cost of neglecting more difficult fiscal adjustments. In this sense, exchange controls were part of a larger package of instruments, which included price and interest-rate controls, that allowed the authorities the illusion of targeting nominal variables without addressing the fundamental causes of inflation and balance-of-payments problems. Because such stabilizations were inherently unsustainable without complementary fiscal measures, periods of disinflation inevitably gave way to periods of accelerating inflation, accounting for the "stopgo" pattern of inflation observed in Argentina during the 1980s (see Chart 3). Exchange controls also permited the government to maintain an exchange rate long after it had become overvalued, thereby adding to the

needed magnitude of the devaluation when it finally was implemented.

“SMeT Xe oy YITM ATduoo 03 AuewjTuMOD s,oTTqnd ey. euTWAepuN 02 ATOATL ST osTe sTorquUo. eZueyoxe Jo Juewso10jue xeT &q porepueSue souet{Tduos preMoq epnqtjje pexeter ey ‘reAoeI0W ‘ADUeZe UOTIOeTTOO xe oy Worz Wey epTY OF eATeS OSTe YUe_ [eAQUeD eYQ pUe suo jsND ay. wor SuoTJoOeSUeTZ TBUI9IXe epTYy 02 pedoTeaep seianpeooad ay ‘ZeAeMOH =‘ VOUPTeEq TeOSsTF PASTYOR 02 PTq sqy uy eSueTTeyo utew s,eurqueSazy squeseaderz UoTSeAS XB} JO UoTIONper ey. ‘ATQUeAIND ‘svete AeYyIO OWUT AeAO

TItds 03 ATOATI ST eJTT oTWoUCDe Jo verze ouUO uy eourtTdwoo-uoy ‘Auouods punos3zepun ayy jo uoTsueqxe atofem e Joz snqedwt eyQ peqeero SeTotTtod eZueyoxe s,eutquesiy ‘aouenbesuod ul ‘eulTZex1 eSueyoxe ey peuTwzepun reYyAAnNJZ ‘uotsseazddns TeZeT eATOeFJe Jo souesqe oy. uT ‘yoTym Jeyzeu TeTTeazed e Jo uoTqeeI9 eYyi 07 BuTpeeT ‘uoTseae ATey rz0F SeATIUSOUT Bu0IAS pe.eeID sTorqUOD esueyoxe Jo uoT3Tsoduy ayy ‘uousuioueyd sty, jo eTdwexe quejiodwt ue st JeyxeW eZueyoxe Te TTeazed s,eutquessy (6861 ‘Taesstny ees) ‘suotje[nZe2r esey2 YITM souRTTdwoo-uou Jo aerZep YsTYy e 02 peT sey ‘queweor0sue zood AzeA YIATM petTdnoo ‘azt{T oTwWou0[s JO seaeie soul BuTqQoOezZzJe suotjze[ner JO uoTIeAeFZTTOAd e ‘euTQUesAzy UT soueT [duos TeseyT fo woTjeaoTazojog ‘quewureAcd ey. Jo AAT IqTpero ey pepore ATsnoTies pue ‘jQuewyseauT peonper ‘AQTATIOe@ OTWouCDe peqdnastp YoOTYA ‘UOTIBTJUT UT sUOTIeIeTeD0R Jo ssoudzeys ey. eJenQUseD0e 03 pepueq SUOTIENTeASP-TXeW QUeITNsel sy, ‘uoTsstTwe AreAoUOW TeUOTITppe quenbesqns pue ‘3TOTjJep [eosTF AeBzeT & ‘sqsoo vsoTaAres Agep OTIsouop AeYySTY 09 peT eseyi eouTs ‘umtwerd qeyzew Te TTered eyQ sonper 03 seqer 4ser0qUT ester

02 sqdweq3e [nzJsseoonsun TeTaTuT Aq peorosuter sem uaqjo AoezZFO stTyL

-0OS-

-51-

REFERENCES

Baldwin, Richard, "Hysteresis Bands and Market Efficiency Tests," unpublished draft, June 1990.

Banco Central de la Republica Argentina (BRCA), Memoria Annual, Buenos Aires, various issues.

Bhagwati, Jagdish N., Anne Krueger, and C. Wibulswasdi, "Capital Flight from LDC’s: A Statistical Analysis," in Bhagwati, ed. ,Illegal Transactions in International Trade, North-Holland, Amsterdam, 1974.

Carta Economica, Estudio M.A.M. Broda y Asoc., Buenos Aires, various issues.

Calvo, Guillermo A., "Fractured Liberalism: Argentina under Martinez de

Hoz", Economic Development and Cultural Change, Vol. 34, No. 3, April 1986, pp. 511-533.

Connolly, Michael D., Alvaro Rodriguez, and William G. Tyler, "The Use of the Exchange Rate for Stabilization Purposes: An Arbitrage Model Applied to Argentina," unpublished draft, May 1991.

Cumby, Robert and Richard Levich, "On the Definition and Magnitude of Recent Capital Flight," in Donald R. Lessard and John Williamson, eds., Capital Flight and World Debt, Institute for International Economics, Washington D.C., 1987.

di Tella, Guido and Rudiger Dornbusch, eds., The Political Economy of Argentina, 1946-83, University of Pittsburgh Press, Pittsburgh PA, 1989.

Dornbusch, Rudiger, Daniel Dantas, Clarice Pechman, Roberto Rocha, and Demetri Simoes, "The Black Market for Dollars in Brazil," Quarterly Journal of Economics, Vol. 98, February 1983.

, and Juan Carlos de Pablo, "Debt and Macroeconomic Instability in Argentina," in Jeffrey Sachs, ed., Developing Country Debt and Economic Performance, Vol. 2, University of Chicago Press, Chicago, 1990.

Edwards, Sebastian, Real Exchange Rates, Devaluation, and Adjustment, MIT Press, Cambridge, Massachusetts, 1990.

Fundacion de Investigaciones Latinoamericanas (FIEL), El Control de Cambios en la Argentina, Ediciones Manantial, Buenos Aires, 1989.

Gordon, David B. and Ross Levine, "The Capital Flight ‘Problem’ ," International Finance Discussion Papers, No. 320, Board of Governors of the Federal Reserve System, Washington D.C., April 1988.

Guissarri, Adrian, La Argentina Informal, Emece Editores, Buenos Aires, Argentina, 1989.

‘0661 ‘'O°q ‘uoQZuTYyseM ‘Gqmory pue APTTTqeas sottg AO SWACfoy .eUTQUOsAY ‘YUPG P[IOM sul

"6861 Azenaigey ‘serTy soueng ‘TqI_ ‘3geAIp peysttqntdun ‘,etqeperseseq BoTIeWATAY ep seW Od0g un KA ‘soUOTO.eTA sey ‘ourzeA Tq, ‘

“8861 AequieAoN ‘SeaTy soueng ‘TgqIa ‘3yeAp peysttqndun ‘,elTqepesaZesaq BOTIeUATAyY ep odog un A soUOTO.eTY se] ‘eAeAeUTIg BT, ‘enbtzuq ‘yormezs

“Git-7se “dd ‘6861 ‘€8-976T eUTjUesay JO AWlOUOOY TeOTAT [Og eUuL ‘*Spe ‘Yyosnqui0g

pue eTT°9L TP UT ‘18-9461 ‘AOTTOd OTWoucdg ouTQUeZIy, ‘*y ArzeT ‘peaseels

‘6/61 AIne ‘seaty soueng ‘3jzeAp peystiqndun ‘eutjquesszy ep soTWMOUdD.z0Z0RPR soTpnjasy ep orquep ‘,gpepT[eey o OATW :eUTQUesIy ue ofzeTqueD oselay Td, ‘peaseels Axe] pue Operz{y soTze9 ‘zen3tTspoy

“1661 YtePW *ZE9 SeTtes zedeg Sutyi0m Ydad ‘FOPT PTAOM ouL ‘,euTQUeZAY Jo esey oUL ‘3TOTJEQG 104995 OTTQNg e42 Jo soTWMOUODeOIOeW eUuL, ‘OpeTs[y so[ze9 ‘zen3BtTzpoy

"8861 tequeqdes ‘7g seTieg rzedeg Suty10m Ydd ‘AUP PTAOM SUL au ‘BOTAJY UPTBYRS-qnsg UT WAOJoeY TenperzD snsAevA AQUSTUIeEAQ :UOTIeTJUL pue ‘soqey oSueyoxg [Rey ‘eSueyoxy ustexroq AoF sqoyeW Yoe[g, ‘uetazag ‘oqutg

"0661 tequejades ‘*y°q uoqQSuTyseM ‘weqsks earTeSey [eAepeg eyQ Fo SIOUIT®ACDH JO preOg ‘/ge ‘ON ‘SaodegZ Wotssnostq soueurq [Leuotjeudtejuy ‘,AITTTqeag pue eainjonajs :ysnorzyl-sseg BupQeUt sg, ‘°Y WeITLTIM ‘AOTTOW

"0661 3sn3ny ‘euTquesay eottTqndey eT ep [eAQUeD ODUegG ‘QJeAp peustTtTqndun ‘,0861 ep epeoed BI ep uUoToOeUTISY seuTQUaesAY OT[TGndey eT ep [eAQUeD ODUeYG [Te ered souoToORqIOdxy

ep UoTOOWOAg ep SOASTOUPUTY SoUeUTZey SOT ep soqsoy, ‘sIn] esor ‘ete

"6861 YOAPW ‘Z9T SeTIeg tedeg SuTyI0M Add ‘AUST PTAOM SUL ‘,ueta Terqasny eyQ soUTS Oo} pue doqs :euTQUeZAY UT uoTIeTJUI, ‘Ten3qtyn ‘pTen3ty

"8861 ‘€ "ON ‘Z ‘TOA ‘MOTADY STWOUOSyT yueg PIIOM SUL ‘,POTTeuy UTIeT WOAF suOSseT :seTOT[og uoTIeZTTTqQeIaS xopoyuzI9 pue soTaTprI3ty ATeUOTIAeTJUL, ‘UPQRETATT UPSSTN pue Toen3TW ‘Ten3tTy

"6861 APW ‘'O'"d uoqZutTysem ‘weqasks sAresoy TeLepey eyQ Fo sAouATeaAcCy Jo preog ‘7ZGE “ON ‘saodeg UOTSSNosTqg ooUeUTY [TeUOTIeUTeqQUT ,‘YyQmomzy AAQUNOD BuTdoToeveq

pue 4qeq [ePUuIeQXY, ‘sUuTASeT ssoy pue ‘uyRy Areqoy ‘

"8861 20q02090 ‘°9°q UuCQZUTYSeM ‘weqQsds eALTISey [eLepeq oyQ Jo sAoursao0y Jo przecg ‘Hee ‘ON ‘Saodeg WoTssnostq DoueUTA TeUOTJeUAeQU] ,‘SO8TIQUNOD ButTdoTevseq uT e8ueyoxyg usTer0g

TOF SROYARHY AOPL[G pue ‘sjTorzquop eBZueyoxyg ‘uotTjenypeasq, ‘usereqg ‘utTuey

"2861 ‘'O'd UoaBuTYyseEM ‘TEST SUOTIOTAISOY SsUEYORY pue Sjuowesuelaly osueyoxd uo qz10dSy LTenuuy ‘puny Azejeuopy TeuoTtzeuszsquy

-CS-

-53-

Appendix A : The Equilibrium Parallel Exchange Rate

This appendix provides additional detail on the model of the equilibrium parallel market rate discussed in Section IV.1. Equation (A.1) reproduces the steady-state equilibrium condition shown on page 27 in the text.

dF=S -D=0 (A.1)

S: flow supply of dollars into parallel market

D: flow demand for dollars in parallel market

F: stock of gross foreign assets (in dollars) owned by private residents

The supply of dollars into the parallel market is assumed to derive from underinvoiced export receipts, as well as dollars purchased in the commercial market but sold in the parallel market:

* * S = oXP,, + 5IP, (A.2) X: volume of merchandise exports ¢: under-invoicing ratio I: volume of legal merchandise imports 6: financed overinvoicing as a fraction of legal

merchandise imports financed in the commercial market

P..: dollar export price

P

<< ie

dollar import price

Total merchandise export volumes depend upon the average of the parallel and commercial exchange rates, weighted by the fraction of receipts sold in each market, as well as export taxes, relative prices, and the level of total output:

(+) (-) G) ©) X= X(gEP + (1-¢)E(1-ty) , P, PY ,Y~) (A.3) ty: export tax rate

P : domestic price level Y : real GDP

The underinvoicing ratio is presumed to depend only upon the parallel market premium and the export tax rate:

(+) $ — $(EP/ E(1-ty)) (A.4) The quantity of legal (not overinvoiced) imports financed in the commercial exchange market is presumed to depend upon the official

exchange rate, relative prices, the level of economic activity, the import tariff rate, and the number of prohibited import categories:

ey. JO uoTIeNTeAep Y ‘sSnon3tque s} osTe eer YoexIeEW TeTTered eyQ uo oe3ei a8ueyoxe [eToOAemOD syQ Jo uoTAenTeAsp eB Fo AoedUT sy

‘AATOTAsSeTS Atun ueyA sseT Jo ere sqiodut Az0j spueuep

qeyQ sumsse OSTe eM JT eat ATsod A[TsnonZtqueun soeuioosq seotad jzodut

ABT [Op uo USTs ay. pue ‘eATaTNQUT ST se ‘aaTatTsod ATsnon8tTqueun souoseq

seotid ofTQsouwop uo UBTSs 9YQ ‘usu, ‘“ButToToAuTstu jzodxe 02 eaTAeTeA eToOr

[jews AjTeatqeqTer e sXketd Butotoautstwu qazodut jeyq sumsse om JT peatoser eq ued (g°VY) uotQenbe ut qQueptae setorutureqepuytT 9YyQ Jo suUloS

(8°¥) (Mab «Nga 6 Mg Mat ta a ' a adn = gd (+) () (2) () (&) ¢-) C) C)

!U9Q3TIM oq UPD SqUeUTUAEQEp , [eQUOMepunj, JO os BuTMOTTOF ayQ 02 e3er QoyTeM TET [ered umTAaqt{tnbe syQ BuTyAUTT WIOFJ peonper [ersues ATA & ‘SUOTIeAepTSUOD seaOoge ey} UO peseg

sqiodut peatqtyord (Aa Aq perxeTsep) [eer AOJF uoTAounj puewep : ( )a

(L°¥) Ma( Mab‘ x‘ Mata‘ ga a-a (+4) G) (-) G) (¢)

*sqiodut 09 sizeTzAAeq JFTAeQ-uouU Jo AUseRxe

aya pue ‘AQTATQOe OTWOUODe Fo TeAeT 9YyQ ‘saoTad saTQeTeA ‘eQer JoOYyTeUW

TelTezed eyQ uo spuedep ay ‘peatqtyord esTmzeyjOo Ato AexTeM eZueYyoxe

[eToOTemuod oyj UT eTqetToueUTy Jou (WSTInoqI BuypNToUT) seTATpommos jo sjiodwy worz puewep peATAep & ST Q SABT [OP A0F puewep MOTF YT,

‘MoTeq peqtzosep ‘sqiodut peatqtyoad eouruTz 03 pepesu jeyq eaoqe q size[ [op 10; puewep MOT} 94Q BuTSeeTDUT Sse Jo AYZnoYQ eq Aeu Ay ‘ATeaTQeUZEQTe ‘“axP? sadteoe1 jrodxe peotoauTAepun TeI03 MoTeq ‘ZeyTeW TeTTerzed ey 03 ste[top jo Atddns 750 e se peqeadzequy eq pInoys yoTtym ‘§ sonper p[nom sty] ‘AexTeW TeTTered oy ut peseyoind oq 02 APY PTNOM sjiodwWT peodtoauTAepun ey eoUeUTF 0F SABT[OP ‘SUFIFOAUTISAG ueyQ 1eyjeI BUPoTOAUTASpun Jrzodwmy But eaTIom ‘eaTQeZeu sulodeq

p[Inoo umtquezd eyQ ‘eqez FFtzeq YSTY ATJUeTOTFInS e Z0OF QeYI 220N

(9"¥) (Mga41) 9/49 9 = 9 (+)

:eqel JFTTeQ eyQ FO eaTSNTOUT ‘umTMerd JoyreM [ENP ey. YITA AyTeatqtsod ZuTAIeA SB peTepoul st ereYys BuToOToOAUTIeAO-jAodmT eyQ ‘ATT eUTA

sieTiizeq qiodwt JyTzeqQ-uou Jo TeaeT : Wib

aqer JFTIeQ AaodutT :; W5 sjiodwt [eSeT [eet AOJF uoTjounj puewsep : ( )/)I

(s"¥) ( Wap © Wyq + x Ma ‘a‘@d)rIe=1

(-) (-) G) ¢-) G ¢)

-G-

-55-

official exchange rate should lower export underinvoicing and reduce supplies to the parallel market, forcing up the parallel market rate. However, this effect can be reversed if the devaluation induces a sufficiently large export volume response. Most evidence (see Kamin, 1988, Edwards, 1989) supports the former story, and so a positive effect of the official rate on the parallel market rate is assumed in the text.

Similar reasoning applies to the effect of export taxes. Increases in export taxes should divert receipts to the parallal market, thus appreciating the parallel market rate, but also should reduce total exports, thus depreciating the rate. Assuming, as above, that the export volume response is weaker than the invoicing response, we assume a negative impact of export taxes on the parallel rate.

Finally, the impact of real output on the parallel exchange rate also is ambiguous. Increases in output associated with increased aggregate demand should raise import demands and depreciate the parallel market rate; conversely, increases linked to higher productivity or favorable weather conditions may be associated with higher exports and an appreciated exchange rate. As disentangling these effects is beyond the scope of this paper, we retain the ambiguity of this variable’s effect.

Equation (A.9), reproduced as equation (3) in the text, incorporates the assumptions described above.

(+) (+) (-) (+) (?) (-) (+) (+) EPr = "E"( EYP, Py , Py ,¥, ty, thy » aty) (A.9)

APPENDIX TABLE B.1 : EXCHANGE RATES AND TERMS OF TRADE QUARTERLY AVERAGES REAL EXCHANGE

NOMINAL EXCHANGE RATES (1) RATES-MAY 1981=100 DOLLAR PRICES-1970=100 (2) PARALLEL TERMS OF COMMERCIAL PARALLEL PREMIUM COMMERCIAL PARALLEL EXPORT IMPORT TRADE 1977 I 03 .03 2.18 222.98 227 .86 192.50 194.70 98.87 II 04 .04 62 223.13 224.54 201.10 206.40 97.43 Ut .04 .04 “1.15 206.91 204.51 197.50 212.60 92.90 IV .05 .05 -.07 196.16 196.03 181.70 212.30 85.59 1978 I .07 .07 -47 187.05 187.92 191.10 215.50 88.68 II .08 .08 -.99 173.99 172.27 194.10 207.60 93.50 III .08 . 08 ~1.18 154.55 152.71 201.00 223.50 89.93 IV .09 .09 -.05 140. 36 140.27 207.80 214.90 96.70 1979 I .11 -1l -1.05 127.47 126.13 233.00 219.70 106.05 II .12 12 -.91 121.58 120.47 240.90 243.40 98.97 III 14 .14 -.90 110.54 109.55 260.40 282.50 92.18 IV 16 -16 2.28 106.79 109.23 288.30 257.80 111.83 1980 I -17 17 -.07 102.67 102.60 311.10 253.00 122.96 II -18 .18 -.00 94.50 94.50 286.10 264.50 108.17 III -19 19 .00 89.46 89.46 286.40 267.20 107.19 IV 20 .20 .00 80.93 80.93 329.40 245.80 134.01 1981 I 22 22 .00 81.37 81.37 354.50 247.50 143.23 II 36 . . 38 6.08 109.20 116.78 310.50 245.60 126.43 III .51 75 45.50 123.06 179.04 294.20 248.50 118.39 IV 68 1.02 50.12 132.71 198.68 295.30 259.20 113.93 1982 I 1.03 1.01 “1.73 159.69 156.93 301.30 250.40 120.33 II 1.37 2.13 54.39 185.15 287.31 254.90 254.90 100.00 III 2.45 4.91 101.55 234.31 471.97 242.70 238.20 101.89 IV 3.93 5.94 52.91 255.96 388.99 231.00 237.60 97.22 1983 I 5.85 7.76 32.80 264.86 351.67 238.60 230.00 103.74 II 7.83 9.71 24.44 259.86 323.59 229.20 232.30 98.67 III 10.76 17.13 57.36 237.78 372.62 225.50 223.50 100.89 IV 17.82 24.66 42.27 235.50 333.92 239.00 228.50 104.60 1984 I 27.84 40.45 43.91 239.22 342.51 250.30 226.40 110.56 II 41.18 62.93 53.36 213.96 328.22 254.20 215.90 117.74 III 69.32 95.66 37.96 206.60 285.06 245.70 223.20 110.08 IV 133.21 156.64 17.83 225.88 266.26 227.40 226.60 100.35 1985 I 249.98 320.06 27.23 233.17 296.66 217.70 220.20 98 .86 II 552.83 648 .00 19.71 248.27 295.88 212.60 229.80 92.52 III 801.00 944.77 17.95 262.65 309.80 216.50 236.10 91.70 IV 801.00 892.60 11.44 246.71 275.10 237.00 239.30 99.04 1986 I 801.00 889.66 11.07 223.29 247.93 198.80 236.50 84.06 II 850.34 905.68 6.59 205.49 219.11 191.00 242.10 78.89 III 973.05 1074.55 10.06 193.78 213.24 188.70 241.00 78.30 IV 1152.45 1370.92 18.62 192.00 227.79 200.50 241.70 82.95 1987 I 1405.66 1768.54 26.11 195.64 246.60 186.20 246.90 75.42 II 1612.66 2060.90 27.99 194.79 249.33 180.00 270.60 66.52 iT 2155.56 2920.80 34.90 197.20 266.07 193.60 281.20 68.85

3429.43 4198.76 22.41 218.57 267.41 186.70 272.60 68.49

-55b-

Td) JNILNIONV/ CI dd

69°SL 02° 862 68°Tl 06°OTS 80°92 00°90 29° 9L Of °90¢ 06°82 09° 262 92°99 09° OTS SL2°9L OL° 682 62°92 02° 622 daval LYOdWI dO SWa3L

02° 0s”

G22 £22 “2E? “be? “o£?

LYOdx4

OOT=026T-S39T¥d YV1100

“S°N * JLVY JONVHOXA) (2) 4YV110d Yad SIVYLSNVY JO SHLAGNVSNOHL CT)

99° bl? T2'O0e¢ 8b° 82 92° 24S 88° 282 T6'O0Te 8L°' 6b? LL° 892 86°062

1411VaVd

OOT=T86T

02°Sle2 T8°692 LU Tle 86° 12b SS 9ST L6°<LT G2' 902 £6°T2e2 gl°st2

Te" LT 6T G8°b £L°6¢ 66°98 Te'T2 66°02 90°Te2 88° oe

WIDYSWWOD WNIWSYd

(2) AVW-S3ALVY

JONVHOXS 1V359

14711

Vad

JOOBB0BBRUE Pees ees fs 3 JOOGB0G000E 36 30 30 3 ED IEE EE 99° 99bLE 98° 6SEST £L°2eSeT L2' Tees

1311VaVd

JIE 303033 IE JO 00G0E JOG 0OOK £o°SBooT £8°51L92T 22° cdtett oT 1989 T2°£S¢b

CT) S31LVa JONVHOX]

IWNIWON

- Le 92°6S S6°0L Sl 85°22 06°LS SL°- 7s" 02 L6°S9 £y 68°ecl OL°?L SZ yL°6e L8° ol 6¢ c2°' So°02 62 90°CL L0°St cL 6L°Zl eZ’ Ol £¢ 6S*Ll 46°72 Ll S6°6 £Ll°l 02 os’sl s8°9. L9 coe of°S 8y¥ ly°9 6£°OL Ol 48°8 88° DL>- St°tl £l°9 yL°- 64° ye L0°L- yL°? ses Ly 80°¢e- 48°6 Lo°- 96°S 9£°CE £0 02°S2 86°Z1L Ol’- s9°0¢ SL°6L £0 o9°Zl YL°SL 02° - el st Obl £0 ZS- Ll ¥6°9L 9S s9°92 Le°7e7 20 S9°4 ee"? ve OL 8? se°?l Le 65°S bo Le £0°- ef’ ol Ly"? Zt yc Ol Sb°S 00 9S9°SZ 40-2 vl £S°9¢ 99°S 97 60°- 59°6 00 BL°cl 6L°8 OL s9°sl SZ°S £s 0S" 2s 9S" eLol L£9°S £8°s OL col L8°s LiL 69° 1 cee Ly 20°72 S6°S ss°l £L£°c 49°OL SO°L £9°¢ co's eel £8°c Sb OL 66 cs 99°8 Ov L 82°¥ L0°6 LL Le°s Lo’s oY £6°2 DL°LL S6° 9S°S ss°8 £6° 8s°9 S2°Ll cs" ec°s ¥o°6 LL” Le°9 elo £>° ss°¢<

Z8° “I'd'd aLva cH “s’n AONVHOXSA TAT Ivavd INZONSd NI

FONVHO dO SALW ATHINOW :

2@°a FIAVL XIGNaAddV -9G¢-

dLva SONVHOXG IWIOYINNOD

H-naonek

ee |

eer mM KKK NRK NN

Q ° . . . . e . 2 DANHDODDKROTONTNNDMORRINDOEMRERRORARTOMAMTRIOR NRK KOK RRENR RO

YINNSNOD

066L

686l

88é6l

L86lL

986L

S86l

v86L

csél

c8é6l

L86L

Osél

6Z6L

BZ6L

LZ6L

1377

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

GROWTH OF REAL GDP FROM YEAR EARLIER

— NWA

~-5. -6.

“9. -5.

—~—ANOMm—

NON-]—] 1 wn

to ND WOhIN—]hoNAHROX

(n—

CAPACITY

UTILIZATION RATE

-55d- APPENDIX TABLE B.3 : RATE OF ECONOMIC ACTIVITY

CIN PERCENT)

z0'SE ¥0°9ZL Lo" LS 99°%

98°L

9S" 12 S0°ZL 10°61 SS°9

BLEl 92°01

Le°c7

tZ°S8l cy cl

alva TIId AUNSVIUL “s7n

aa.Lsnrdv- NOILVIOWNdIG

CLINZDN3d NI) (SISV@ ATHLNOW) NUNLIY JO SALVU * ¥°A TIGVL XIOGNAddV

LS°62 L9° £7? AI Lo° 9? £o°sl III ee°OLl 00°66 Il Zs’°s8l ZS°ZL I 49°0L Zs"Ol AI £9°¥l Oc vl III oOL’sl L£9o° ZL II £9°¥l co°vt I Oo cl o9°Ll AI £9o°el £2°6 III 04°8 £l°s II eo°'£ 2B I 09°8 ee°S AI Lo°9 Lo IIt es" ZL°e II 0c°S oOL’e I 0Ss’S OL’ AI Lo’ oSs"f III Oc be £o°L<? II Looe 0s sl I L0°L72 oo'-Zl AI £9°61L os’slL III Ofc sl 00°<l II £o°el 0s "Ol I Off? OS‘ yl AI £9°0¢ OL'eL III Z8°sl £9°6 II £o°sl OL’ OL I L9° 72% 66°2 AI £0°6L 69°S ITI S6°91L Sik II ses 80°L I yS°8 60°2 AI Z9o° Ll cB 6 III SZ°6 ss’8 II £e°8 6L4°9 I 69°S 6L°¥ AI £e°9 cL’S ITI OL°S BL Il OLS 9¢°S I Le°“£ £y°9 AI ¥6°L cool ITI BSL°Z £s°9 il Oc L 0s°9 I SS°L cL’9 AI a £9°9 III II I AI III II I ALVU dLvu NVOT aaud LSIAAINI LIS0d3ada

066L

686L

886L

L86L

986L

S86L

bB86L

esé6l

csé6l

L8é6l

OgélL

6Z6L

8Z6L

LL6l

-55f- APPENDIX TABLE B.5 : BALANCE-OF-PAYMENTS COMPONENTS

($ MILLIONS)

MERCHANDISE MERCHANDISE CURRENT PRE-EXPORT IMPORT EXPORTS IMPORTS ACCOUNT FINANCING FINANCING 1977 SI II III IV 1978 I 1330.00 858.00 267.00 317.00 89.40 II 1808.00 871.00 652 .00 193.30 -101.80 III 1966.00 1046.00 739.00 -484 .20 94.20 IV 1294.00 1059.00 32.00 -73.40 -24.80 1979 I 1549.00 1199.00 -68.00 332.80 110.50 II 2383.00 1380.00 456.00 368.60 77.40 III 2161.00 1913.00 -~76.00 -120.20 355.80 IV 1717.00 2220.00 -898.00 160.80 201.70 1980 I 2059.00 2282.00 -767.00 143.20 347.30 II 1924.00 2286.00 -1003.00 58.50 35.50 TII 2036.00 2782.00 -1151.00 37.50 318.00 IV 2002.00 3190.00 ~1870.00 -358.90 471.10 1981 I 1990.00 2614.00 -2080.00 -124.40 -176.30 II 2848 .00 2622.00 -909.00 -429.50 -5.40 III 2719.00 2195.00 ~212.00 -830.90 -435.80 IV 1586.00 1999.00 -1491.00 -246.50 -175.10 1982 I 2170.00 1484.00 -309.00 216.00 -159.50 II 2346.00 1333.00 -316.00 -619.40 323.20 III 1624.00 1217.00 -624.00 -64.80 -465.00 IV 1483.00 1303.00 -1143.00 -74.50 395.90 1983 I 1934.00 977.00 -607.00 225.80 -338.40 II 2107.00 1185.00 -485.00 402.20 81.90 III 2003.00 1210.00 -792.00 -98.10 -216.10 IV 1793.00 1133.00 -592.00 -266.20 812.30 1984 I 2159.00 865.00 ~ =-200.00 6.80 ~56.80 Il 2449.00 1099.00 -136.00 182.40 -117.00 III 2081.00 1326.00 -750.00 -91.60 120.50 IV 1418.00 1294.00 -1307.00 579.70 209.30 1985 I 1803.00 977.00 -778.00 -31.70 -79.50 II 2570.00 927.00 185.00 401.60 -300.20 III 2310.00 979.00 109.00 ~254.40 153.60 IV 1713.00 931.00 -471.00 42.60 85.40 1986 I 1513.00 921.00 -803.00 53.60 24.20 II 1968.00 1154.00 -455.00 255.00 147.50 III 1897.00 1356.00 -511.00 -442.10 232.10 IV 1474.00 1293.00 -1092.00 76.40 62.70 1987 I 1441.00 1200.00 -1018.00 539.10 -27.40 II 1741.00 1379.00 -795.00 -281.20 135.60 III 1618.00 1633.00 -1096.00 -456.00 351.30 IV 1560.00 1608.00 -1321.00 309.10 33.10 1988 I 1713.00 1165.00 -851.00 -11.80 -371.20 II 2304.00 1407.00 -427.00 294.30 -33.10 III 2673.00 1458.00 2.00 468.60 256.00 Iv 2444.00 1294.00 -339.00 276.10 -18.10 1989 I 2117.00 1168.00 -822.00 5.90 -202.70 II 2546.00 1034.00 -125.00 -845.40 162.20 tt 2870.00 1090.00 203.00 -20.90 61.10

1990 I

Appendix Table B.6: Measures of Mis-invoicing and Capital Flipht

($ millions)

Measures of Capital Flight:

Federal

Im ort Overinvoicin 1 World Bank? Reserve Board? Cuddington”

1977 630 -287 940 -618

1978 703 -1,055 1,852 1,497

1979 626 -2,917 3,128 -1,693

1980 400 -2,097 5 ,036 2,301

1981 661 -876 5,751 9,800 8,680

1982 852 -617 8,455 6,100 5,210 4 1983 942 -1,568 2,615 1,700 1,955 a 1984 1,028 -1,160 -2,617 -400 -1,635

1985 1,097 -526 500

1986 65 , -568 -1,200

l/ Source: FIEL, 1988. 2/ Source: Cumby & Levich, 1987. 3/ Source: Kamin, Kahn, and Levine, 1989.

1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987

1988

Appendix Table B.7: Measures of the Public Sector Deficit

Cavallo-Pena”

11.8 13.6 10.5 13.0 16.7 16.4 19.6

Balance of non-financial public sector.

Reproduced in di Tella and Dornbusch, 1989.

Rodriguez, 1991.

World Bank, 1990.

-55h-

(% of GDP)

: 3 Rodriguez

13.3 15.1 15.2

11.9

World

13. 15. 16.

12.

Bank

1

Sseqejseq Tala :90anos ‘sTseq squoudeg-jo-aouelTeg ‘sqiodxe esTpueyorew APT [Op TeTOTFFO

aseqeqeq Tdld :2903no0s “sTseq squowdeg-jJo-sourjTeg ‘sqiodut estTpueyorew ABT [Op [eTOTFFO

0661 eUNe ‘BUTIUSSIY op SOoTWUOUCDSOAOER SSAOPeOTpUTL ‘yd :eoAnos ‘sooTad qyaodxe Ae[[op jo xepuly

0661 eUNe ‘BUTQUSSAY op SOSTMOUCDSOTOeR SeAOpPOTpUuL ‘TVdaD :e01NO0S ‘seotad qazodut Aze[[op jo xepuy

SOTJSTIJeqN9 [@loueuTy [TeuoTjJeutojU, :90A1n0sg ‘yquow ou. JoF seSereae ere YOTYM ‘sTeaeT ATUQUOU

Fo e8erz0ae ATAeQAeNb - Axoird se poesn xepuUy eOoTIg AeONporAg ‘s'n :

aseqeaeq dia :eo0anosg ‘yquow oy TOF seZereae ore YOTYM ‘STeAeT

ATyQuow jo e8ezeae ATAeqzeNb - xXepuy eoTAgG AeUMSUOD sUTQUeZIy :

SOTASTIEIS TeToueuty pTeuotTjeurequy :eoAznos soqei ATYQUOU Fo e8er2Ae Ajaeqazenb - sT[tg Aanseezl ‘S'n Uo (stseq ATYQUOU) eQe1 QSeTEQUT

BotTuoUudoY eqaey {suotqeottTqnd qwI pue yUueg P[AOM SnoTAeA :e0zNES *soqer A[Ttep Jo sosdereae o27e YOTYM ‘STeAeT

ATyquow jo e8ezeae ATAeqAenb - sueoT peyoeq XANOG A0OJ O2"A WATS

-lequy stenbe ‘spotied [ozquod eqer AsezZequT Buting ‘jueuwuzeaocs Aq peT{Torzquos jou (stseq ATYQUOU) suUBOT UO e4e1 QYSeTeQUT

BoTMoOuoog Sjaeg {suotjzeortqnd yWwI pue yueg P[zA0M snoTzAeA :a07N0S *soqez A[Tep Jo seBeroae

are yoTyM ‘sTeaeT ATYQuou Jo e8eteae ATAeRIeNb - syUeq ouTQUeZAy qe sqtsodep owt Aep-o¢ uo (stseq ATYQUOW) e2e1T JsezTeQUT

(2)

(1)

aseqejeqd Tala :eoanos ‘soqea A[Tep Jo so8exzoae ore yoTym ‘seqex ATYQuOU jo o8erzeae ATAeqAZenb - APT [Op ey} ASUTeSe eet eBueyoxse [eToremmog

sseqejeq Tdla :90an0S ‘soqez A[Tep jo sedezoae ore yoTym ‘seqez ATYQUOW jo e’ezeae ATA0RIZenb - AeT Top ey. AsuTeSe aqerI YeYyAeU TET [ered

s}TsA[euy UuoTsseiZey uy pesg eyeq :9 x}pueddy

-9C-

> XO

-I0

PXF:

qq"

-57-

Pre-export financing inflows in dollars. Calculated as CX-OX, where CX is merchandise exports, Exchange-Balance (Balance Cambiaria) basis (i.e., actual receipts by exporters, including pre-export financing).

Source: FIEL Database

Percentage of all tariff code positions subject to quantitative restrictions. Source: World Bank staff

average tax rate on exports. Source: World Bank staff

average tariff rate on imports. Source: World Bank staff

Real Gross Domestic Product of Argentina - quarterly. Source: FIEL Database

Real Gross Domestic Product of the OECD countries - quarterly. Source: QECD

“TSSOZ «6° O'"d ‘u0j3utTysemM ‘ueqsks eALVSOY [elsepsy oy FO STOUTAOCD Fo pAreog ‘Hz doqjs ‘goueuty Teuotjzeurzsequy Fo uotstatq ‘sqedeg UoTSsnosTd soueUuTY [PUOCTQeUAeqUT 03 setdoos 1t0zy sisenber sseippe eseatg

eee

Jedee] “W otag

enbtataig y wear8ut yjeg uoTjenjTeaq AoT[Og oTAQeuoUCDY qsog C6E O66L SeoTAtes TeToUrUTY UT epeiry Jo Aroay] AZewo3juoW “gq uyor Be preMolL :2661 pue uoTReQUeUZeS JoyAeY H6E uUOSSUSAS ‘QO “Yq Sie SWH 943 SuTanp oqey e8ueyoxg Wd/d4 esoy ‘y Merpuy eYUL ‘squowUsTTeeYy peqotperg pue peqoedxy S6E Sjiodwl “s$*n Fo r0Taeyeg aya Jo stskTeuy Teotatdwuyq uy :sotaqeuouccg jo AQTOTASeTY ZenbireW outer eyA AO SeTITOTISeTY Jo soTAqewmouccg ayy 96€ AoTAeyeg sZuTaesg Jo eToy au :A4T[ITqeas uTARBy TOeYyoTW pue ‘sourteg eperl ey ‘epeazy jo swI9], L6E¢ sewou, ‘qd seTzeyy a2eT TOG ea epee “Y uel Tq aseoor0g 09 AATTTqeuTeqsns Teureqxg SuIsg 86¢E AQUTeQe0uQ eje8e138y II uwewetop uyor anqTIm YaTM seouetTeg Asuow AzteuoTAnede1rg 66€ sqjueulertnbey ~Teqatdes Ayeelt “dq Teeyotw yueg pue ysTy eSueyoxy uSter0oy SuTutuszeqeg 007 sueaeqs “9'aA And AeTAeIS “| stToyT eed Jueuljseaul Joe1TG@ “s'p Jo Aoenbepy ayy lov SeTaqunop Zutdoteasg ut suer801g uoTAeTJUISTGC uTwey ‘gq ueraeqs Jo 3a0ddng ut setny oejey aSueyoxy ZOv uopr0y “gq ptaegd aedeey "HW oTaAg qoeFFa AITPInbTT eyq Jo yorveg uz €0v SeTpeuley peqIses3ng pue seoanos :SJuNoOooY suOTIOPSURAL [eUCTIeUIeAUT A9T4PIS “A sToyT "S'Q 043 ut Aouederostq TeoTasTqeag ey vOV ATBAI 09 uotAeotTTddy TeoTatdug ue YIM Tepow wy :AoTTog BTeI9 ures “y Testy pue syTetquerezjsIgd eqey 4serequl swyq SOv uell yuy-3u0y sodwey ettne ApolTopoujey uossoTayq “Yy TION OFAJeuloucoy s,Aipusy pTAeq pue AAID-od 90% O66T-T86T :siexTeW eBueyoxy ulwey ‘gq uaaeqs TeTTered Y2IM souetiredxy s,eurquesiy LOv T66T — CS) aOHINV SHILIL aH aWON dda

szedeq WOTSSnosTg SoueUTY, TeuoT eure UT

- 8S -

IFDP NUMBER

392

390

389

388

387

386

385

383

382

381

380

378

377

376

- 59 -

International Finance Discussion Papers

TITLES 1990

Mercantilism as Strategic Trade Policy: The Anglo-Dutch Rivalry for the East India Trade

Free Trade at Risk? An Historical Perspective

Why Has Trade Grown Faster Than Income?

Pricing to Market in International Trade: Evidence from Panel Data on Automobiles and Total Merchandise

Is the EMS the Perfect Fix? An Empirical Exploration of Exchange Rate Target Zones

Estimating Pass-through: Structure and Stability

International Capital Mobility: Evidence from Long-Term Currency Swaps

Is National Treatment Still Viable? U.s. Policy in Theory and Practice

Three-Factor General Equilibrium Models: A Dual, Geometric Approach

Modeling the Demand for Narrow Money in the United Kingdom and the United States

The Term Structure of Interest Rates in the Onshore Markets of the United States, Germany, and Japan

Financial Structure and Economic Development

Foreign Currency Operations: An Annotated Bibliography

The Global Economic Implications of German Unification

Computers and the Trade Deficit: The Case of the Falling Prices

Evaluating the Predictive Performance of Trade-Account Models

Towards the Next Generation of Newly Industrializing Economies: The Roles for

Macroeconomic Policy and the Manufacturing Sector

AUTHOR(s

Douglas A. Irwin

Douglas A. Irwin

Andrew K. Rose Joseph E. Gagnon Michael M. Knetter

Robert P. Flood Andrew K. Rose Donald J. Mathieson

William R. Melick Helen Popper Sydney J. Key Douglas A. Irwin David F. Hendry

Neil R. Ericsson

Helen Popper

Ross Levine

Hali J. Edison Dale W. Henderson

Lewis S. Alexander Joseph E. Gagnon

Ellen E. Meade Jaime Marquez Neil R. Ericsson

Catherine L. Mann