Hyperinflation and Stabilization in Nicaragua

. .

* I

February, 1992

HYPERINFLATION AND STABILIZATION IN NICARAGUA

Josi! Antonio Ocampo l /

I. INTRODUCTIONThe Nicaraguan economy faced in the 1980s and early 1990s

massive macroeconomic disequilibria. Economic activity neverrecovered the large losses incurred during the 1979 revolutionwhich brought the Sandinistas into power. Moreover, GDP per capitalfell steadily from 1983 to 1991. As a result of production lossesand rapid population growth, by the late 1980s GDP per capita hadreturned to levels comparable to the 1940s. Throughout this process

of economic collapse, private consumption per-capita and real wages

fell even more?L

This ,' process was accompaniedc

disequilibria. As a result of these, I..accumulated by the early 1990s a foreign

by massive externalimbalances, the. countrydebt close to $10 billion

(including interest arrears). more than seven times GDP, the worstdebt ratio in a heavily indebted region. Finally, the collapse ofreal economic activity has been accompanied by equally massivedomestic financial disequilibria, which exploded intohyperinflation in 1988 and 1990. From January 1988 to January 1989,and from March 1990 to March 1991, when this process was at its

peak, inflation reached 43,000 and 64,000% (equivalent to monthlyrates of 66 and 71%), the records so far in Latin America and someof the highest rates in'world history.

Macroeconomic management faced a complex set of constraints,quite different to those confronted by other Latin American

'/ Senior Researcher, FEDESARROLLO, Bogoti, Colombia. Thispaper is based on the author's experience as member and head of theSIDA and WIDER-SIDA Hissions to Nicaragua from 1989 to 1991.

2countries in the 1980s. Through the 198Os, Nicaragua continued toreceive massive financing from abroad. Also, according to ECLACestimates, the terms of trade did not fared badly, either '/.However, these favorable events were overwhelmed by the impacts onproduction and resource availability of the revolution and thecontra war, the US trade embargo and veto on multilateral lending,excessive reliance on relatively inflexible bilateral assistancefrom the former socialist countries, and a series of naturaldisasters.

,The building up of macroeconomic disequilibria was alsoclosely associated to economic policy. In the first years of therevolution, the government adopted an expansionary publicexpenditure program, to improve the poor social record inheritedfrom the Somoza years and accelerate economic growth. These goals,

mI, particularly the latter, were sacrificed when the government was

iforced to increase defense expenditure to face the contra war. Upto 1988, the central government ran massive budget ,deficits.Monetary financing of the deficit, together with equally massivesubsidies on the use of foreign exchange and credit resulted, with

a lag, in hyperinflation.

The magnitude of existing disequilibria forced the governmentto adopt more ambitious adjustment programs in 1988 and 1989. Inthe former year, the program emphasized the correction of relative.price distortions, particularly the simplification of theinefficient and costly multiple exchange rate system. In 1989,continuing efforts to 'correct exchange rate overvaluation werecombined with a contractionary fiscal policy.

Adjustment efforts were completely abandoned during the firstmonths of 1990. The new Chamorro administration inherited again

_' '/ This is not true according to alternative estimates byBulmer-Thomas (1987), Table A.14.

.i

Kiara CaleroResaltado




.-

3

substantial macroeconomic disequilibria. Moreover, as opposed tothe pattern typical during the Sandinista period, it faced strongtrade union resistance. This factor, combined with increasingdollar indexation, resulted, once more, in hyperinflation. InMarch, 1991, a new stabilization program was put in place, withstrong backing from the major bilateral donors (particularly theU.S.) and multilateral agencies. This program was able to build onprevious stabilization efforts and to stop inflation, though at thecost of significant overvaluation of the cordoba.

Iah,ifls,,paper,,,. L analyzes m,acroeconomic policies and performance:, I,,, .I ,II #..I . .i :: .:,',.5-..~ .',", ., ,INicaraqu&:in. the 19.80s and. thg,onset .of the.l,,9:9$s.w.n*'$t3&,&... w &I, ~~?~.~~~rY.,t:.d~,:i,~.j~?~~,~~~~~~:~-.~e ,.-l,t+ ,, It is divided_ ,( :. \9 .:a.,.::* ..I', :, \'i'.;... 7

inin

nine sections, the first of which is this introduction. The secondsummarizes some features of the Nicaraguan economy prior to therevolution. The third considers the effects of revolution and the

.c period of recovery which followed it. The fourth analyzes thebuilding up of macroeconomic disequilibria during the transition to

'Y and full fledged war economy. The fifth shows the characteristicsof the 1988 adjustment program and hyperinflation. The sixth takesa close look at the 1989 stabilization program. The seventhconsiders the 1990 hyperinflation and its relation to the politicaltransition from the Sandinista to the Chamorro administration. Theeighth summarizes the major features of the March 1991 program andthe post-stabilization period. Finally, the ninth presents themajor characteristics of the structural reforms underway.

II. THE NICARAGUAN ECONOMY PRIOR TO THE REVOLUTION

The recent study by Bulmer-Thomas (1987) indicates that therewas little growth in GDP per capita in Nicaragua from the 1920s to

the late 1940s (see Figure 1). This period of relative stagnationwas followed, however, by an export-led boom from the 1950s to justbefore the revolution. GDP per-capita multiplied by 2.5 during thisperiod. As this process was matched by rapid population growth, GDPexpanded at an average rate of some 6% a year, the fastest inCentral America. The rapid growth of cotton exports was the initial

I.

4

basis for expansion. Later on, the process was reinforced by new

primary exports (beef, sugar, shellfish, etc.) and a boom ofagroindustrial and other manufacturing exports to members of the

Central American Common Market, CACM (Bulmer-Thomas 1987, CEPAL1981, Gibson 1987a).

Rapid economic expansion was not translated into an equallyrapid improvement of social indicators. At the end of the boom,illiteracy, child mortality and life expectancy levels were among

the worst in Latin America --comparable, however, to other CentralAmerican countries, excluding Costa Rica and Panama '/. Income

distribution remained highly skewed, at levels also similar to theCentral American neighbors (Brundenius 1987, Table 2). There islittle evidence on how distribution evolved during the period ofexpansion. However, available data on labor incomes indicate that

-c real wages were basically trendless in the 1960s and 1970s '/. Asthis was accompanied by widespread and growing informality in the

'.I labor market (Gibson 1987a, Table 2), it may indicate that income

distribution deteriorated in the last phases of the boom. On theother hand, the concentration of wealth in hands of the Somoza

family and his political clique was remarkable, as the data onnationalizations following the revolution later revealed.

Economic management was fairly orthodox throughout the boom.From the late 1950s to just before the revolution, the exchange

rate was pegged at a rate of 7 cordobas per US dollar. Since 1963,the currency was fully convertible. Orthodox fiscal and monetary

policies guaranteed low-inflation rates but also the transmissionof external shocks to the domestic economy. As a reflection of

'/ See CEPAL (1988a), pp. 13, 45 and 50 and footnote 5 below.

'/ Using the average wage estimated by INSSBI, and the GDPdeflator as a price index, real wages (1981=100) increased slightlyfrom 1960-1964 to 1965-1969 (from 100.2 to 107.0) but thenstagnated and declined (106.0 in 1970-1974 and 103.2 and 1975-1979).




56I.

limited fiscal and current account deficits, foreign indebtnessremained within close bounds (Gibson 1987a and 1987b).

The economy and economic management experienced, however,increasing hardships in the 1970s (CEPAL 1981). Reconstructionefforts after the 1972 earthquake broke the tradition of fiscalconservatism. In the last years of the Somoza regime, budget

deficits increased to an average of over 5% of GDP (see Table 2below). This was also refelected in increasing foreign indebtness.According to ECLAC estimates, the external public sector debt

quadrupled from 1972 to 1979 (from $230 to $961 million). The

counterpart of this process was persistent current account

disequilibria, enhanced by the adverse effects of the 1973 oil

shock, the slowing down of growth of trade within the CACM, growingovervaluation of the cordoba and capital flight in the months

-L before the victory of the Sandinistas (see below). To face growing

disequilibria, the Somoza government established mild exchange

\: controls in late 1978. In April, 1979, it devalued the basic

exchange rate to 10 cdrdobas per dollar and introduced a multiplerate system.

III. REVOLUTION AND RECOVERY (1979-1981)

The economic legacy of the last years of the Somoza regime and

the revolutionary uprising was complex. Economic activity severely

contracted in 1978 and 1979, by an accumulated 32% (Table 1). The

capital stock was also severely affected. Losses associated to thedestruction of buildings, equipment and stocks, looting of

inventories, slaughter of immature beef cattle and smuggling of

herds were estimated by ECLAC at $381 million (CEPAL 1981),equivalent to 18% of 1980 GDP. National Accounts records indicate

that the loss of inventories in 1978-1979 was equivalent to 14.4%of GDP (see Table 1). To these, we must add capital flight for $535

million in the 18 months preceding the revolution (CEPAL 1981),portfolio losses by industrial and commercial firms and, of course,

.' the casualties inflicted by the war.





6The revolutionary government brought with it some emergencymeasures, a plan for economic recovery but, above all, an agendafor structural change. The latter was presented as a program for a"mixed economy", in which the State would assume control of theproperties of the Somoza family and his clique and some lrkeyrreconomic sectors, and considerably expand social expenditure andits contribution to capital accumulation. The State would alsoencourage the organization of the popular classes, throughunionization in urban areas and cooperativization in thecountryside. As a result of the enhanced role of the public sector,new rules of the game for the private sector would be designed.

One of the first decrees issued by the government aftermilitary victory on July 19, 1979, was the nationalization of theproperties of the Somoza family and his allies who fled the

-L country. It was followed by the nationalization of the financial

system, foreign trade, large scale (particularly gold) mining,forestry and fishing. Few other important nationalizations tookplace in the following years, but the government periodicallyexercised the right to confiscate the properties of capitalistssuspected of counter-revolutionary activities or practices whichled to the decapitalization of their businesses (Stahler-Sholk &al. 1989). Government's share in GDP rose from 15% to slightlyovers 40% in the early 198Os, but then stabilized. The privatesector retained a dominant share of agriculture, manufacturing,domestic commerce and most services (World Bank 1981, Baumeisterand Neira 1986, Ruccio 1987, Brundenius 1987). In 1990, at the

onset of a new phase of structural change, government's share inGDP stood at 47%, but this figure is not fully comparable to

earlier estimates '/.

'/ CORNAP (1991) estimates that the 350 public sectorenterprises, excluding utilities, contributed to 31% of GDP in1990. On the other hand, 15.7% of GDP was made up of governmentservices, financial activities and utilities, all fully controlled

,:, by the government in that year.




7The initial nationalization decrees also brought some 20% ofland property under state control. Land redistribution acceleratedas a result of the Agrarian Reform Decree issued at the secondanniversary of the revolution.,As a result of both measures, morethat 50% of rural property was affected in the years following therevolution. During its first phases, the government emphasized thedevelopment of parastatals and cooperatives, but soon evolved intoencouraging small scale farming. The redirection of agrarian policywas largely induced by the need to erode peasant support for the

Contras in some regions of the country. Nonetheless, it alsoreflected the social programs of the revolution and the policy ofself-sufficiency in food staples (Enriquez and Spalding 1987, Neira1988, Wheelock 1989).

The nationalizations created a large parastatal sector. As in. most countriesL undergoing similar processes, the management

problems generated by such a sudden expansion of the state sector

\ were costly (Colborn 1990). On the other hand, the redesign of newrules of the game for the private sector proved difficult and, infact led, rather early in the process, to violent confrontations(Vilas 1987). At a purely economic level, the private sectorresented excessive state intervention in their businesses andgovernment predilection for public-sector enterprises. Moreimportantly, however, the exclusion of the bourgeoisie frompolitical power and the practice of intermittent political

confiscations generated a general sense of insecurity of propertyrights.

The inadequate functioning of state enterprises andconfrontations with the private sector may explain the failure ofeconomic activity to recover rapidly in the years following therevolution. A partial recovery was, nonetheless, experienced, basedon an expansionary demand policy and an ample supply of externalfinancing (Fitzgerald 1989). By 1981, central government

L' expenditure, as a share of GDP, had doubled with respect to levels




8,c typical before the revolution. The initial fiscal expansion

included many social programs, --which induced a rapid improvementin key social indicators "/--, but also defense and generalbureaucratic expenditures. A large part of this expansion wasfinanced by rising taxes. The resulting deficit, of some 9% of GDP,was, nonetheless, reasonable in the short run, given the amplesupply of external financing (Table 2).

In fact, other domestic macroeconomic indicators were notparticularly troublesome. As a result of the disruption of thedomestic distribution network during the last stage of therevolutionary uprising, inflation peaked at 70% in 1979. Assupplies stabilized, this price surge was followed by moderateinflation in the early 1980s --some 20% a year (see Table 1).Domestic liquidity ratios increased with respect to those typical

. \ before the revolution "/, but were stable (Table 2). Finally,nominal wages increased, but there was no attempt to raise them in

., real terms (see Table 1 and note 3). This required, in fact, asignificant political effort by the Sandinistas to control labor

demands (Vilas 1987). The policy strategy adopted by the governmentthus implied that workers will receive increasing real income

through government services --a llsocial wage", a it was called--butwould contribute, through wage restraint, to the recovery ofeconomic activity.

"/ Life expectancy at birth increased from 56.3 years in 1975-1980 to 62.3 years in 1985-1990, as child mortality fell from 9.3to 6.2%. At the same time, the illiteracy rate fell from 42.5% in1970 (and a similar figure just before the revolution) to 13.0% in1985. See CEPAL (1988a), pp. 13, 45 and 50.

"/ Estimated on the basis of end-of-year monetary aggregates,the ratio of M,/GDP increased from 13.1% in 1974-1978 to 22.6% in1980, whereas MJGDP increased from 20.7 to 30.5% (see IMF,International Financial Statistics). The methodology used in Table

- 2 puts such liquidity indicators at 20.9 and 33.0% in 1980.



9The core external sector indicators moved, however, in thewrong direction. Neither traditional nor non-traditional exportsever reached pre-revolutionary levels (Table 3). The reduction of

exports was combined in the early years by a deterioration of theterms of trade. On the other hand, the revolutionary government

inherited a clearly overvalued cordoba, and a rate of inflationclearly incompatible with a fixed exchange rate. There was noattempt to correct such imbalances. A steady real appreciation of

the cordoba then ensued. It was accompanied by a strong

depreciation of the black market rate (Figure 2). The politicalclimate generated by growing confrontations between the Sandinistasand the private sector accentuated this trend.

Although strong import and exchange controls became a centralfeature of external sector management during the first years of therevolutionary government, the former were not particularly harsh.

Indeed, the import coefficient reached a historical peak in 1980and 1981 (Table 3). Growing external imbalances generated by large

imports and weakening exports were financed by record' capital

inflows. Thus, as outstanding debts were renegotiated, the country

had ample access to new financing. Resources came from multilateral

agencies, bilateral sources, both in the developed countries

(including the US) and the Third World (Mexico, in particular), andonly secondarily from socialist countries (see Stahler-Sholk 1987,Arana et al. 1987 and Table 3). The result of this strategy was, of

course, the rapid growth of the external debt. By 1981, the debthad already reached extremely critical levels (Table 3).

10

IV. WAR ECONOMY AND MACROECONOMIC DISEQUILIBRIA (1982-1987)A. General features of macroeconomic manauement '/The expansionary demand policy adopted during the first years

of the revolutionary government could be defended on the groundsthat the access to external financing should be used to ensure afast turnaround of economic activity and an equally rapidimprovement in key social indicators. On the other hand, as we haveseen, the macroeconomic package typical of the first years revealedsome prudence on behalf of the government, as reflected in its wageand tax policies. Nonetheless, by itself, external disequilibriawould have called for a significant policy shift as early as 1981.

The government did not grasp the urgent need for action.Indeed, the systematic lag in the adoption of the stabilizationpolicies and the partial nature of such efforts once they were

adopted became central features of Sandinista macroeconomicmanagement early in the post-revolutionary period. Expectedly, thegovernment was unwilling to give up what it thought to be theessential goals of the revolution, or to adopt policies which itthought would affect the economic recovery and, even more, riskmilitary defeat. Nonetheless, the political process worked in

peculiar directions. Understandably, defense and social expenditure'became the most inflexible components of the budget. Paradoxically,however, the government was at the end more willing to sacrificereal wages and capital accumulation than to reduce the massive

subsidies to the productive sector. Its strong political control ofthe labor movement and public-sector enterprises and, on the

contrary, its feeble relations with the private sector and the needto guarantee the support of the peasants in the contra war, go along way to explain this paradox.

'/ For a more extensive analysis of this period, see Arana &al. (19871, Fitzgerald (1989), Gibson (1987b), IMF (1988), Medal(1988), Pizarro (1987), Taylor et al. (1989) and World Bank (1986).Stahler-Sholk et al. (1989) presents also a very useful chronology,which would be extensively used below.



11

Although the first signs of government concern for the balanceof payments --the adoption of export-promotion policies--came asearly as 1982, the expansionary of expenditure policies were infull swing up to 1984. By then, domestic disequilibria had reachedclearly explosive levels. Forced by the circumstances, thegovernment adopted the first important stabilization measures in1985, including cuts in non-defense expenditure, adjustment ofgovernment-regulated prices and devaluation. This was followed bysimilar steps in the subsequent years. However, the inconsistency

of the stabilization packages implemented from 1985 to 1987enhanced macroeconomic disequilibria. Particularly, risinginflation eroded the tax base, and attempts to repress inflationand defend exporters against official exchange rate overvaluationled to massive relative price distortions and booming blackmarkets. As a consequence of these imbalances, the government wasfinally forced to adopt more drastic stabilization measures in 1988and 1989.

. .

On top of the dynamics generated by expansionary expenditure

policies and inconsistent macroeconomic management, therevolutionary government also had to face during this period thedestabilizing impact of the contra war and the US anti-Sandinistacampaign. The war had significant demand effects, as it forced afurther expansion of defense expenditure. However, it also hadimportant supply effects (Fitzgerald 1987, Gibson 1987b). Asidefrom the destruction of resources and production, it createdmultiple labor shortages, associated with the diversion of young

workers into military service, rural-urban migration, scarcity oflabor in some crucial (particularly coffee-producing) regions andthe flight of skilled workers abroad. On the other hand, the 1985US trade embargo forced an inefficient substitution of tradingpartners. Finally, the suspension of direct US aid soon after

Reagan was inaugurated in 1981 and the American veto onmultilateral lending in the following years, forced the country to




12

rely increasingly on inflexible bilateral assistance from socialistcountries (Table 3 and Stahler-Shock 1987).

B. Fiscal and nonetarv disecnailibria and the f&&. . .Stabr&.zation efforts

As a reflection of policy decisions and defense needs, centralgovernment expenditure continued to increase rapidly after 1981,peaking at 58.7% in 1984 '/ l /. As Table 2 indicates, the mostdynamic element from 1982 to 1984 was the expenditure ininfrastructure and production (largely investment outlays).However, all components of central government expenditure continuedto increase at rapid rates. Efforts to raise government revenueswere successful, and by 1984 the country had one of the highest taxrates of Latin America and the Third World. Nonetheless, the growthof expenditure clearly outpaced the tax effort. In the same year,the central government deficit reached 23.5% of GDP --26.6% for theconsolidated public sector deficit, according to a partialestimate using IMF data lo/.

Growing pressures generated bymacroeconorPic disequilibria andthe contra war led the government to undertake significant

*/ As pointed out in note 2 of Table 2, total expenditureaccording to central government accounts does not coincide withdata on destination of expenditure by ministries, which is used tomake up the breakdown shown in the second part of the same Table.The former figures are used in the text when referring to totalexpenditure.

'/ Total expenditure was actually higher, as not all militaryexpenditure financed bythe socialist countries was budgeted. IMP(1991) estimates such expenditure at 13.4% of GDP in 1990, but thefigure for earlier years is unknown. Since 1991, al militaryexpenditure is budgeted.

lo/ We have excluded from this figure both unpaid foreigninterest and deficit estimates for the nrest of the public sector".The former are unlikely to be ever paid. The latter have beenestimated by the IMF on the basis of domestic lending, which is apoor approach in a highly inflationary economy. The estimates ofcentral bank losses may also subject to controversy.

13

expenditures cuts starting in 1985. However, the war forced afurther increase in defense expenditure, which peaked over 18% ofGDP in 1986-1988 'I/. Thus, the government was forced toconcentrate cuts in civil expenditure. From 1984 to 1987,expenditure in infrastructure and production fell to very modestlevels and foreign interest payments were all.but suspended, as theexpansion of public administration costs earlier in the decade wasreversed. Expenditure in social services was maintained, however,at historically peak levels.

Overall, central government expenditure was reduced from 58.7%

to 44.1% of GDP from 1984 to 1987. Non-interest civil expenditurefell even more, by some 18% of GDP, but remained slightly above1980-1981 levels (Table 2). Unfortunately, most of the expenditure

cuts were defeated by the adverse Olivera-Tanzi effect on- government revenues "/. Thus, the central government deficit

remained at 16.5% of GDP in 1987. As we will see shortly, othermajor components of the public sector deficit, particularly CentralBank losses, were even more inflexible. Thus, the overall publicsector deficit never fell below 20% of GDP, even if unpaidinterests on the external debt and the deficit of several public

sector enterprises are excluded.

The monetary impact of deficit financing was dramatic.However, up to 1984, the economy absorbed it through an impressive

c

-/ Or higher, as pointed out in footnote 9.

lz/ This was the dominant element in the erosion of taxrevenues in 1984-1987 and through 1988. Given a month's lag in thecollection of tax and other current incomes (a lag which seems tohave been reached by the end of this period), the 1984 share ofcurrent government income in GDP would have fallen to 29.1% in 1987and 22.4% in 1988 as a result of faster inflation. Thus, additionaleffects on government income, such as domestic recession, had asecondary role in the erosion of the tax base. They may beimportant, however, is the more recent stabilization of the taxrate at fairly low levels.

14

increase in liquidity, with only a modest acceleration of inflation

(see Tables 1 and 2). Although the lack of an inflationarytradition goes a long way to explain this result, it was also

supported by a fixed exchange rate and strong price controls. Theimportance of the latter factors is supported by the significantrole played by explicit adjustments in the official exchange rateand other controlled prices in the inflationary dynamics after 1985(see below).

Oddly enough, up to 1984, the demand for money grew fasterthan that for term deposits (Table 2). Several factors may explain

this 'result. First of all, nominal interest rates were hardlyreadjusted with inflation up to late 1988 13/. With rising

inflation, this meant that term deposits became a close, thouuhilliauid, substitute for money. In a more orthodox pattern, excess

domestic liquidity was reflected in the increasing demand for blackImarket dollars, as the evolution of the relevant real exchange rate

indicates (Figure 2.C). The demand for dollars was enhanced bypolitical instability and the growing overvaluation of the officialexchange rate (Figure 2.A). The role of political factors mayexplain why devaluation in the black market overshoot the rapid

increase in liquidity levels and monetary aggregates actually

collapsed, if measured in (black market) dollars.

By 1984 the official exchange rate was only a minimal fraction

of the black market rate (Figure 2.B). This finally convinced the

government to devalue the official rate from 10 to 28 cordobas perdollar in February, 1985. As we have seen, the devaluation wasaccompanied by some austerity measures in the fiscal area. The needfor fiscal austerity also led the government to massively readjust

13/ The most important increase in interest rates took placein early 1986. Most lending rates were then established in the 20-30%. range. The highest rate (for loans to commercial firms) was

1. then placed at 45% a year. See Medal (1988), Table 32.

15 *

controlled prices (basic consumer goods and gasoline) at the sametime.

In an attempt to regulate the wage structure, the governmentdecreed in 1984 a complete wage scale (SNOTS), to which public andprivate firms should abide. As a result of the price adjustments

adopted in the first months of 1985, the government then attemptedto defend them against inflation and thus adjusted the scale threetimes from February to May 1985, increasing the average wage by146%. The adjustment was slightly higher than inflation duringthese months, but not enough to compensate for the fall in realwages in previous years. This and similar attempts in the followingyears to index wages were soon abandoned. Thus, wage policy wasineffective and in fact did not seriously try to avert the collapsein real wages which accompanied the explosive inflationary dynamics

. L (Table 1 and 4, and Figure 3). Under these conditions and theincreased demand for labor generated by growing black markets,

incentives to work in the lVformalV' sector (including thegovernment) were reduced. The result of this process was a general

fall in labor productivity, high labor rotation and growingpayments in kind l'/.

C. The outburst of inflationary Dressukes

The stabilization package adopted in early 1985 clearlyinduced a f'regimell change: from an atypical excess liquidity/low

domestic inflation/rapidly rising black-official exchange ratedifferentials, to a more 81classical" flight against the currency

and explosive inflationary effects of monetary expansion. Theformer regime was undoubtly one of "repressed inflation" (m

l'(/ In 1986 labor rotation in the central government was 50%.As a result, 44% of government employees in 1987 had one year orless in service (SPP, 1989). For payments in kind in thegovernment, see note 18 below. In mid 1989, some privateentrepreneurs informed the SIDA Mission that the costs of differentpayments in kind were three times the costs of the nominal wage

Y bill.

16

foreign exchange speculation). The fffundamentalsff were thus boundto prevail at some point. However, in the transition from oneregime to the other, the exulicit pricing decisions adopted by thegovernment in the first months of 1985 played the crucial role. Infact, as Figure 4 shows, the first dramatic acceleration ofinflation in the post-revolutionary period came as the directeffect of these policy decisions.

After this turning point, the price-monetary dynamics becameexplosive. The average monthly inflation rate constantlyaccelerated until it reached hyperinflation in 1989 (Table 1).Under these conditions, price controls became totally ineffectiveand only led to widening differentials between the legal and thefree markets for goods subject to regulations "/. The monetaryfuel was provided by the budget deficit, but also by the losses ofthe Central Bank in foreign exchange transactions and the need tofinance most of the nominal expansion in domestic credit throughmoney creation. The latter was made necessary by the decision tofix nominal interest rates at artificially low levels. Moreover, as'

the basic official exchange rate was only devalued once more duringthe period under analysis (on February 1986, when the official ratewas devalued to 70 cordobas per dollar), the costs of dollar-denominated domestic debts (foreign trade financing) were also keptat very modest levels. '

Accelerating inflation was accompanied by a great variabilityin monthly rates. Moreover, as Figure 4 indicates, rather than thestep-wise acceleration typical of ffinertialff inflationaryprocesses, it adopted a neat cyclical pattern. The length of thecycle was annual from 1985 to 1987. Hyperinflation was basically

15/ In May 1989, just before the major liberalization ofdomestic prices (see part V), the ratio of black to official marketprices was the following for some important consumer goods: rice5.5, kidney beans 5.2, soap 12.7, detergent 16.6, and toilet paper2.4 (SPP, 1988b).

.17

associated with the dramatic shortening in the length of the cycleto some 4 to 5 months in 1989. What is more interesting, someturning points, but not the intensitv of the cvcles, wereassociated with explicit decisions to correct basic pricingimbalances: February 1985, the same month in 1986 and, as we willsee below, February and June 1988.

As traditional monetary theory predicts, acceleratinginflation was accompanied by falling demand for domestic liquid

assets. For reasons which have already been mentioned, the demandfor term deposits declined ahead of that for money. The latterremained, in fact, surprisingly high even at fairly advanced stagesof the hyperinflationary process (Table 2). The strongunderdevelopment of the domestic financial market goes a long wayto explain this result. Finally, despite the gross and increasingovervaluation of the official exchange rate (Figure 2.A) and thedramatic widening in black/official rate differentials (Figure

2-B), falling liquidity was accompanied by an appreciation of thereal black market rate (Figure 2.C). Some policy measures may havesupported the process, particularly the creation of a rgreyff(parallel) foreign exchange market in 1985 I"/, where foreign

remittances and a fraction of export earnings could be legallysold. Massive US aid to the Contras may have also supported thisparadoxical outcome.

The parallel market was actually part of a more generalmultiple exchange rate regime. Since 1982, this regime became

increasingly complex, reflecting the decision to defend exportersagainst the growing overvaluation of the cordoba. It included twobasic mechanisms: exporters were authorized to keep part of the

16/ As part of the package of February, 1985, foreign exchangehouses were allowed to operate, under the regulation (and, in fact,ownership) of the Central Bank. The first and most important of thetwo existing houses, NECSA, started to operate in June of that

I year. BICSA started to do so in August, 1988.

18

foreign exchange earned, and domestic support prices for exportcrops were fixed at levels higher than those compatible withprevailing international prices and the official exchange rate. Thebasic difference between the two systems was the mechanism by whichthe implicit "export incentive" was financed. In the first case, itwas paid by importers of goods and services who bought the foreignexchange in the parallel market. In the second, it was financed bythe Central Bank.

As Table 5 indicates, both mechanisms were quite effective inraising the average exchange rate for exports significantly abovethe official rate (almost 100 times by January 1988). The latterwas increasingly relevant only for a few exports (mainly from stateenterprises) and most imports. Given the high import content ofsome exports of Nicaragua, the multiple exchange rate system thusoperated as a mechanism to increase.effective protection to exportactivities. Also, given the massive implicit subsidy on imports, .

the government had to rely on direct import controls to rationimport demand. As most imports were sold by parastatals, thismassive subsidy was largely passed on to the final user, subject,in any case, to significant resource misallocation, rationing andgrowing secondary black markets. Late in the process (June 1987),the government adopted a surcharge for most imports (the tasa de

estabilizacion monetaria, TEM) to finance the foreign exchangelosses of the Central Bank. By January 1988, this mechanism had

raised the average import rate significantly above the officialrate; still, the average export rate was almost 13 times higher

than that applicable to. imports.

Given the features of the multiple rate'system, the collapseof exports which took place through most of this period (Table 3)was only associated in part to exchange rate policies. A myriad offactors, affecting both the domestic supply and the external

demand, thus account for the evolution of exports: the effects ofwar in some areas of the country; lack of confidence by the private

.19.

sector; stronger incentives (price and, particularly, credit) given

to food crops; inefficiencies of state enterprises; exodus ofskilled labor and other labor supply shortages; and the collapse of

the CACM, which was decisive for non-traditional exports. Thesesame factors were responsible for the decline in economic activitysince 1984 (Table l), as production for the domestic marketcontinued to grow at moderate rates up to 1987 (1.2% a year in1984-1987).

Although imports fell with respect to the early post-revolutionary peak, they remained at historically high levels. In

any case, the country was able to finance its record externaldeficits, despite skyrocketing debt ratios and the interruption of

capital flows (Table 3). Three sources were basically used to

finance the deficits: mounting payments arrears, bilateral

. assistance from socialist countries and pre-financing of exportcrops.

V. 1988 STABILIZATION AND HYPWINFLATION "/

By early 1988, economic conditions were critical. The mosttransparent to all economic observers were the massive distortions

associated with the multiple exchange rate system (Table 5).However, this was only a manifestation of generalized macroeconomicdisequilibria. Monetary and fiscal imbalances were already

reflected in extremely high inflation rates --an average monthly

rate of 24.9% in 1987 (see Table l)--, which had led to the virtualcollapse of price controls. External deficits had also resulted in

near generalized moratoria on the foreign debt. Finally, the

country had already experienced a substantial fall in GDP per

capita and an even stronger contraction of real wages and privateconsumption per head. This dramatic deterioration in economic

"/ For a more extensive analysis of the 1988 and 1989stabilization packages, see Arana (1990), Ocampo and Taylor (1990),* Ocampo (1991a) and Taylor et al. (1989).

20.. conditions were combined by clear signs that the Contras were in

disarray, that the war was losing intensity and that peace talksamong Central American presidents were being successful, asreflected in the Esquipulas I Accord of August, 1987.

These conditions were the background to the two massivestabilization packages implemented in February and June, 1988. Thegoals of these programs were multiple and ambitious (SPP 1988a).They included: (1) the realignment of relative prices: (2) areduction of inflation rates by austere fiscal and monetarypolicies: it was stated early in the year that the centralgovernment deficit would be reduced to 10% of GDP in 1988 andeliminated altogether by 1990; (3) reversing the deterioration ofthe formal sector of the economy generated by price controls andfalling real wages; and (4) reconstituting the normal economic

-. functions of the wage payments system. Wage policy aside --whichwas explicitly conceived as a suuulv-side policy--, the objectives

-\ and instruments of the stabilization plan werethe IMF (1988) acknowledged later in the year.

fairly orthodox, as

Although these stabilization packages were more ambitious thanany previous effort, they tended to reproduce patterns which hadbeen common to macroeconomic policy since 1984. Particularly, thedifferent goals were not pursued with the same vigor, nor were thepackages globally consistent. Emphasis was placed on relative price

realignment. This fact was reflected in the outcomes of theprograms, as we will see below. On the contrary, fiscal andmonetary policies were not made consistent with the inflationtargets. Also, as in 1985, the attempt to defend or even increase

real wages was soon abandoned, giving way to a different policylater in the year.

The February package included five major provisions. The firstwas a monetary reform, by which 1000 old monetary units wereconverted into one new cordoba. This reform included the

..

21

demonetization of some 20% of existing liquid assets, which had,

attached to it, explicit political goals lo/. The monetary reformwas accompanied by the consolidation of all explicit and implicitexchange rates into two legal rates: 10 new cordobas in theffofficialff and 10.25 in the ffparallelff market. In relation toJanuary levels (Table 5), this implied that the official andaverage import rates were multiplied by 143 and 19, respectively,but the average export rate was devalued by only 46%. Thus, thehigh rate of effective protection to export activities implicit inthe multiple exchange rate system was altogether eliminated and thetrade deficit of the private sector, measured in domestic currency,massively increased. Nonetheless, the new legal rates were setsignificantly below the black market rate. Thirdly, the governmentdecreed significant increases in controlled prices. This wasaccompanied by a 675% increase in the average SNOTS wage level;

-. Finally, it announced a 10% cut in central government expenditure.

-. The major successes of this package were associated withexchange rate policies: the official rate was massively devalued inreal terms, as the black market rate appreciated and exchange ratedifferentials narrowed (Figures 2.A to 2.C). Nonetheless, the

official rate remained clearly overvalued and no mechanism wasadopted to avert its further real appreciation (only two minordevaluations of one new cordoba per dollar each were adopted inApril and May).

The major weaknesses of the February package were related,however, to fiscal and, particularly, monetary policies. Theinitial cut in central government expenditure was clearly

l*/ The short period to make the conversion in the banks (3days) was planned to leave the contra with a sizable stock ofuseless bills. It was also determined that households convertingmore than 10 million old cdrdobas had to leave their money indeposit at the banks for 12 to 14 months. This was aimed atspeculators and black market arbitreurs holding sizable amounts of

s cash.

..

22

insufficient to reach the target deficit, as the Olivera-Tanzieffect was eroding the tax base at a fairly rapid rate (seefootnote 12). On the other hand, the maximum domestic lending ratewas kept at 45% a Year and the government decided that thedevaluation of the official rate would not be passed on to dollar-denominated liabilities. Under prevailing conditions, thesedecisions were equivalent to a generalized debt forgiveness. Theyalso implied that the Central Bank had to incur in very largelosses in foreign exchange transactions (Table 2) and that anynominal increase of domestic credit had to be financed by moneycreation.

The mix of massive exchange rate, price and wage adjustmentsand weak demand policies initiated a new inflationary cycle, moreintense that those experienced in previous years (Figure 4). Under

-. these conditions, price controls were totally ineffective and realwages soon fell (see Table 1 and Figure 3). The government then

-. abandoned any attempt to arrest the collapse in the real income ofwage labor. In the second semester, real wages were only a fourthof its 1985 level (one-sixth in the case of central governmentemployees), if the GDP deflator is used to estimate them (Table 4).

.

The June package liberalized most prices and wages, decreed

large increases in those prices which remained under thegovernment's control (particularly gasoline) and deepened the

exchange rate reforms, but did little to make the global

stabilization policy more consistent. The official exchange rate

was then devalued by 700% and the parallel/black market

differential considerably narrowed. In the following months, the

parallel and, since late August, the official rate were devalued

more frequently (the latter five times between August 31st, 1988,and January 4, 1989). As a result, the overvaluation of theofficial rate was considerably reduced. Although the black/official

exchange rate differential remained substantial, it narrowedconsiderably with respect to previous years.

23

Nonetheless, fiscal policy was not significantly affected bythe June decisions. There was also no attempt to control the growthof domestic credit. However, two important reforms in monetarypolicy took place in June. First, the government did not assume theexchange rate risks on dollar-denominated domestic debts. Givendevaluation policy, this decision considerably raised the costs ofsuch liabilities, if contracted after February "/. Secondly,

authorities decided to index domestic interest rates. However, the"indexing rule" used was imperfect, particularly in the first few

months 'O/. Thus, from mid-June to mid-September, the maximumeffective lending interest rate was set at 14.9% a month. Sincemid-September, the rule was improved. Still, in the last months ofthe year, interest rates ran significantly below inflation levels(see Table 6).

In June, government wages were adjusted by 30%. Given massive.price increases accumulated since'February (790%), this was anextremely moderate rise. They were adjusted more frequently afterSeptember (monthly, except in December) but systematically below

inflation rates. To compensate for this fact;government employeeswere granted a food subsidy (AFA) in August 'I/.

The series of maxi-devaluations and massive adjustments inregulated prices, together with the inability of the authorities to

19/ For a debt contracted just after the February devaluationand paid in mid-January, 1989, the monthly interest rate was 61.9%,somewhat below inflation (63.7% a month in the same period).However, the closer the debt was contracted before the Junedevaluation, the higher the implicit interest. rate. Thus, aliability contracted just before that devaluation and paid inJanuary 1989 had a monthly cost of 107.7% or 19.3% in real terms.

20/ An annual interest rate was determined by adding UP themonthly inflation rates.

'l/ The subsidy took the form of the right to buy a basket ofbasic food products (10 lbs, of rice, 10 lbs of beans and 5 lbs ofsugar) paying between 5 and 10% of their nominal wages.

..

24

control the major sources of monetary growth were the fundamentalsources of the 1988 hyperinflation. As Figure 4 indicates, theeconomy underwent three distinct price cycles between January, 1988

and the first months of 1989. The first two of them were clearlyunleashed by the adjustment programs of February and June. Thethird was more closely associated with the effects of HurricaneJoan, which hit the country in October, generating losses estimated

by ECLAC at $840 million (CEPAL 1988b). The third cycle was themost intense. In total, the inflation rate ran close to 100% amonth between September, 1988, and January, 1989.

Overall, the monthly inflation rate was 62.4% in 1988.Following a classical pattern, this process was accompanied byrapid demonetization. By January, 1989, M, as a share of GDP hadfallen to 6.8% (Table 6). On the other hand, reductions in

-. aggregate demand (largely associated with the rising balance ofpayments deficit of the private sector measured in domestic

currency, as fiscal policy was not contractionary), relative pricechanges induced by the adjustment programs (real devaluation and

wage cuts, in particular) and supply shocks (the hurricane andelectric supply failures during the first semester) led to a 13.4%

fall in GDP. This was accompanied by a renewed deterioration ofexports, as the effective protection to export activities wasactually curtailed. The current account deficit improved somewhat,however, as the result of rising external transfers and a moderate.cut in imports.

VI. THE 1989 ADJUSTXENT PROGRAH

If massive relative price distortions associated with themultiple exchange rates and price controls were the dominanteconomic feature of Nicaragua in January, 1988, hyperinflation hadtaken over that place one year later. The urgent need for actionwas reflected in the rapid pace of demonetization and thegeneralized lack of confidence in government policies. Moreover,the authorities had few instruments to handle the explosive price

.25

.. dynamics. Price controls had collapsed in mid-1988 after severalyears in which they had become increasingly ineffective. The

official exchange rate was still overvalued and too distant fromthe parallel and black market rates to .be used as an anti-inflationary weapon. Finally, scarce foreign exchange placed severe

restrictions on any attempt to fix the exchange rate or liberalizeimports.

-.

Under these conditions, the government correctly understood

that a very orthodox policy was called for, combining fiscal andmonetary austerity with additional relative price adjustments. Thepackage adopted by the authorities in January included six major

provisions. First of all, central government e x p e n d i t u r ewas

massively cut to reach an expected deficit of 5.6% of GDP

(Ministerio de Finanzas, 1989). In practice, expenditure was cut

even further by transferring to the Ministries in the first monthsof the year less resources than were demanded according to budgetallocations z2/. As we will see below, an essential element offiscal austerity was a significant cut in public sector employment

(comoactacion).

Secondly, the government adopted a restrictive credit policy,accompanied by active interest rate management. The authorities

aimed at keeping positive real returns on term deposits and real

Costs for all (or most) types of credit. Thirdly, a system ofgradual devaluation was adopted in late January. In practice, thisled to small or medium-size devaluations some three times a month.

This was accompanied by, important readjustment of real regulatedprices during the first months of the year. On the other hand, as

22/ Transfers were cut by 33% in January, 25% in February, 20%in March and 18% in April. See SPP, Sintesis evaluativa de lasprincinales variables economicas de abrilde 1989 Y urouramacion demayo 1989, May 1989, p. 6. Similar documents will be quotedthereafter as Sintesis evaluativa.

..

26.

in 1988, the authorities stated the objective of arresting furtherdeterioration of public sector real wages.

In the speech in which President Ortega made public the newprogram, he also announced the willingness to establish new rulesof the game for the private sector: as a first step in thatdirection, he informed that expropriations would cease 2J/.Finally, the government adopted a financial programming systemcoordinated by the Planning Secretariat (SPP) and significantlyimproved the data base for short term macroeconomic analysis.

In terms of some of its major targets, the stabilizationprogram was initially very successful. Inflation rates fell rapidly(Figures 4 and 5). Actually, by March, the CPI increased by 8%,excluding regulated prices (public utilities and bus fares --Table

-- 6). On the other hand, the government was quite successful indevaluing the official exchange rate in real terms and instabilizing the parallel and black markets. By March, differentialsbetween the different foreign exchange markets had been reduced toless than 10% (Figure 6). The attempt to increase real regulatedprices was, on the contrary, short-lived. Since April, these prices

started to fall. The process accelerated in June, when thegovernment used these prices to repress inflationary pressures

(Figure 5.B).

In the face of falling inflation rates, demonetization ceasedin February (Table 6). The demand for term deposits also increased

since that month, but remained fairly low by historical standards.The government cut and maintained central government expenditure atlow levels. Indeed, as Tables 2 and 6 indicate, such expenditurestabilized around 22% of GDP, less than half its average 1988

level. Supported by some liquid foreign aid, since May, the

./ See "Esfuerzo National por la Paz y la Reconstruccidnf*,e Barricada, January 31, 1989, pp. 3-4.

..

27

government actually ran fiscal surpluses during a few months.

Finally, the fall in real wages was also arrested and partiallyreversed (Figure 3 and Table 4).

The major initial cost of stabilization was a strong

recession. In the first quarter of the year, industrial productionfell by 17% with respect to the same period in the previous year.However, it started to recover in the second quarter (Figure 7). Ingeneral, with few exceptions (agricultural foodstuffs and

electrical energy), inward-oriented were severely affected,

whereas, also with a few exceptions (cotton), exportables

experienced a boom (see Table 3 and below) 24/.

Hmployment effects were significant. By June, central

government employment had fallen by 14.3% with respect to the same

-. month in 1988 --11.000 employees approximately c-p, 1989).

Interestly enough, there were also a significant number of unfilled

-. vacancies in the central government, as the way budget allocations

were transferred to the different Ministries actually encouraged

this practice "/. In the same month, 16.500 civil employees,

including those in public sector enterprises, had been affected bycomnactacion. By October, this figure had increased to 17.000 26/.

This was equivalent to 2% of the labor force of the country.

Managua household surveys reflected this massive reduction of

public sector employment. However, they indicated that it did not

lead to increased open unemployment (which remain surprisingly low,at 5 to 6% of the labor force) but to growing informality (rising

24/ SPP, Sintesis evaluativa, June 1989 and suceeding months.

25/ The wage costs of vacant positions were transferred by theMinistry of Finance. The different Ministries used those resourcesto selectively increase wages of existing employees.

"/ SPP, Sintesis evaluativa, August 1989, p. 21 and Decem&;1989, p. 23. See also Nathan Associates (1991), Ch.2,alternative estimates of the effects of comuactacion, referring

I exclusively to central government employment.

..

28

proportion of self-employment and workers in very smallenterprises) and longer unemployment spells 27/.

The major problems faced by the stabilization program in thefirst months of the year were both related to monetary policy.Aside from the central government, other domestic agents weresubject to a credit crunch. The most important exception were thegovernment trading companies which, at the same time, continued toreceive massively subsidized credit. The profits made by thesecompanies by the joint effect of credit subsidies and realdevaluation were transferred to the producers of export crops(particularly coffee and cotton) by periodic resettlement ofaccounts (reliauidaciones), fuelling the money supply.

Interest rate policy became ' also a major source ofcomplications.-* Nominal rates were raised effective February 15. Inthe face of rapidly falling inflation, ex-post real rates were

-. extremely high from February to April (Table 6). Pure backwardindexation rules and significant lags in decisions --rates were

adjusted only once a month-- contributed to the same phenomenon.Some of these problems were eventually solved: "forward" indexationcriteria were introduced in April and weekly readjustments in June.However, the political opposition to high interest rates led thegovernment, in a meeting with agricultural producers on April 17and 18, to agree to maintain lending rates stable, to establishceilings on lending rates and new subsidized long-term rates, andto grant a mix of debt forgiveness and debt restructuring at lowinterest rates for foodstuffs and cotton producers. Starting inMay, these agreements led to government to fix some and, in June,

a lending rates below deposit rates, thus creating new sources ofCentral Bank losses (Table 6).

"/ SPP, Sintesis evaluativa, September 1989, Appendix II.

29

More generally, the authorities .were unable to controlsources of monetary growth different to the central government (seethe line "Emission - Deficit" in Table 6). From March to May, thisled to a sizable expansion of liquidity. Monetary expansion wasreflected in moderately rising inflation rates in May and,

particularly, in a wave of speculation in the foreign exchangemarket. The latter process was interpreted in some parts of thegovernment as a sign that the official rate was still overvaluedand that a maxi-devaluation was called for. These sectors were

apparently successful in restricting the official supply of dollarsto the parallel market. Expectations of devaluation then becamegeneralized and were reflected in massive speculation in theparallel and black markets. The authorities then decided to "follow

the market" and devalued the official rate by 111% on June 12.

The devaluation led to a rapid increase in gross internationalreserves --from an average of USs71.6 million during the first

semester to US$116 to 135 million after August--and in privatesales of foreign exchange in the parallel market--from an average

of USs2.7 million during the first semester to USS5.5 millionduring the second semester--. However, it was soon reflected inmassive inflation, which rapidly eroded most of its real effects(see Table 7 and Figure 6). Five weeks after devaluation, non-

regulated prices, excluding foodstuffs, had caught up with the

exchange rate. Devaluation was also fully reflected in food priceswith a somewhat longer lag (15 weeks); such lag may have been the

result of state intermediation in the food market through theagricultural marketing board (ENABAS). Thus, some four months afterdevaluation, it had been fully eroded, except for the incomplete

adjustment of regulated prices.

The June 12 devaluation also initiated a new "stop-go" cycle,not unlike that experienced during the first semester. The recoveryof economic activity was temporarily arrested (Figure 7), as

inflation came down fairly rapidly, reaching 10.5% in August, if

.30

regulated prices are excluded (Figure 5 and Table 6). This wasinitially accompanied by a dramatic fall in liquidity. However, asthe government was unable to control all sources of monetaryexpansion, liquidity, inflation and economic activity started topick up. Once more, this was reflected in speculation in theforeign exchange market in November. This time, the government didnot cut the supply of dollars to the parallel market and maintained

the system of gradual devaluation, thus averting major foreignexchange speculation and a new inflationary shock. However, it alsokept liquidity at high levels.

Aside from this basic change in policy reactions, there werealso two important changes with respect to the inflation cycleearly in the year. First, the negative interest rate marginswidened (Table 6) and became a major source of distortions and

'. Central Bank losses (Table 2). Secondly, the size of the parallelmarket doubled. This was equivalent, in fact, to an unplanned

I import liberalization.

Overall, the 1989 stabilization program was somewhat lesscontractionary and much more effective in terms of the inflationand exchange rate targets than its 1988 predecessor. GDP

contracted, nonetheless, by 5.1% (Table l), or 3.3% if the strong

cut in government services (15.5%) is excluded. Such betterperformance was solely associated with export dynamism, which led

to a 35% increase in the real exports of goods and services in theyear. On the contrary, domestic demand experienced a severe decline

(13.5%), inducing a strong recession of most inward-oriented

sectors.

Moreover, the inconsistency and fragility of the stabilization

program were increasing apparent to most economic observers (see,for example, Fishlow et al. 1990). First of all, it was clear that

the cut in central government expenditure could not be indefinitely. maintained: in particular, the extremely repressed real wages in

31

crucial sectors (particularly education and health) could not lastfor too long. On the other hand, as we have seen, monetary andinterest rate policies remained a source of considerabledifficulties. The sensibility of the foreign exchange marketcontinued to be a major source of instability. This reflected, inturn, the inability of the government to raise an adequate supplyof liquid foreign aid. Indexation increased in 1989 to levels whichwere incompatible with permanent reductions in the inflation rate.Finally, although the room for private initiative considerablywidened, no major advance was made in terms of designing stablerules of the game for the private sector.

VII. POLITICAL TRANSITION AND THE 1990 HYPERINFLATION

Stabilization efforts ceased during the first months of1990 '*/. This process had started before the victory of the United

National Opposition (UNO) in the February 25 elections, butaccelerated after the electoral results. From January to April, theofficial exchange rate and government regulated price werevirtually freezed. The real exchange rate and real regulated priceshalved, totally reversing the results of the 1989 stabilization. Asa consequence, the differential between the black and the officialexchange rate rapidly widened, as the supply of dollars to the

parallel foreign exchange market fell. In an attempt to compensate

for the adverse effects of overvaluation on exports, the governmentdecided in April to pay exporters the parallel market exchangerate, at the cost of increasing Central Bank losses.

Simultaneously, public sector finances became rapidly

expansionary. Taxes were cut as expenditure doubled with respect to

"/ Aside from official accounts of this process by the newgovernment (Republica de Nicaragua, 1990a), see the "Statement ofthe IMF Representative at the Pledging Conference on Nicaragua inRome on June 6-7, 1990", Mimeo. The last Sintesis Evaluativa of SPP(April, 1990) made also clear that major disequilibria werebuilding up.

32

1989 patterns (Table 8). In particular, a series of wage hikesincreased public sector real wages by 271% with respect to the lastquarter of 1989 (Figure 3 and Table 4). This was accompanied by anew public service law, which significantly increased the cost offiring government employees. By April, the central government

deficit peaked again at 35% of GDP. Despite the virtual freeze onprices under its control, inflation followed with a lag. In April,

inflation, excluding regulated prices, was approaching the criticallevels which are used in the economic literature to definehyperinflation --50% a month (Figure 5 and Table 8).

'Interest rates also lagged, and turned again strongly negative'in real terms in April (Table 8), inducing a reduction in thedemand for term deposits. Although the government eliminated thenegative spread between lending and borrowing rates in April, it

l simultaneously decreed a series of domestic debt write-offs andrestructuring at low interest rates for foodstuffs and cotton

'. producers. Thus, the correction of the negative interest ratespread did not eliminate Central Bank losses in 'financial

transactions.

As in 1979, the new administration, inaugurated on April 25,brought with it an agenda for structural change (see part IX

below), but had also to face the massive macroeconomic

disequilibria inherited from the last months of the Sandinistas.Indeed, during the first months in power, the Chamorro

Administration was overwhelmed by the sheer magnitude of thestabilization effort. The guidelines of the stabilization programwere summarized in the document presented to the internationalcommunity at the June Donors' Conference in Rome (Republica deNicaragua, 1990a). The implementation of the program was, however,

constrained by the unexpected magnitude of macroeconomic imbalancesand by the strong political opposition of the radicalized laborunions to both the stabilization effort and the agenda ofstructural change.

33

The basic element of the initial stabilization program was thegradual introduction of a new currency, the cdrdoba oro, fullyconvertible at a parity of one to one to the US dollar. Three

stages in the process were envisioned. During the first, the newcurrency would only serve as a financial and fiscal unit ofaccount. During this stage, it was expected that the governmentwould correct most of the macroeconomic disequilibria inherited

from the previous Administration. During a second stage, the newcurrency would be gradually introduced into circulation, at a

variable exchange rate with respect to the old currency. Finally,once fiscal and financial discipline had been restored, the oldcurrency would be taken out of circulation.

Following the program, during its first days in power, theCentral Bank decreed that new deposits and loans would bedenominated in cordobas oro. Annual interest rates for a three-month deposit in the new currency were set at 10.5%; those forshort-term loans fluctuated between 13 and 22%, depending on thesector involved. The Bank also decreed that the exchange rate forthe old for the new currency would be that prevailing the parallelmarket. Interest rates for old deposits and loans were placed at

levels which created strong incentives to transform them intosimilar assets and liabilities denominated in cordobas oro.

No credit cealings were initially established. This was meantto guarantee adequate incentives for plantings, as the new

government took over when the new agricultural season was starting.In turn, given the dominant share of agriculture in exports, thiswas thought to be an essential element for a strong export supplyresponse in 1991. Credit cealings were reestablished in August andbecame more stringent since October. In any case, in opendivergence to the experience of 1989, the non-fiscal sources ofmonetary expansion remained within close bounds since June

(Table 8).

.34-

An initial maxi-devaluation of both the official and parallelmarket exchange rates was also adopted during the first days of thenew Administration (close to 50 and lOO%, respectively). The

differential between both rates increased as a result. This wasmeant to facilitate a gradual adjustment of the domestic price ofimportables tied to the official rate (gasoline, in particular), asit placed the exchange rate for financial transactions (theparallel rate) at what was thought to be an equilibrium level. Inthe following weeks, the rates of devaluation were fixed so as toeliminate the differential between the two official rates. This wasfinally achieved in mid-June. With exchange rate unification, theelimination of the negative interest rate spread in May and the'firm decision not to decree additional write-offs and subsidized

restructuring of domestic debts, the sources of the quasi-fiscalCentral Bank deficit were thus eliminated. For the year as a whole,

-* Central Bank losses were only 2.8% of GDP, a significant reductionwith respect to previous years (Table 2) "/.

.-

Regulated prices were also massively adjusted during the firsttwo months of the Chamorro Administration. This process wasfollowed by similar efforts in later months (Figure 5.B). At thesame time, it implemented a major tax reform, which was effectivein the second semester of 1990. It included six major provisions:(1) the indexation of taxes, by converting all tax liabilities intocordobas oro; (2) the elimination of most exemptions; (3) thesimplification of the income tax, including the unification ofrates for corporations and major reductions in those for non-wagepersonal income: (4) the substitution of the withholding income taxsystem based on gross sales for more technically designed

29/ Some minor sources of Central Bank losses remained,particularly high administrative costs and some subsidies on theuse of USSR loans, which tended to compensate for the overvaluationof the ruble. More importantly, given the chaos characteristic ofCentral Bank accounting during the Sandinista years, not all debtwrite-offs, restructuring, etc., were fully reported. This hasgenerated continued accounting losses through 1991.

.. mechanisms; (5) an increase of the basic sales tax rate from 10 to

15%; and (6) a reduction in most other taxes, including theelimination of most selective consumption taxes for domestic goods.Later in the year, a presumptive income tax system was alsoadopted.

The net effect of the tax reform on government revenue wasexpected to be positive (see Pardo and Perry, 1990). However, thedomestic revenue of the central government remained fairly lowthroughout the second semester of 1990 and the first few months of1991 (Tables 8 and 9). Indeed, domestic revenues during the secondsemester were only 13.6% of GDP vs. 20.2% during the same period in1989 (Table 6). What is equally important, foreign grantscontributed only marginally to the finances of the centralgovernment and, overall, did not significantly increased withrespect to 1989 (Table 3).-m

Although government expenditure simultaneously decreased, itremained above the levels which had been typical in 1989.' Two major

obstacles to a rapid reduction of expenditure were faced by the newadministration. The first was the strong resistance of labor unionsto real wage reductions. Attempts by the government to reduce realwages during the first few months led to two major strikes in Mayand July, the latter with major political implications. Agreementswith the Sandinistas after the July strike finally facilitated a

reduction in real wages in the following months: even then, realpublic sector wages remained significantly above the levels which

had been typical in 1988 and 1989 (Figure 3 and Table 4). Thesecond obstacle was the large size of the Army --80.000 men in May,

larger in fact than the rest of the government bureaucracy--.Thanks to the agreements reached with the armed forces, it waspossible to reduce it to 28.000 men in December. At the same time,the government designed late in 1990 a voluntary retirement

program, the effects of which were only felt, however, in 1991 (seebelow).

36

The resulting deficit has been incompatible with a rapidreduction in the rate of inflation. Corrective measures adopted atthe onset of the new administration led to a major price shock inMay, which eliminated most the initial excess liquidity (Figure 5.Aand Table 8). After the initial shock, inflation rapidly declined,if regulated prices are excluded from the calculations. The Julystrike sharply reversed this trend. From August to October, itremained at high though declining rates. Moreover, the substantialreduction which it experienced in October and November (to 30% a

month) was not sustained, as the budget deficit continued toexercise major expansionary effects.

The introduction of the cdrdoba oro may have accelerated thegeneralized use of indexation rules and, particularly, dollarindexation. This process was clearly discernible since 1988 and,especially, 1989, when frequent correction of exchange rates,government-regulated prices and wages became regular practices.However, since mid-1990 domestic prices were increasinglydenominated in the new currency (i.e., in dollars). This fact nodoubt arrested the real effects of nominal devaluation after aninitial spur which left, however, the cordoba overvalued; rather,throughout the second semester the real exchange rate tended tomoderately appreciate, despite significant nominal devaluations(Figure 6.A). Nonetheless, the exchange rate differential betweenthe black and the official markets remained at low and decreasinglevels since June (Figure 6.C). Indeed, the black market exchangerate has appreciated in real terms since April. This behavior isconsistent with the close historical association between the realblack market exchange rate and liquidity levels (Figure 2.C).

Persistent high inflation was accompanied by a strong domesticrecession. In particular, the industrial recovery which had takenplace during the last months of the Sandinista Administration was

sharply interrupted (Figure 7). For the year as a whole, however,GDP grew by a moderate l.O%, the first such positive rate since

37

1983 (Table 1). Such moderate growth was associated with thecontinued export recovery (Table 3), as domestic demand remainedstagnant during the year.

In this unsettled macroeconomic environment, the decisionadopted in August to place the cordoba oro in circulation waspremature. In fact, in the last months of 1990, it was clear thata growing proportion of the issues of the new currency werereturning to the Central Bank for conversion into dollars and,thus, contributing to the drain in foreign exchange reserves.Indeed, after peaking at USS182.8 million in June (USS124.3excluding AID funds), gross international reserves experienced asteady decline during the second semester of 1990 (USS74.9 millionin December) and, if AID funds are excluded, in the first twomonths of 1992 (USS69.9 million in February). Under these

. . conditions, the convertibility of the new currency was suspended inJanuary, 1991, when it was decided that it could only be used topay for imports and some services (hotels and airline tickets).

More generally, persistent imbalances eroded the confidence ofthe public in the capacity of the government to control the pricelevel and led to growing distrust of the multilateral agencies andthe donor community on the effectiveness of the stabilizationprogram (see, for example, World Bank, 1990). Indeed, by late 1990,it was clear that macroeconomic conditions were out of control and

that the lack of confidence and overvaluation had extended to thenew currency (Ocampo, 1990 and 1991b). The success of the

tripartite negotiations.(concertacion) between the government, thelabor unions and the entrepreneurs which took place in Septemberand October (Republica de Nicaragua, 1990b) indicated, however,

that some of the social tensions and major political constraintswhich had affected the stabilization effort and the speed of thestructural reforms in 1990 would be more moderate in the new year.

38

VIII. THE 1991 STABILIZATION PROGRAMThe urgent call for action was finally grasped by the

government on March 3, 1991, when a new major stabilization programwas adopted. The program included seven major provisions: (1) a400% devaluation of the cordoba oro followed by a fixed exchangerate system; (2) the conversion of the old currency at a rate of 5million cordobas per cordoba ore; (3) a firm commitment toeliminate Central Bank credit to the government; (4) a 160% generalincrease in public sector wages, with additional hikes foreducation and health workers (35 and 55%, respectively), whichraised the average increase to some 200%; (5) the backwardindexation of term deposits and loans to the dollar: this wasenacted by readjusting the former by 400% and the latter by 240% to

400% depending on the nature of the liability: (6) the adoption ofa new interest rate policy, by which term deposits rates formaturities longer than a month were liberalized, that for one month

. .deposits raised from 9.5 to 12% and-indexed to the dollar, and thetraditional differentiation by sectors of the lending rates waseliminated and a simple system of dollar-indexed lending'rates wasadopted --18% for short term and 14% for long term loans; and (7)the maintenance of credit cealings for commercial banks 30/.

The fixed exchange rate system has been maintained since Marchand, as we will see, fiscal accounts have actually generated asurplus since then. Interest rate policy has been subject, however,to important changes. In particular, on April 9, the Central Bankimplemented, through the Banco Nicarautiense de Industria YComercio, BANIC, a temp,orary savings plan, denominated "Plan 30",in which deposits were not indexed to the dollar. The rate was setat 25% for April and rapidly reduced in the following months, to20% in May and 10% in June. This was a peculiar decision, as it

'O,/ Banco Central de Nicaragua (1991a), Ch. 1, and (1991b); IMF(1991); World Bank (1991); for public sector wages, NathanAssociates (1991), Ch. 1.

39

created a new source of Central Bank losses, at a time at which theelimination of the public sector deficit (which includes the quasi-fiscal deficit of the Central Bank) had become the priority of

economic policy. It was, however, rapidly abandoned.

In June, BANIC was allowed to issue 15-day deposits. On theother hand, in September, coinciding with the signing of the stand-

by agreement with the IMF, the interest rate cealings were

transformed into floors. In December, reserve requirements fordollar-denominated deposits were reduced from 100 to 25%; thisapplied, however, as a marainal requirement on deposits above thoseoutstanding on October 1, 1991. Although this liberalized for the

first time the use of dollar deposits, stringent conditions on

their use were maintained: in particular, banks cannot use them tofinance cordoba-denominated debts. Reserve requirements on domestic

deposits were maintained at 10% (a,level which had been fixed inlate 1990), but such provision remained largely nominal, as

commercial bank discipline in this area has not been established.Finally, the January, 1992, Letter of Intent expressed the

willingness to liberalize interest rate intervention altogether,

except of the regulation of lending rates for rediscount

facilities, including those financed with external funds.

Contrary to what the World Bank (1991, p. 7) has argued, no"cash budgeting" rule has been followed by the central government.

Moreover, expenditure has remained rather high: it fluctuated sinceApril around 24.5% of GDP, a level only somewhat lower than that

typical during the second semester of 1990 and above that of 1989(26.1 and 21.9%, respectively --see Tables 6, 8 and 9).Nonetheless, the elimination of the extra-budgetary military

expenses financed by foreign aid implied a significant reduction ofmilitary expenditure (some 13% of GDP according to IMF, 1991, Table

3). This had been made possible by the massive cut in the army

throughout 1990 (see Section VII above).

40

Moreover, through 1991 the government implemented a majorvoluntary retirement program, which reduced public sectoremployment, including the military and parastatals, by 20% --from156.600 to 124.700 =/. Estimates using INSSBI (Social SecurityInstitute) data are only slightly different. They indicate thatpublic sector employment (central government, public sectorenterprises and autonomous institutes) decreased from 165.275 inFebruary to 131.284 in December, i.e, by 20.6%. Together with theestimated cut in the army in 1990 (70.000), this implies that over100.000 persons --equivalent to 8% of the economically activepopulation of the country-- have enlarged the private labor supply,at the time when creation of formal jobs in the private sector hasremained rather limited (some 4.400 new jobs between February and

December, 1991, including mixed enterprises, according to INSSBIstatistics). The major reduction in formal sector employmentgenerated by the structural adjustment in the size of the publicsector has, thus, shifted macroeconomic disequilibria to the labormarket. Unfortunately, the lack of recent household surveys do notallow us to analyze the implications of this process J2/.

Part of the resulting cut in government expenditure has been

passed on through higher real wages, which after bottoming in

April, recovered during the rest of the year. By December, realcentral government wages were more than twice the level reachedduring the 1989 stabilization (Table 4). After March, only amoderate wage increase was decreed in June (some 20%); moreselective wage hikes have taken place in other months. The AFA foodsubsidy created in 1988 was eliminated in September, 1991. Finally,in the Letter of Intent presented to the IMF in January, 1992, thegovernment announced than no general wage increases would be

"/ See Letter of Intent of January 31, 1992. See, also, NathanAssociates (1991), Ch. 2.

32/ See some considerations (no doubt imprecise) on this topicin FIDEG, Observador Economico, No. 1, January, 1992.

. .

41

negotiated in 1992, but selective hikes might be granted on thebasis of savings generated by a more moderate retirement programwhich would be implemented in the year.

The decisive factor in the turnaround in central governmentfinances and the elimination of Central Bank budget financing, was,nonetheless, the increase in revenues. Domestic tax revenuesexperienced, indeed, a strong recovery in the post-stabilizationperiod, particularly during the second semester of 1991 (See Table

9; Ministerio de Finanzas, 1991; and Nathan Associates, 1991). Muchmore important, however, massive foreign aid allowed the governmentto run budget surpluses from March to October (Table 9)

The strong external support to the stabilization program wasalso reflected in the significant increase in official transfers:

from USS201.6 million in 1990 to USS528.1 million in 1991 (Table

3), equivalent in the latter year to 38.6% of Nicaragua's GDP. InApril, Taiwan lent the country US$60 million. In May, firmcommitments by the donor community to facilitate the country a mix

of grants and a bridge loan to repay arrears to the World Bank andthe IDB were finally made. Arrears were finally cleared in

September. In December, the Paris Club refinanced arrears and theprincipal and interest due through March, 1993, on extremely

favorable terms (approximately equivalent to a 50% debt servicereduction). Moreover, a different times in 1990 and 1991 a virtual

write-off of debts with Venezuela, Mexico and Colombia had beenagreed and the U.S. wrote off AID debts. Despite that, a

significant debt burden,remains (see part IX below).

Generous external financing was obviously crucial, not only toeliminate Central Bank budget financing but to improve the balanceof payments and to sustain a fixed exchange rate. Indeed, despitea significant reduction in export revenues, the large magnitude ofexternal transfers was reflected in a current account surplus in1991, if unpaid interests on the foreign debt are excluded (Table

..42

3). Gross international reserves improved by $97.8 million. Thisfigure is only slightly lower than the total monetary base (high-powered money) of the country in December, 1991 --CO$547 million or$109 million. It thus indicates that reserve accumulation played acrucial role in the rapid remonetization which took place in thepost-stabilization period (Table 9). Nonetheless, as in the 1989stabilization, net domestic credit to the private sector continuedto be a major source of monetary expansion. Indeed, the 1992monetary program implies that the public sector would generate a

massive surplus to finance a persistent and even growing deficit inthe domestic financial system (Table 10).

Given strong dollar indexation, exchange rate stability wasalso the clue to the sudden interruption of hyperinflation. Aftera very large initial spur during the week after devaluation,inflation was very moderate in the.following weeks and was followedby some deflation since May (Figure 5.A and Tables 7 and 9).However, this exclusively the result of a reduction in realregulated prices (Figure 5.B). Indeed, as Table 7 indicates, theinflationary effects of the exchange rate adjustment were fasterthan those of the June 12, 1989, devaluation, confirming thatdollar indexation had significantly increased since then. One weekafter devaluation, it was almost fully reflected in non-regulatedprices, except foodstuffs. In the latter, it took about 15 weeks todo so, no doubt as a result of state intervention through ENABAS.

The same factors which made the fixed exchange rate extremely

effective as a price stabilization devise thus made the March 3,1991, devaluation almost totally'ineffective as a way to correctthe overvaluation the cordoba. Figures 2.A. and 6.A indicate,indeed, that the real exchange rate has stabilized at a level some25% below that achieved during 1989 and only slightly above the1980 (overvalued) level. Moreover, in terms of relative prices, nomajor advance took place over the second semester of 1990. Thisimplies, in turn, that the real exchange rate is likely to hinder

43

rather than contribute to export dynamism, which will then have todepend on other factors ("crowding-in" effects of public sectorinvestment, positive effects of the structural reforms under wayand special supply-side programs).

Contrary to the previous stabilization experiences (see partsv to VII), the 1991 program was expansionary. This is clearlyreflected in the monthly evolution of domestic sales and productionof the industrial sector (Figure 7). For the year as a whole, GDPdeclined by 0.7% (Table 1) but this was the effect of a strongcontraction of exports (13.6%). The domestic sources of demandexperienced, on the contrary, a moderate recovery (3.6%) andprivate consumption a very strong one (Table 1). Although exchangerate overvaluation may have adversely affected export performance,other factors were also present, including, in particular, adverse

climatic effects on coffee production, excessive meat and shrimp.extraction in previous years and, perhaps, disruptive effects of

the structural reforms underway on sectors where public enterprisesare still dominant.

Price stability was, thus, obtained in 1991 without

significant contractionary effects, thanks to massive external

support and significant public sector reforms undertaken since

1990, but at the cost of persistent overvaluation of the cordobaand the significant cuts in formal sector employment. On the other

hand, as in previous stabilizations, the expansionary effects of

private domestic credit have not yet been brought under control;this thus remains the major area where major stabilization efforts

still lie ahead.

IX. STRUCTURAL REFORMSAs in 1979, the structural reform program adopted by the

Chamorro Administration is an ambitious one, but it aims at

reducing rather than strengthening the role of the State in .. economic activity. In particular, in involves: (1) the

.44.

reestablishment of solid private property rights: (2) a majorrestructuring of the public sector: (3) the redesign of thedomestic financial system; (4) the definition of new institutionsto manage the social costs of the transition to a more liberaleconomy; and (5) a major trade reform "/. Other areas of reformare probably less important. In particular, since price controlscollapsed in 1988 and hav

Hyperinflation and Stabilization in Nicaragua

Documents

Transcript of Hyperinflation and Stabilization in Nicaragua