Inertial Inflation and the Cruzado Plan

LUIZ BRESSER PEREIRA* Fundacao Getulio Vargas, São Paulo, Brazil

Summary. - On 28 January 1986 the Brazilian government decided to freeze ali prices. This article examines the "heterodox shock" and the dcveloprnent by Brazilian economists of the thcory of incrtial inflation, which scrvcd as a thcorctical basis for the shock.

1. INTRODUCTION inflation underwent a new acceleration, after a partia! price freezc from April to July, which On 28 February 1986, Brazilian inflation, artificially maintained it at an annual levei of which was already at a levei of approximately 150%. Inflation jumped from 270% a year 350'10 per annum, was halted by a heterodox (almost 12% a month) to a new annual levei of shock, that is, by a freeze in prices, wages, and about 360% (almost 13.5% per month), as can be the exchange rate. This sarne phenomenon had seen in Table 1. taken place in Argentina nine months earlier. ln both cases the economic policy adopted was Table 1. Annualized inflation derived directly from a new development of the Latin American structuralist theory of inflation: Periods IPCA* IGPt the theory of inertial inflation. In both cases we witnessed an important advance in the think- ing on economic policy accornpanied by a Dec. 1984-!Vlar. 1985 275.8 268.8 corresponding development in rnacroeconomic Apr. 1985-Jul. 1985 155.8 149.6 Aug. 1985-Jan. 1986 305.3 334.4 theory. Based on these experiences anti 011 the Nov. 1985-Jan. 1986 360.2 453.7 theory of autonomous or inertial inflation which preceded them, it is here maintained that orthodox economists - monetarists as wcll as Keyne- Source: FIBGE and Getulio Vargas Foundation. *IPCA = Consumer Pricc Index, sians - should rethink their models and give tlGP = General Pricc Index. more importancc to administrative policies for stabilizing prices .. The objective of this article is to make a When, in January 1985, monthly inflation general analysis, right after the Brazilian (Consumer Price Index - IPCA) reached Stabilization Plan, of the shock itself, of the 16.2%, and in February 14.3%, the situation theory on which. it was base d, and of its pros- became unbearable. The economic authorities pects. and the President of the Republic quickly Iearnt When this article was written, one month after ali they could about the theory of inertial the Brazilian heterodox shock, everything inflation which some Brazilian economists had seemed to indicate that inflation would be developed at the beginning of the 1980s. They brought under control. If this prediction is also noted the success of the monetary reforms confirmed, and the 1,000% inflation in Argen- and general price, wage and exchange rate freeze tina and the 350% inflation in Brazil are elimin- in Argentina and in Israel. It became increasingly ated with almost no cost, economic theory and clear that the only way to eliminate Brazilian policy - via the theory of inertial inflation - will inertial inflation would be a general prize freeze have taken a great step forward. and a monetary reform, or, in other words, a heterodox shock. The decision to carry it out on

2. THE ST ABILIZA TION PLAN "Since wnnng this article , Dr. Bresser Pereira has ln the second semester of 1985, Brazilian become Finance Minister of Brazil.

1035 1036 W O R LD D E V E LO PM E N T

28 February was hastened by the high inflation De-indexation was not total, thus maintaining rates of January and February. certain guarantees for economic agents. Mone- The basic measures of the Stabilization Plan tary correction was kept on deposits in savings consisted of: (1) freezing ali prices, wages, and accounts. The ORTNs (Readjustable National the exchange rate on D day; (2) de-indexing the Treasury Bonds), which served as a base for the economy; (3) introducing a new currency, the whole indexation process, were transformed into Cruzado, in place of the Cruzeiro, from which OTNs (National Treasury Bonds), but they kept three zeros were removed; and ( 4) converting ali a guarantee of a yearly correction. A "trigger term contracts (wages, debts, rents, school tui- price adjustrnent" was established for wages, tions, mortgage payments) from Cruzeiros to guaranteeing that they will be readjusted when- Cruzados via formulas which would guarantee ever cumula tive inflation reaches 20%. At the the recomposition of the average real price of the last minute, due to pressure from the workers, it last six months. Ali other stipulations of the Plan was also established that, in the respective annual complemented or established exceptions to these data base for wage negotiations, each category of four basic guidelines. workers would have its wages readjusted by 60% lt was a general frecze accompanicd by the of the past inflation, even if the "trigger" of the setting of prices for thc most important consumar ' movable scale is not reached. goods (almost 500 articles, with prices differcnti- ated according to geographic regions). The prices 3. WHY A HETERODOX SHOCK? wcre controlled formally by professional inspec- . . tors and informally by the whole population. The success of the program dcpends mainly on Wages could not be directly converted frorn one condition: that on D day, 28 February, the Cruzeiros to Cruzados due to the different frceze does not provoke the emergence of any months in which wage readjustments were legally big winners or big losers. ln other words, rela tive executed. They could also not be readjusted to prices should be reasonably adjusted, without thcir pcak real value on D day by incrcasing any distortions causcd by partia! freezes and nominal wages by IOO'Yo of the period's inflation subsidies. Aside from this, thc time lags in the and then be converted into Cruzados, because, in increases should be small enough so that, on this dealing with a term contract , this would imply an D day, rela tive prices are not unbalanced be- increasc in the real wages of thc workers. lt ca use thc last rcadjustmcnt was closer or Iarthcr would have made thc freeze inviable for thc firms frorn the D day. as it would have eliminated their profits. The If there are big winners and big losers, this will solution was a simple conversion table for wages mean that the reform did not respect the law of and benefits received in the last six months into value; that market prices deviated too much frorn Cruzados in a way which guaranteed the reproduction prices; that relative ~rkes were dis- constitution of average real wages and therefore torted benefiting those who had increased their of workers' buying power. Monthly wages rcceived prices recently and penalizing those who had during the Septernber 1985 and February 1986 planned to raise their prices soon after D day; period were multiplied by corrective coefficients that the differences between the sectoral profit which transformed them into real wages in terms rates incrcased rather than dccreased; and that of 28 February prices, totaled, and then divided by wages and other time contracts were not con- six to obtain the average real wage. An 8% bonus verted according to their average real values. If was added on to this amount. this happens, the pressure from losers could As in Argentina, another table was established prove unbearable. Their prices will have to be to reduce the value of term contracts in general. readjusted sooner or !ater, and the stabilization Corporations had included their expectations for program will be a failure. future inflation in the prices for ali term sales. Contrary to what happened in Hungary in Once prices were stabilized, it became necessary 1946, and in Argentina and Israel in 1985, there to guarantee a discount for debtors which corres- were no adjustments made in relative prices on ponded to these inflationary expectations. The the eve of D day in Brazil. Prices of state daily table for converting Cruzeiros into Cruza- corporations were not increased because it was dos established this discount, corresponding to thought, perhaps a bit hastily, that most of them the geometric average of inflation in the last had already been adjusted, and because it was thrce months before the shock ( 14.65%, ). Spccial wrongly predictcd that there would be no oper- formulas for conversion were establishcd for ational public deficit in 1986. The Cruzeiro, on rents, mortgages and school tuitions, ali of which the other hand, was not devalued before being attempt to reestablish the average real value of converted into Cruzados, and therefore no safety the contracts. margin was established in relation to the ex- INERTIAL INFLATION AND THE CRUZADO PLAN 1037 change rate w hich w ould allow for a srnall price freeze and on administrative measures arnount of inflation in the follow ing rnonths. instead of market rneasurcs to cornbat inflation. The predicted fiscal balance carne, on the one It is also heterodox because it does not provoke a hand, frorn a fiscal reforrn approved at the end of recession. An orthodox shock begins frorn the 1985, w hich increased the tax base progressively, assumption that economic agents spend more and, on the other, frorn the Stabilization Plan, than they earn. For this reason it would be w hich elirninated the real loss of taxes w hich necessary to provoke a recession which would carne frorn the difference between the rnom ent in reduce spending. The heterodox shock already w hich taxes are cornputed and w hen they are recognizes that the market no longer functions - paid. For exam ple, in the case of the tax on since it no longer maintains stable prices, even industrialized products (placed on the though there is no excess of demand - and that consurner), w ith inflation the governrnent w as it is necessary to administer it. ln these terms, a losing receipts relative to the devaluation of the recession is unnecessary, if not even Cruzeiro of a three to four rnonth period. This counterproductive. Ao expansion of the loss disappeared w ith the stabilization, increasing economy would make it easier to stabilize prices, the taxes collected. ln 1985, the operational as an increase in productivity would make it public deficit (Public Sector Borrow ing Recruit- possible to reduce costs. rnents in real terrns) w as 3.4% of the GN P . W ith ln Argentina, the economy was already going the fiscal reforrn it w as estirnated that this deficit through a recession when the heterodox shock w ould be reduced to 0.5% of the GN P . After the was applied. This recession was aggravated be- Stabilization Plan, ali Governrnent estirnates cause it was necessary to sct a very high intercst pointed to a balanced budget. Tw o rnonths later , rate in order to avoid capital flight and an however, this idea was disrnissed, as there w ere inventory buildup; also, because it was necessary new projections of a public deficit of around 5% to implement a very large and abrupt cut in the of the GN P . The effects of the fiscal reforrns public deficit. The Stabilization Plan in Brazil w ere not as positive as previously thought, the was introduced under completely different cir- decision not to raise the prices of the public cumstances. The economy was in full expansion, sector irnrnediately before the shock w as a the externai sector balanced, international re- rnistake and the control of public expenditures serves relatively high and public finances under w as sornew hat loose. relative control. Also, Brazil has a powerful G iven the dorninantly inertial nature of Brazi- industrial economy which is internationally lian inflation, there was no other alternative to a competitive, and where opportunities for invest- heterodox shock. A n orthodox shock, inspired ment are enormous. For this reason, capital flight by rnonetarist or K eynesian econornics, is based is a much less serious problem, and does not on a cut in state spending and an increase in demand a high interest rate. Also, the risk of taxes, on a drastic reduction in the rnoney supply, speculative hoarding of stocks is smaller in on ao increase in the interest rate and on a Brazil. This is another factor which dispenses recession w hich w ould have indirectly )ed to a with a high interest rate, because, as there were reduction in w ages and in profit rnargins. This no drastic adjustments made in public prices on kind of shock w ould not have been viable the eve of the D day, economic agents did not because inflation w as inertial rather than due to , have a special reason to expect a future increase dernand pull. lt w as necessary . to break the in these prices. inflationary inertia, that is, the' ability of econornic agents to forrnally or inforrnally index their prices, thus autornatically passing their increases 4. THE THEORY OF lNERTIAL in costs on to prices. To do this, indirect lNFLATION rneasures w hich airn to reduce this inertia l increase of prices via the rnarket are inefficient, The theory of autonomous or inertial inflation, as they present an extrernely high cost-benefit which served as the base for the heterodox shock ratio. A ctually, there are only tw o correct w ays of 28 February 1986 in Brazil and, previously, for for cornbatting inertial inflation, both of w hich the Argentine shock of 15 June 1985, is part of are of an adrninistrative nature. One can either the wider structuralist theory of inflation. Actu- follow a policy of gradually controlling prices, ally, it could be considered as the third para- w ages, and the exchange rate in accordance w ith digmatic moment of the development of this a declining future rate of inflation, or, if the levei theory. The first moment, which was marked by of inflation is already very high, introduce the the works of the economists of CEPAL (The only alternativc, a heterodox shock. Economic Commission for Latin America of the The shock is heterodox because it is based on a United Nations), especially those of Sunkel 1038 W O R L D D E V E LO PM E N T

(1958), Furtado (1959), Pinto (1973) and Oliveira 5. CONSTRUCTING THE MODEL (1964), introduced two basic concepts: (1) supply bottlenecks which provoke sectoral elevation of ln order to construct a model of inertial prices, and (2) the propagating effects of infla- inflation, there was a need for a clear distinction tion, which spread the initial price hikes to the - which the conventional literature on inflation rest of the economy. Simonsen, who always uses does not rnake, or to which it does not give an eclectic approach - both monetarist and enough importance - between the factors which structuralist - to analyze inflation, made a accelerate inflation ( clemand or supply shocks) contribution to the theory of the propagating and those which maintain it (the "tendency," effects with his concept of inflationary feedback "mornentum," or inertial component of infla- ( 1970). tion). By starting with zero inflation, the theories The second paradigmatic moment for the of inflation were always oriented towards theory of inertial inflation took place with the explaining the causes of inflation in terms of the publication of Rangel's book, A lnflãcao Bra- causes of inflationary acceleration. This resulted sileira (1963). The endogenous nature of the in a debate to determine if, in each concrete case, money supply, inflation as a defense mechanism we were dealing with demand inflation (Keyne- of the economy itself when faced with a chronic sian, if the excess of demand has a fiscal origin; insufficiency of demand, and the concept or monetarist, if the origin is monetary; and structu- administered, or oligopolistic, inflation were the ral, if the excess of demand is sectoral), or with main ideas developed by Rangel. administejed. or cost inflation, provoked by the The third paradigmatic moment for the theory monopoly power either of corporations, of trade of structural inflation took place at the beginning unions or of the government. of the 1980s in Brazil. There are some works, After making the distinction between factors dealing especially with formulas for indexing which accelerate and those which maintain infla- wages, which contain some elements of the tion, it became necessary to clarify, when asking future theory of inertial inflation: Resende about the causes of inflation, whether the ques- (1979), Lopes and Resende (1980), Lopes and tion referred to the causes of the acceleration of Bacha (1981), Arida (1982). The idea of inflation, or to the maintenance of the levei of autonomous or inertial inflation gained con- inflation. If one was dealing with the first sistency with the works of Luiz Bresser Pereira question, the old debate between dernand infla- (1981, pp. 15-20), Silva (1981, pp. 67-75), and tion anel cost inflation continueel to be valid. Bacha (1982, Chap. 7), in which the idea that the However, if one was elealing with the seconel present inflation is a mere reproduction of past question, it then became necessary to search for inflation, that it is the result of the formal and the causes of the autonomous or inertial nature informal indexation of the economy, and on a of inflation. lt became necessary fo. know why wider levei, of the distributive conflict, begins to past inflation tended to automatically reproeluce become defined. ln Chile, Ramos (1977) macle a itself in the present. pioneer contribution to the theory of inertial As the theory of inertial inflation is an aelvance inflation, even though he adheres excessively to or a new conceptual stage of structuralist theory, the problem of inflationary expectations. Eck- it looks for this cause in the elistributive conflict. stein in the United States made an important For the structuralists, inflation is a real pheno- contribution to the theory of inertial inflation menon which always has monetary consequences with his concept of "core inflation" (1981). (and possibly causes). Economic agents always The theory of inertial inflation, meanwhile, try, either inelivielually or in groups, to maintain only became fully developed with the works of their share in income, anel if possible, even to Bresser Pereira and Nakano (1983 and 1984b), increase it. Besides, ali try to maintain a positive . Lopes (December 1984a), Arida (1984) and growth of income. ln the process of defending Resende and Arida (1984). This starting point their income share - anel given the current was the concept of stagflation, that is, of the inflation rate - economic agents increase their coexistence of inflation with recession, idle capa- prices alternately anel systematically. If the city and unemployment. ln arder to explain this economy consisteel of only three economic agents fact, it was necessary to construct a model of - A, B anel C - anel if the current inflation were inflatión which, contrary to what happens in x%, then corporation A woulel increase its prices Keynesian models of inflation, (1) assumes inertially by x% on the first day of the month, B unemployment and idle capacity, and (2) does on the 10th, anel C on the 20th, thus causing A to not begin from a situation of stability (zero again increase its prices by x% (as longas there is inflation) to explain inflation, but admits the no new accelerating factor) on the first elay of the existence of a given prevailing inflation rate. following month, anel so on. If any one of the INERTIAL INFLATION ANO THE CRUZADO PLAN 1039 econom ic agents were to stop increasing its to defend their share in the incarne by basing prices, it would have its incarne share reduced . their expectations principally on past inflation, A li were facing the prisoner's dilem m a. which is concrete, instead of acting in accordance ln his more detailed model of intlation, M ilton with expectations of declining future inflation, Friedman m ade a distinction between accelerat- which in the end may or may not confirm current ing factors and inflationary tendency. H owever, expectations. There is a radical divergence along as he understands intlation to be essentially a these lines between the structuralist or neo- monetary phenom enon, he attributes inflation, structuralist theory of inertial inflation, based on its acceleration and its "tendency" (inertia) di- the real distributive conflict, and the theory of rectly to the behavior of the m oney supply, and rational expectations, which uses the idea that indirectly, to expectations relating to this m oney inflation is a psychological phenomenon, which supply. A s a result, the distinction between depends on the way in which economic agents accelerating factors and m aintaining factors is change their expectations in terms of the changes lost, as ali are explained by the sarne cause: the of the regime of economic policy. money supply. On the other hand, by reducing The higher inflation is, the clearer the incarne the whole problem of inflation to the money effects of price increases become for ali economic supply, the monetarists, and especially the agents. Monetary illusion disappears. As a result, followers of the theory of rational expectations, the distributive conflict becomes sharper and turned inflation into a problem of expectations, inflation increasingly inertial. ln cases of and thus into a psychological problem . The hyperinflation - the study of which was very determ ining factor for inflation would bc the important for the formulation of the theory of expectations of econom ic agents in term s of the inertial inflation - the inertial component of money supply. inflation becomes absolutely dominant. On the A s infiation is an econom ic problem , and as other hand, the non-synchronization of price econom ics is a social science, it is tautological increases becomes minimal and relative prices that inflation is based on the behavior of indi- reach a reasonable balance in which no one gains viduais, on their expectations of the future, and or loses with inflation. Price increases become on their attempts to face uncertainty and to m ake almost simultaneous and instantaneous. The the most of the profits or wages which they are to differences between peak prices and real average receive. M eanwhile, it is far from clear that these prices almost disappear. For this reason, any expectations can be easily changed by means of significant exogenous factor - such as a mone- econom ic policy. lt is equally unclear that the tary reform accompanied by the receipt of expectations of econom ic agents influence their foreign loans which guarantee the fixation of the behavior to the extent that these expectations are exchange rate, as happened in Central European confirm ed in practice. There are many expecta- countries after World War I - allows for the tions which never m aterialize. H irschm an once immediate elimination of inflation without the defined disappointm ent, such a com m on pheno- need for a price freeze. Taking the viewpoint of menon, as a kind of m istaken expectation. He the theory of rational expectations, Sargent added: "it is much m ore' cornm on for expecta- (1982) mistakenly attributed the end of hyper- tions to exceed reality than for reality to exceed inflations to the change in economic policy. expectations" (1982, p. 13). Consequently, Actually, inertial inflation reached such dimen- econom ic agents know that they cannot take sions, the time lag between price increases their expectations too seriously. became so small, that inflation lost any Nevertheless, orthodox economists keep plac- redistributive effects and therefore its reason for ing increasing importance on expectations. They existing. The exogenous change in economic believe that governments can change expecta- policy was justa signal for the economic systern, tions through changes in the "economic policy which was ready and anxious to stop hyperinfla- regime." Once these changes are implemented, tion as long as the exchange rate was maintained they believe that economic agents will behave stable. rationally, in terms foreseen by the economic theory. As opposed to what monetarist economists 6. THE ROLE OF MONEY SUPPL Y assert, inflation is not essentially a monetary phenomenon, but rather a real phenomenon with ln the theory of inertial inflation, the money monetary consequences (and eventually causes). supply is considered to be a factor which sane- Inflation is a real phenomenon directly related to tions inflation. The endogenous nature of the the distribution of income. Actually, monetarists money supply was advanced by Bogtliew, underestimate the tendency for economic agents Schumpeter, Keynes, and Robinson in the 1930s1 1040 WORLD DEVELOPMENT but in a very imprecise form. Finally, this idea Perhaps the endogenous nature of money would was fully developed by structuralist economists, be the point where the two theories come closest particularly by Rangel. More recently it became to each other. ln the post-Keynesian model, popular among post-Keynesians, mainly due to however, there is no disti nction between the contributions of Kaldor (1970). Given the accelerating factors and maintaining factors, exchange equation and assuming relative stability there is no adequate explanation for stagflation, for the income-velocity of rnoney, the inertial the distributive conflict does not receive the sarne increase of prices necessarily leads to an increase emphasis, and shocks in supply are almost in the money supply. The alternative is a reduc- exclusively related to real increases in wages tion in the real money supply, a liquidity crisis, above productivity, thus ignoring or under- an increase in the interest rate and recession. To estimating oligopolistic increases in profit avoid this, the econornic system tries to defend margins and measures of "corrective inflation." itself by increasing the nominal money supply, The distinction is also clear of the levei of either directly, through an expansionist monetary economic policy. While monetarists are policy, or indirectly, through the automatic fundamentally oriented towards controlling the mechanism of the financial market. Faced with money supply and Keynesians towards managing the nced to create nominal moncy in order to fiscal .and incolne policy (thus admitting the prevent a reduction in the real money supply, • importance of the distributive conflíct), thc and thus to make the volume of current transac- neo-structualists emphasize administrative con- tions viable (macroeconomic reason), and given trais of prices and wages. the credit available for financially sound corpora- If inftatión isinertial, if it is not the result of an tions (microeconomic reason), the banking sys- exccss of demand, but rather of the ability of tem automatically expands the nominal amount economic agents to automatically pass on in- of credit. Even when the Central Bank succeeds creases in their costs to their prices, the natural in avoiding an increase of the nominal money solution to break this cycle is administrative pricc supply through particularly restrictive measures contrais. This solution becomcs even more natu- of monetary policy, the economic system still ral when we learn that this ability to automatic- manages to defend itself by increasing the velo- ally reproduce past inflation in the present would city of circulation of the existing money stock, becorne greater not only if inflation is higher, but with a result which is similar to a nominal also as the market for goods and services as well increase in the money demanded by the market. as the labor market · are olígopolízed and Naturally, there are special cases in which an nationalized. increase in the money supply can have an Administrative wage-price contrais can be exogenous component and transform itself into carried out either gradually, in keeping with a an accelerating factor of inflation. If the govern- forecasted declining inflation, or abruptly, via a ment decides to act in a populist way and finances general freeze of prices, wages and the exchange its public deficit by emitting money, or, more rate. As long as inflation is at relatively low precisely, by increasing the real money supply, leveis, it is still possible to think in terms of a the resulting acceleration of public and private gradual administra tive control. However , when investment (given the reduction in the interest inflation reaches high leveis (more than 300% in rate) would lead the economy to a classic demand Brazil, and more than 1,000% in Argentina and inflation. lt must be made clear, however, that in Israel) there is no other choice than a freeze to the simple existence of a nominal public deficit abruptly cut inflation. financed by an increase in the nominal money The first economist who perceived the need for supply does not transform this increase into an an administrative type of shock treatment in accelerating factor of inflation. While the money order to eliminate inflation was Octavio Gouveia supply increases nominally when there is inertial de Bulhões, when, at the beginning of 1983, he inflation, in order to maintain the real money proposed a total de-indexation of the Brazilian supply, it would merely sanction inertial infla- economy. De-indexation is an administrative tion. measure. However, as he is a monetarist economist and does not make use of the concept of inertial intlation, he did not propose a freeze, 7. THEORETICAL ANO POLICY preferring to support de-indexation with a radical IMPLICA TIONS reduction of the money supply. The proposals for a freeze and de-indexation Naturally, there are numerous points in com- are natural results of the theory of inertial mon between the neo-structuralist thcory of inflation. Consequently, as soon as or immedi- inertial inflation and the post-Keynesian theory. ately after this theory was formulated, proposals INERTIAL INFLATION ANO THE CRUZADO PLAN 1041 formulated by these sarne economists appeared. industry conflicts on discounts in credit sales Bresser Pereira and Nakano (July 1984b, pp. could create obstacles to production. 123-124) proposed a "heroic solution for ln both countries, meanwhile, the freeze has controlling inflation," with the choice of a D day not been suspended yet - in Brazil, beca use for a general price freeze. Lopes ( 1984a; 1984b) it is still very early; in Argentina, because the introduced the expression "heterodox shock" government does not yet feel secure enough for and made a more complete proposal along these that. Of course, the big question is to know lines, which would later serve as the main what will happen after the end of the price theoretical source for the Argentine and Brazi- freeze. Iian shocks. Resende (1984a; 1984b) and Arida ln terms of the inertial inflation theory, the (1984, pp. 5-18), both separately and together first big risk for the plan is that the shock was not (1984), proposed a "rnonetary reform" and an neutral enough from the distributive point of "indexed currency." Simonsen ( 1984) supported view on D day. The second big risk is that the and further developed the original proposal of imbalance of relative prices - which expresses Resende. Leite (1985) presented a proposal of the lack of distributive neutrality - would "overcoming inflation in 100 days." Modiano increase instead of decrease during the period of (1985) made a formal proposal for converting the freeze. ln the event that these two problems wages to an average real wage. Lastly, Dorn- converge, inflation would start to accelerate busch (1986) proposed a freeze while maintaining again frorn the moment when the freeze, having indexation. become insupportable, is suspended. ln Argen- Taken together, these proposals served as the tina, up to March 1986, the accumulated inflation base for the Austral and the Cruzado Plans. ln for the year was 9.6%. The first devaluation of the Argentine case, there was more preoccupa- the Austral, of 3. 75%, was announced at the tion witlí correcting relative prices at the time of beginning of April, anticipating monthly de- the shock, as occurred in Hungary in 1946.2 ln valuations for the next two months of 2%. Brazil, the major preoccupation was with Perhaps this is a consequence of the Argentine guaranteeing distributive neutrality via severa) government's delay in thawing prices, without, at formulas for conversion. ln Argentina, the the sarne time, managing to correct relative economy was in a recession; in Brazil, in an prices. expansion. ln both cases, however, the hetero- No matter how inertial inflation is, on the D dox shock was based on the choice of a D day to day there will always be some imbalances in end inertial inflation through a general price relative prices due to controlled prices or to freeze. ln this way Brazil avoided the deep exceptional behavior in the supply or demand of recession which an orthodox policy (either certain goods in the period immediately before gradualist or by shock) certainly would have the freeze. These imbalances will be frozen along provoked to eliminate such high inflation. with the prices, and, moreover, will become Argentina was not as successful because its more visiblc. On the other hand, ali of the economy was already in . a recession when the formulas and tables for converting long-terrn shock was applied. contracts into Austrais or into Cruzados would not prevent that, at the beginning, those imba- ' ,...... lances increase instead of decrease, due to the 8. PROSPECTSfOR SUCCESS different timing of price increases and to the fact that the D day is an arbitrary day among the Now the big question is whether or not both differently timed price increases. Even though stabilization plans will be successful. Orthodox there is an attempt to convert prices to their economists are generally skeptical, or else they average and not to their peak value, there are pretend that rather than a heterodox shock in always some prices which could not be converted their countries, there was an orthodox one. The in this way, or that were converted imperfectly. structuralist economists are confident, because in ln Brazil, during the days following the shock , both stabilizations, as well as in Israel, they see a there were two themes which dominated discus- confirmation of their theory of inertial inflation. sions: one was the conversion of wages in Both in Argentina (10 months later) and in Cruzeiros into Cruzados via the average buying Brazil ( 45 da ys later) inflation has been con- power of the last six months. The other was trolled and prices have stabilized. The recession interindustry relations, given the need of the in Argentina was already occurring before the corporations which were selling on credit to offer shock; recent information gives indications of the a discount relative to the inflation which was built beginning of a recuperation. ln Brazil, therc has into their selling prices. As for wages, although been no sign yet of a recession, although inter- it was naturally difficult for the workers to 1042 WORLD DEVELOPMENT understand the conversion of wages based on the else if there were a large deflation in the first average real wage rather than on the peak real 111011th. Obviously, neither of these two alterna- wage , it seem s to be sufficiently clear that there tives is realistic. ln Brazil, for example, deflation has not been any loss for wage earners. The in the first month was only 0.11 %. ln April, the shock maintained the relation between wages second month, we had a positive inflation of and profits more or less unaltered. 0.78%. The problem of interindustry relations is more ln any case, the freeze cannot last for long complicated, since the govcrnment decided not because , in the first months after the shock , it to distinguish between imbalances in relative would be theoretically possible to decrease irn- prices, which already existed on D day, from balances in relative prices by administrative those resulting from the inflation built into term means; in the long run, however, the risks are contracts, nor to establish a clear rule for very great that administering prices instead of discounts for new supplies of merchandise sold letting them be controlled by the market would on a credit basis. lnstead, it intended to let the increase rather than decrease those imbalances. market resolve these two problems jointly, For this reason it will he necessary to suspend the through discounts frcely set between thc corpora- frecze intime, preterably gradually, and definite- tions. Thc result of this policy could be favorable, ly before this suspension can no longer be a finally reducing the imbalances in relative prices," decision made by government but rather be- but it is more likely that it will be negative, comes an inevitable fact, an imposition made by increasing these imbalances. ln the sarne way in market itself. This would happen if, rather than which a conversion formula was set for the being respected - hecause the law of value was long-term contracts which were in force on D obscrved in determining relative prices - the day, a formula for calculating a minimal discount market was violated during the period of the for new supplies should have been adopted. One freeze. · month after the Cruzado Plan , interindustry As was expected, the stabilization of prices relations continued to be tense. W holesalcrs and provoked a greater demand for money. The retailers were not accepting the discounts offered economic agents no longer needed to recycle by industry. There was a shortage of merchandise their Cruzeiros quickly, preferring instead to in stores at the time that stocks were accumulat- increase their cash deposits. As a result, cash ing in the factories. The difficulties, however, deposits in the banks increased by 83.4%, the seemed to be resolved more easily than expected. monetary base increased 36.1 % and the money As opposed to what happened in Argentina. the supply (M-1) 85.3% in March 1986. For a while, Brazilian econom y was expanding at the time of until the remonetization of the economy is the shock. ln spite of the interindustry difficulties, completed, the emission of currency could help this expansion continued after the shock, stirnu- to finance the public deficit in a non-inflationary lated by higher wages (since many corporations way. It is difficult to determine the new ideal went against the conversion formula and did not levei of money with which the economy can reduce nominal wages when the formula called operate. If the growth of the money supply for this) and by the monetary illusion of con- provokes an excessive lowering of the interest sumers. Actually, the expansion is helping to rate and pressure on demand, these would be solve the interindustry distributive conflict. signs that it is necessary to interrupt the Strictly speaking, a general freeze of three or remonetization process. Until then, even though four months should be enough to break inertial there may be some excess demand, the real inflation. A longer freeze could only be justified interest rate for depositors should be maintained if', during this period, the government succeeded at approximately the sarne levei as before the in diminishing imbalances in relative prices shock: 15% per annum. through a judicious administration of prices, and One month after the shock, concerns about the thus prevent the shock from representing large public deficit returned. The projections for the tosses for some and large gains for others. lt is operational public deficit (Public Sector Borrow- important to note that this administration of ing Requirements in real terms) fluctuated be- prices is inconsistent with zero inflation, which tween 3.5 and 5.4% for 1986, showing that the was the airn of the Brazilian government. A small fiscal reform of December 1985 was not enough amount "of inflation is necessary, resulting from to balance the public budget. This deficit could the increase of outdated prices. Zero inflation be financed with the internai savings of the would only be feasible if the increase in these private sector, hut it is a cause for concern in the outdated prices were compensated for by a sense that the private sector also is showing signs decrease in the prices adjusted more recently, or of wanting to recover investments. ln this case, INERTIAL INFLATION AND THE CRUZADO PLAN 1043 there would be fewer resources available for the the plan deserves some restrictions, especially for public sector, thus allowing for a substantial not having regulated interindustry relations, it increase in the ínterest rate. was definitely carried out with technical compe- The conditions necessary for the success of the tence, However, there are many decisions to be Cruzado Plan are clear. Actually, the Plan is made until the moment arrives in which the already an extraordinary success, a great con- market can go back to regulating the economy. quest of theory and economíc policy. The pre- These decisions must continue to be made dominantly inertial nature of Brazilian inflation competently so that the success of the plan can be before the shock is indisputable. And although guaranteed.

NOTES

1. ln this respect, see Merkin (1982). 2. See Bomberger and Makinen (1983) and Hegedus (1986).

REFERENCES

Arida, Pérsio, "Reajuste salarial e inflacão," Pesquisa e Furtado, Celso, Formacão Econômica do Brasil (Rio Planejamento Econômico, Vol. 12, No. 2 (August de Janeiro: 1959). 1982). Hall, Robert E., Inflation: Causes and Effects (Chic- Arida, Pérsio, "Economic stabilization in Brazil," ago: University of Chicago Press, 1982). Discussion text No. 84 (Rio de Janeiro: Pontifícia Hegedus, Georges, "A hiperinflacão Húngara de Universidade Católica, December 1984). 1945-46," Revista de Economia Política, Vol. 6, No. Arida, Pérsio, lnflacão Zero: Brasil, Argentina, Israel 2 (April-June 1986). (Rio de Janeiro: Paz e Terra, 1986). Hirschman, Albert O., Shifting lnvolvements (Prince- Bacha, Edrnar, lntroducão à Macroeconomia. Uma ton: Princeton University Press, 1982). Perspectiva Brasileira (Rio de Janeiro: Editora Cam- Kaldor, Nickolas, "The new monetarism," Lloyds pus, 1982). Bank Review (July 1970). Bomberger, W. A., and G. E. Makinen, "Toe Hunga- Leite, Antonio Dias, "Idéia fundamental de um plano rian hyperinflation and stabilization of 1945-46," de domínio de inflacão em cem dias," Boletim de Journal of Political Economy, Vol. 91, No. 5 Conjuntura Industrial of the Instituto de Economia (December 1983). Industrial of the Universidade Federal do Rio de Bresser Pereira, Luiz, "Inflacâo no capitalismo de Janeiro, Vol. 5, No. 1 (January 1985). Republished estado (e a experiência brasileira recente)," Revista in Antonio Dias Leite, A Transicâo para a Nova de Economia Política, Vol. 1, No. 2 (April-June República (Rio de Janeiro: Nova Fronteira, 1985). 1981). Republished in Bresser Pereira and Nakano Lopes, Francisco Lafayete, "Inflacâo inercial, hiper- (1984a). . inflacão e desinflacão: Notas e conjecturas," Revista Bresser Pereira, Luiz, and Yoshiaki Nakano, "Fatores de ANPEC (December 1984a). Republishcd in aceleradores, mantenedores e sancionadores da in- Revista de Economia Política, Vol. 5, No. 2 (April- flacão," em X Encontro Nacional de Economia June 1985). (Belém: Associacão Nacional de Centros de Pós- Lopes, Francisco Lafayete, "Só o choque heterodoxo Graduacão em Economia, ANPEC, 1983). Pub- .- pode derrubar a inflacão," Economia em Perspectiva lished in Revista de Economia Politica, Vol. 4, No. 1 (Conselho Regional de Economia: August 1984b). (January-March 1984). Republished in Bresser Lopes, Francisco Lafayete, Choque Heterodoxo. Pereira and Nakano (1984a). Combate à inflacão e reforma monetária (Rio de Bresser Perira, L., and Y. Nakano, Inflacão e Recessão Janeiro: Editora Campus, 1986). (São Paulo: Editora Brasiliense, 1984a). Lopes, Francisco Lafayete, and André Lara Resende, Bresser Pereira, L., and Y. Nakano, "Política adminis- "Sobre as causas da recente aceleracão inflacion- trativa de controle da inflacão," Revista de Economia ária," in VI/ Encontro Nacional de Economia (Nova Poltüca, Vol. 4, No. 3 (July-September 1984b). Friburgo: Associacão Nacional de Centros de Pós- Bresser Pereira, L., and Y. Nakano, "Inflacâo e Curva Graduacão em Economia, ANPEC, 1980). de Phillips," Revista de Economia Política, Vol. 6, Lopes, Francisco Lafayete, and Edrnar Bacha, "Infla- No. 2 (April-June 1986). tion, growth and wage policy," Discussion text No. Dornbusch, Rudiger, "Reforma monetária no Brasil," 10 (Rio de Janeiro: Departamento de Economia da Folha de S. Paulo (3 January 1986). Pontifícia Universidade Católica, 1981). Eckstein, Otto, Core lnflation (Englewood Cliffs: Merkin, Gerald, "Towards a theory of the German Prentice-Hall, 1981). inflation," in Gerald D. Feldman et ai., The German Friedman, Milton, "The role of monetary policy," The Inflation (Berlin: Walter de Gruyter, 1982). American Economic Review, Vol. 58, No. 1 (March Modiano, Eduardo, "O choque argentino e o dilema 1968). brasileiro," Discussion text No. 12 (Rio de Janeiro: 1044 WORLD DEVELOPMENT

Pontifícia Universidade Católica, 1985). Published in Resende, A. L., "Moeda indexada: Nem Mágica nem Revista de Economia Polttica, Vol. 6, No. 2 (April- paracéia," Revista de Economia Politica, Vol. 5, No. June 1986). 2 (April-June 1984b). Oliveira, Julio G., "La inflacion estructural y el Resende, André Lara, and Pérsio Arida, "Inertial estructuralismo latinoamericano," in O. Sunkel et inflation and monetary reform in Brazil" (Rio de ai., lnflacion e Estrutura Economica (Buenos Aires: Janeiro: Pontifíca Universidade Católica, 1984). Paidos, 1967). Originally published in English in Presented at a seminar in Washington sponsored by Oxford Economic Papers, Vol. 16, No. 3 (November the Institute of International Economics, December 1964). 1984. Published in Arida (1986). Pinto, Anibal, Inflacion: Raices Estructurales (Mexico: Sargent, Thomas J., "The ends of four big inflations," Fondo de Cultura, serie Lecturas, No. 3, 1973). in Hall (1982). Ramos, Joseph R., "Inflacion persistente, inflacion Silva, Adroaldo Moura da, "lnflacão: Reflexões à reprimida e hiperinflación. Leciones de inflation y margem da experiência brasileira," Revista de Eco- estabilizacion en Chile," Cuadernos de Economia, nomia Polltica, Vol. 1, No. 3 (July-September 1981). No. 43 (Instituto de Economia da Pontifícia Univer- Simonsen, Mário Henrique, lnflacão: Gradualismo X sidade Católica de Chile: December 1977). Repub- Tratamento de Choque (Rio de· Janeiro: ANPEC, lished in Desarrollo Economico. 1970). Rangel, Ignácio, A lnflacão Brasileiro (Rio de Janeiro: Simonsen, M. li·, "Desindexacão e reforma mone- Tempo Brasileiro, 1963). Third edition São Paulo: tária," Conjuntura Econômica, Vol. 38, No. 11 4 Editora Brasiliense, 1978. • (November 1984). Resende, André Lara, "Incompatibilidade distributiva Sunkel, Oswaldo, "La inflacion chilena: Un enfoque e inflacão estrutural," in VI/ Encontro Nacional de heterodoxo," El Trimestre Economico, vol. 25, No. 4 Economia, Mimeo (Atibaia: Associacão Nacional de (October-December 1958). Transcribed in Sunkel et Centros de Pós-Graduacão em Economia, ANPEC, ai. (1967). 1979). Published in Estudos Econômicos, Vol. 11, Sunkel, Oswaldo, et al., Inflacion v Estructura Econo- No. 3 (September-December 1979). mica (Buenos Aires: Paidos, 1967). Resende, A. L., "A moeda indexada: Uma proposta para eliminar a inflacão inercial," Gazeta Mercantil (26, 27, 28 September 1984a).