Hostname: page-component-586b7cd67f-r5fsc Total loading time: 0 Render date: 2024-12-02T21:31:29.205Z Has data issue: false hasContentIssue false
  • English
  • Français

Brazilian Inflation from 1980 to 1993: Causes, Consequences and Dynamics

Published online by Cambridge University Press:  05 February 2009

G. Tullio  and
M. Ronci
G. Tullio
Affiliation:
Professor of Economics at the University of Brescia, Italy
M. Ronci
Affiliation:
An economist at the International Monetary Fund, Washington D.C., USA
Get access Rights & Permissions [Opens in a new window]

Abstract

This article focuses on Brazilian inflation: its causes, consequences and dynamics from 1980 to 1993. We argue that the main economic cause of the Brazilian inflation was the excessive growth of money, in turn caused by too high budget deficits. Oil and exchange rate shocks also played a role, together with the greater dependence of the Central Bank of Brazil on the government. We measure the degree of Central Bank independence by the variable ‘turnover’ of Central Bank governors defined as the number of months in office. The effect of this variable on inflation is found to be highly significant and positive.

Type
Articles
Information
Journal of Latin American Studies , Volume 28 , Issue 3 , October 1996 , pp. 635 - 666
Copyright
Copyright © Cambridge University Press 1996

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 See Spinelli, Franco, and Fratianni, Michele, Storia Monetaria d' ltalia (Milan, 1991), Carlo Favero and Franco Spinelli, ‘Deficits, Money Growth and Inflation in Italy: 1865–1990’, Working paper no. 279, Queen Mary and Westfield College, University London, October 1992.Google Scholar

2 See Law no. 4,595 of 31 December 1964.

3 See Complementary Law no. 12 of 8 November 1971.

4 See Law no. 6,045 of 15 May 1974

5 By way of comparison, in Germany the real stock of monetary base was about 1 per cent of GDP at the peak of hyperinflation in November 1923.

6 The actual loss of capital due to changes in indexation rules was estimated in 1991 to be 70 per cent of its real face value since 1980.

7 See Lees, Francis A., Botts, James M. and Cysne, Rubens Penha, Banking and Financial Deepening in Brazil (London, 1990).CrossRefGoogle Scholar

8 Medida Provisória no. 542 of 30 June 1994.

9 For a more detailed discussion of the monetary reform, see Ronci, Marcio, ‘In Search of a Monetary Constitution for Brazil’, in Langoni, Carlos, Ferrer, James and Ronci, Marcio (eds.), The Quest for Monetary Stability (Washington, 1995), chapter 7.Google Scholar

10 Carlo Azeglio Ciampi was Governor of the Bank of Italy from 1979 to 1993. See Goodman, John B., Monetary Sovereignty: The Politics of Central Banking in Western Europe (Ithaca and London, 1992), pp. 170Google Scholar, and Spinelli, Franco and Fratianni, Michele, Storia Monetaria d' Italia (Milan, 1991).Google Scholar

11 See Lees, et al. , Banking and Financial Deepening in Brazil, Table 5.4, p. 123.Google Scholar

12 See Tullio, Giuseppe and Ronci, Marcio, ‘Central Bank Autonomy, the Exchange Rate Constraint and Inflation: the case of Italy; 1970–1992’, forthcoming, Open Economies Review, 1996 and in Grauwe, Paul De, Micossi, Stefano and Tullio, Giuseppe (eds.), Inflation and Wage Behaviour in the EMS (Oxford, 1996).Google Scholar

13 It was estimated in 1989 that Brazil cleared about 7 billion cheques, 95 per cent of which were cleared in 24 hours. By comparison American Express processed on a world wide basis 28 million items a month, about the same number of cheques cleared in Brazil in a single day. During the high inflation period it was not unusual in Brazil to write cheques for less than the equivalent of two US dollars. See Lees, et al. , Banking and Financial Deepening in Brazil, p. 134.Google Scholar

14 See Novaes Walter Filho, ‘Inflação e Preço de Ação de Bancos Commerciais’, Disertacção de Mestrado, Pontifícia Universidade Católica, Rio de Janeiro, July 1988.

15 See Conjuntura Econômica, October 1995, p. 3.

16 Authors' calculation based on data published by Gazeta Mercantil: Balartço Anual, year 17, no. 17 (São Paulo, 10 1993).Google Scholar

17 For an interesting econometric study of Brazilian inflation, see Barbosa, Fernando de Holanda and McNelis, Paul, ‘Indexation and inflationary inertia’, The World Bank Economic Review, vol. 3, no. 3 (1990), pp. 339–57.Google Scholar

18 See Cagan, Philip, ‘The monetary dynamics of hyperinflation’, in Friedman, Milton (ed.), Studies in Quantity Theory of Money (Chicago, 1956), pp. 25–117.Google Scholar

19 See Tullio, Giuseppe, ‘Inflation and Currency Depreciation in Germany 1920–23: a dynamic model of prices and the exchange rate’, Journal of Money, Credit and Banking, 05 1995, pp. 350–62.CrossRefGoogle Scholar

20 For an explanation of Hendry's method, see Banerjee, A., Dolado, J., Galbraith, John and Hendry, David, Co-integration, error correction, and the econometric analysis of nonstationary data (Oxford, 1993).CrossRefGoogle Scholar

21 For previous estimates of a reaction function for Brazil, see Marcos, RobertoMontesano, da Silva, Controle Monetário em Economia Aberta: O Caso Brasileiro no Período 1974–1982 (Rio de Janeiro, 1989).Google Scholar

22 We also performed Granger causality tests between the budget deficit, money growth and inflation. In general, we found that the deficit causes money growth and inflation.

23 We also estimated equation (1) using instrumental variables to take into account the fact that some explanatory variables may have been endogenous. In particular, the nominal budget deficit may be endogenous because the interest payments on public debt are indexed to the past inflation and tax revenues depends on the business cycle. In general, the instrumental variable estimates were very like the ordinary square estimates presented in Table 10. We decided to present only the ordinary least square (OLS) estimates because, although they yield biased and inconsistent estimators, Monte Carlo studies show that OLS estimates have relatively small variances compared with other estimation methods and in a small sample (which is the case) their smaller variances may more than offset their greater bias, so that they have smaller mean squared errors. See Wonnacott, Ronald J. and Wonnacott, Thomas M., Econometrics (New York, 1970), p. 399.Google Scholar

24 The results of these tests are not reported here, to save space. We did not perform recursive parameter estimates, because the number of observations was not large enough. We also performed Granger causality tests between the deficit, money growth and inflation. In general we found that the deficit causes money growth and inflation.

25 For a similar reaction function estimated for Italy from 1970 to 1992, see G. Tullio and M. Ronci, ‘Central Bank Autonomy, the Exchange Rate Constraint and Inflation: the case of Italy; 1970–1992’.

26 We also estimated equation (4) using instrumental variables to take into account that some explanatory variables were endogenous. In general, the results were the same as the ordinary least square (OLS) estimates presented in Table 11. We decided to present only the OLS estimates for the same reasons as explained in note 23.

27 See Cukierman, A., Webb, S. B. and Neyapti, B., ‘Measuring the independence of central banks and its effect on policy outcomes’, World Bank Economic Review, vol. 6 (1992). PP. 353–98.CrossRefGoogle Scholar

28 The lags changed from annual to monthly in the last 30 years. See Holanda Barbosa and McNelis, ‘Indexation and Inflationary inertia’.

29 One may argue that high inflation could bring about a high ‘turnover’ of central bank governors. We carried out Granger-causality tests which showed that ‘turnover’ caused inflation and inflation did not cause ‘turnover’.