Brazil’s 1995–1998 PROER and PROES Bank Restructuring After the Plano Real

When Brazil introduced the new real currency on July 1, 1994, under Finance Minister Fernando Henrique Cardoso’s Plano Real, annual consumer price inflation plunged from more than 2,700 percent in the twelve months preceding stabilization to below 20 percent by 1995 and single digits by 1997. While price stability rescued millions of wage earners from the ravages of hyperinflation, it simultaneously delivered a severe structural shock to Brazil’s commercial banking system. For two decades, Brazilian private and state-owned banks had derived between 35 and 42 percent of their total operating revenues from the “inflationary float” (ganhos com o float)—capturing triple- and quadruple-digit nominal yields on overnight government securities while paying zero or below-inflation interest on sight deposits and tax collections in transit. As disinflation wiped out roughly USD $10 billion a year in float revenues overnight, deep balance-sheet insolvencies and accounting frauds surfaced at major private institutions including Banco Econômico, Banco Nacional, and Banco Bamerindus, alongside chronic fiscal deficits at state-owned regional banks such as Banespa and Banerj. To avert a systemic bank run without derailing currency stabilization, the Central Bank of Brazil (Banco Central do Brasil, or BCB) engineered two landmark resolution frameworks: the R$ 21 billion PROER program for private banks in November 1995 and the R$ 60 billion-plus PROES program for state-owned banks in August 1996. ^[1]

Hyperinflation and the Mechanics of the Brazilian Float#

During Brazil’s chronic high-inflation era of the 1980s and early 1990s, the domestic financial sector evolved into one of the most technologically sophisticated clearing systems in the developing world—not to evaluate long-term credit risk, but to maximize the capture of inflationary seigniorage. Because prices rose by 20 to 40 percent per month prior to July 1994, the real value of non-interest-bearing demand deposits, checks in clearing, utility bill payments, and federal or state tax collections eroded rapidly by the hour. Commercial banks built sprawling nationwide branch networks and automated check-processing systems to attract free sight deposits and payment float, immediately investing those balances in high-yielding, inflation-indexed federal treasury bills (Letras Financeiras do Tesouro, or LFTs). ^[2]

Between 1990 and 1993, according to Central Bank and IBGE national accounts data, inflationary float income generated more than 38 percent of Brazilian banks’ total gross revenue and reached roughly 4 percent of Brazil’s gross domestic product. Easy float profits allowed Brazilian banks to support bloated administrative payrolls, tolerate lax credit underwriting, and roll over non-performing corporate and political loans without recognizing losses. Once the Plano Real anchored the exchange rate and eliminated hyperinflation in the second half of 1994, banking sector float revenues collapsed from over 4 percent of GDP to less than 0.2 percent of GDP by 1995. ^[3]

To compensate for the sudden disappearance of float income, Brazilian commercial banks aggressively expanded retail and commercial lending during the second half of 1994—precisely as domestic consumption surged. Alarmed by overheating demand and capital flight following the December 1994 Mexican peso crisis, the Central Bank of Brazil imposed sky-high compulsory reserve requirements (reaching 100 percent on marginal sight deposits and 20 to 30 percent on time deposits) and raised benchmark annualized overnight interest rates above 60 percent in early 1995. Private-sector loan defaults spiked from roughly 7 percent in mid-1994 to over 16 percent by late 1995, pushing Brazil’s most overextended banks to the brink of collapse. ^[4]

The Collapse of Banco Econômico and Banco Nacional’s Ponzi Books#

The first major private casualty of post-inflation banking was Banco Econômico S.A., founded in 1834 in Salvador, Bahia, and controlled by businessman Ângelo Calmon de Sá. As Brazil’s eighth-largest private bank, Econômico had lent heavily to troubled industrial conglomerates and affiliated petrochemical enterprises while relying on high-cost interbank deposits to stay afloat. By mid-1995, Econômico owed more than R$ 3 billion to the Central Bank’s discount liquidity window. On August 11, 1995, the Central Bank of Brazil, under President Gustavo Loyola, placed Banco Econômico under formal intervention, freezing deposits and triggering severe political backlash from powerful Bahian congressional leaders as well as an immediate run by depositors shifting funds out of second-tier private banks into state-owned giant Banco do Brasil. ^[5]

While regulators struggled to resolve the Econômico intervention, an even more dangerous insolvency was festering inside Banco Nacional S.A., Brazil’s sixth-largest private commercial bank, controlled since 1944 by the influential Magalhães Pinto family of Minas Gerais and famous across Brazil for sponsoring Formula One champion Ayrton Senna. Behind its prestigious retail facade, Banco Nacional had actually become insolvent in the late 1980s following heavy corporate credit losses. Rather than writing off the bad loans and recapitalizing the bank, executive vice president Marcos Magalhães Pinto and senior accounting managers operated a massive internal accounting fraud—described in academic literature by economist Fernando de Holanda Barbosa as an institutional Ponzi game—for nearly a decade. ^[6]

To conceal the capital hole from external auditors (KPMG) and Central Bank supervisors, Banco Nacional maintained 652 fictitious or inactive checking accounts assigned to long-defunct small businesses in lesser-known provincial branches. Every month, bank executives booked fictional interest income and newly fabricated loan balances onto these phantom accounts, inflating the bank’s stated assets and generating paper profits out of thin air—even paying corporate income taxes and shareholder dividends on nonexistent earnings. By late 1995, when the loss of inflationary float and interbank distrust forced Banco Nacional to borrow billions from the Central Bank, the cumulative balance of fictitious loans exceeded R$ 5.4 billion (roughly USD $5.6 billion), turning a reported positive net worth of R$ 1.2 billion into a true negative net worth exceeding R$ 4.2 billion. ^[7]

Launch of PROER: The Good-Bank / Bad-Bank Resolution Model#

Recognizing that an disorderly liquidation of Banco Nacional on the heels of the August 1995 Banco Econômico freeze could spark a nationwide banking panic akin to Venezuela’s 1994 crash or Mexico’s Tequila Crisis, President Fernando Henrique Cardoso issued Provisional Measure No. 1,179 on November 3, 1995, establishing the Program of Incentives for the Restructuring and Strengthening of the National Financial System (Programa de Estímulo à Reestruturação e ao Fortalecimento do Sistema Financeiro Nacional, or PROER), alongside the creation of Brazil’s private deposit insurance fund, the Credit Guarantee Fund (Fundo Garantidor de Créditos, or FGC), which insured retail deposits up to R$ 20,000 per account. ^[8]

PROER pioneered a rapid “good-bank / bad-bank” split funded by Central Bank liquidity lines and tax incentives, ensuring that retail branches never closed their doors for a single business day and that controlling shareholders of failed banks lost their entire equity stakes:

• Banco Nacional into Unibanco (November 1995): On Saturday, November 18, 1995, the Central Bank placed Banco Nacional under Special Temporary Administration Regime (RAET), stripped the Magalhães Pinto family of ownership, and over a single weekend transferred Nacional’s healthy retail branch network, brand, deposits, and performing assets (the “good bank”) to rival Unibanco (União de Bancos Brasileiros), backed by roughly R$ 5.9 billion in PROER credit lines. The R$ 5.4 billion hole of fictitious loans and uncollectible claims remained inside the residual “bad bank” under Central Bank liquidation.
• Banco Econômico into Banco Excel (April 1996): After eight months of paralysis, regulators applied the PROER structure to split Banco Econômico, transferring its retail deposits and sound assets to Banco Excel-Econômico (later acquired by Spain’s BBVA in 1998) while retaining the non-performing portfolio in the Central Bank-administered bad bank.
• Banco Bamerindus into HSBC (March 1997): When Curitiba-based Banco Bamerindus—Brazil’s fourth-largest private bank, controlled by Senator José Eduardo de Andrade Vieira—buckled under paper-pulp conglomerate losses and deposit flight, the Central Bank intervened over the Easter weekend of March 1997. Using approximately R$ 5.7 billion in PROER financing and FGC support, authorities transferred Bamerindus’s 1,200 branches and sound operations to British banking group HSBC, which formed HSBC Bamerindus.

Crucially, PROER contained a financial engineering mechanism that kept Central Bank bailout loans heavily collateralized on paper: failed banks were permitted to use PROER credit lines to purchase deeply discounted, defaulted federal obligations—such as Wage Variation Compensation Fund (Fundo de Compensação de Variações Salariais, or FCVS) housing credits and agrarian debt bonds trading at 40 to 50 cents on the dollar in secondary markets—and pledge those federal securities back to the Central Bank at 100 percent of their face value. Across seven major restructurings (including Nacional, Econômico, Bamerindus, Mercantil de Pernambuco, Banorte, Pontual, and Crefisul), PROER disbursed approximately R$ 21 billion (roughly USD $20 billion, or about 2.5 percent of 1996 GDP).

PROES and the Cleanup of Banespa and Brazil’s State-Owned Banks#

While PROER stabilized private commercial banking, an even larger fiscal black hole lay in Brazil’s network of 35 commercial and development banks owned by individual state governments (bancos estaduais). Throughout the 1980s and early 1990s, state governors routinely treated their regional banks as captive money-printing machines, ordering them to extend uncollectible loans to state treasuries and state-owned enterprises during election years. The largest offender was the Bank of the State of São Paulo (Banco do Estado de São Paulo, or Banespa), which alongside Rio de Janeiro’s Banerj was placed under Central Bank special administration (RAET) on December 30, 1994, with Banespa alone nursing a state-debt hole that compounded past R$ 20 billion under 1995 interest rates.

To permanently end the quasi-fiscal abuse of regional banks, the Cardoso administration enacted Provisional Measure No. 1,514 on August 7, 1996, launching the Program of Incentives for the Reduction of the State Public Sector in Banking Activity (Programa de Incentivos à Redução do Setor Público Estadual na Atividade Bancária, or PROES). Under PROES, the federal government offered to finance 100 percent of the cost of cleaning up a state bank’s balance sheet—refinancing the expenditure into 30-year state debt owed to the National Treasury at real interest rates of 6 percent—only if the state government agreed to privatize the bank, liquidate it, transform it into a non-deposit development agency, or transfer ownership to the federal government for subsequent sale. If a state governor insisted on keeping control of a state bank, the federal government financed only 50 percent of the cleanup.

The incentive structure worked decisively. Between 1996 and 2002, out of 35 state-owned financial institutions across Brazil, 10 were privatized directly by states, 6 were federalized and privatized (including Banespa, sold to Spain’s Banco Santander in November 2000 for R$ 7.05 billion), 10 were liquidated or extinguished, and 4 were converted into development agencies, leaving only 5 state banks under regional control. Total federal bond issuance under PROES exceeded R$ 60 billion (roughly 5 to 6 percent of GDP). By the time the January 1999 currency devaluation tested the Brazilian economy, PROER and PROES—combined with the Basel capital standards enacted under Resolution 2,099—had insulated Brazil’s banking system from systemic collapse at a fraction of the GDP cost suffered in Mexico, Argentina, or East Asia.

Theoretical & Balance-Sheet Transmission Mechanism#

In the Kindleberger–Minsky taxonomy of financial instability (Kindleberger, 1978; Minsky, 1986) and Irving Fisher’s (1933) debt-deflation framework, speculative asset booms become systemic crises when credit expansion shifts balance sheets from hedge finance (cash flows covering principal and interest) to speculative and Ponzi finance (relying on rising asset prices and short-term margin roll-overs).

Once monetary tightening or an exogenous shock halts price appreciation, forced margin liquidations trigger Fisherian debt-deflation: fire-sale asset discounts erode bank collateral values, contracting credit supply across the real economy.

Sources and Further Reading

1. Banco Central do Brasil, “ Reestruturação Bancária no Brasil: O Caso do Proer,” Notas Técnicas do Banco Central do Brasil, No. 38, 2003-06. Official / State Archive
2. Banco Central do Brasil, “ PROER – Program of Incentives to the Restructuring and Strengthening of the National Financial System,” Banco Central do Brasil Historical Archive, 1996. Official / State Archive
3. Fernando de Holanda Barbosa, “ Banco Nacional: jogo de Ponzi, PROER e FCVS,” Brazilian Journal of Political Economy, Vol. 28, No. 1, 2008-03. Peer-Reviewed / Academic Research
4. Fernando Pimentel Puga, “ Sistema financeiro brasileiro: reestruturação recente, comparações internacionais e vulnerabilidade à crise cambial,” Banco Nacional de Desenvolvimento Econômico e Social (BNDES), Textos para Discussão No. 68, 1999. Official / State Archive
5. Roberto Luis Troster, “ Price Stabilization, the Banking Crisis and Financial Reform in Brazil,” The International Handbook on Financial Reform, Edward Elgar Publishing, 2003. Peer-Reviewed / Academic Research
6. Luis Otávio de Souza Leal, “ O PROES e a Eficiência dos Bancos Estaduais,” Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio), 2001. Peer-Reviewed / Academic Research
7. Irving Fisher — "The Debt-Deflation Theory of Great Depressions," Econometrica 1, no. 4 (FRASER Federal Reserve Archive, 1933): 337–357 Central Bank / Monetary Authority
8. Charles P. Kindleberger & Robert Z. Aliber — Manias, Panics, and Crashes: A History of Financial Crises (Palgrave Macmillan) Peer-Reviewed / Academic Research

Have an idea for another financial history documentary or article? Request a story →