Still the century of government savings banks? The Caixa Econômica Federal

ARTICLE

Still the century of government savings banks? The Caixa Econômica Federal

Kurt von Mettenheim

Professor da Escola de Administração de Empresas de São Paulo da Fundação Getulio Vargas. E-mail: [e-mail]

ABSTRACT

This article explores general concerns about government banking, social inclusion, and democracy through case study of the Brazilian federal government savings bank (Caixa Econômica Federal). Review of government savings banks in Brazilian history suggests that these institutions have been at the center of domestic political economy, expanding and contracting under a variety of political regimes and economic conditions. Since capitalization to meet central bank and Basel Accord guidelines in 2001, the Caixa has attempted to modernize, continue to serve as agent for government policies, and expand both popular credit and savings and investment banking activities.

Key-words: government banks; savings banks; bank policy; Caixa Econômica Federal; Brazilian federal financial institutions.

JEL Classification: E5; G2.

"The function of a savings bank, in fact, is not to serve as an institution for investing money. Its business is to enable people to put money aside and even to build up a little capital. But when this capital has been formed, if the depositors wish to invest it — that is to say, to make a profitable use of it — they have merely to withdraw it: the rôle of the savings bank is ended, and it rests with other institutions such as we have already studied in dealing with banks and credit establishments, to take charge of it."

Charles Gide, Principles of Political Economy, London: George G. Harrap & Co. Ltd. 1924, p. 510 (1st ed. 1906)

The original intent of savings banks was to teach popular classes the habit of saving, increase the liquidity of capital, and spur economic growth.¹ Government savings banks were founded in France (1818), Austria (1819), and several German states (1835) while government guarantees provided incentives for a variety of private, community, cooperative, and mutual savings banks in the Netherlands, England, Italy, and the United States. In 1906, almost a century after the Caisse d'Epargne was founded in France, Charles Gide argued that savings banks should serve a limited role, comparable to piggy banks in terms of collecting small amounts of capital and savings. What happened? The first Brazilian government savings bank (Caixa Econômica e Monte de Socorro) was founded in 1860. If savings banks are temporary and limited in character, why are an estimated 70% of Brazilians still sem conta, that is to say bankless, without checking accounts, savings deposits, or bank cards? If other banks and credit institutions are more efficient, why do government savings banks still provide roughly 20% of credit in Brazil and retain over 20% of savings deposits? And why, after two decades of a Washington Consensus about the need to privatize state firms and liberalize (especially financial) industries, is the Caixa Econômica Federal still the third largest bank in Brazil (depending on measure)? In June 2004, the Caixa maintained 60,402 employees, 2013 branches, more than R$ 82 billion in deposits, and over R$ 166 billion in assets.²

Like corporatism, government savings banks appear to have outlived their predicted usefulness on the road to developed markets, societies, and polities.³ Are these institutions still necessary to counter the imperfections of markets and private banking? Alternatively, are critics of government banking right that these institutions cause financial repression and reproduce underdevelopment? For La Porta and theories of financial repression in the tradition of Gurley & Shaw, it is precisely the continued presence of government credit that crowds out more efficient allocation of resources through markets.⁴ These opposing theories about government banking frame this case study.

This article reviews both government savings banks in Brazilian history and the reorientation of the Caixa since capitalization in 2001. Secondary accounts and primary data suggest that government savings banks expanded under Empire from 1860-1889 but declined during the economic liberalism of the Old Republic (1889-1930). After the 1930 revolution, government savings banks became central to national populist strategies of capital mobilization and industrialization. Balance sheet problems during the late 1950s appear to be due primarily to the impact of rising inflation that led depositors to withdraw funds from savings accounts (earning fixed 6% annual returns), while administrative costs also suggest management errors. After the breakdown of democracy, the indexation of savings led to a recovery of deposits at government savings banks while reforms in 1970 centralized control over bank policies. During the 1980s, the Caixa dramatically increased its market share of domestic lending to public and private sectors, suggesting that high inflation and macroeconomic instability left the bank as one of the few major lenders. Since costly capitalization of the bank in 2001, the Caixa continues to focus on core business in housing, urban development, sanitation, and as agent for government social policies. Since 2001, a sanitized portfolio combining high interest bearing government paper and low interest savings deposits has helped produce strong profits,⁵ permitting the bank to pursue a dual strategy of expanding both investment bank operations and new popular credit and savings programs. The Caixa thereby provides a critical case study for the ability of governments to encourage popular savings without crowding out market forces or succumbing to the politicization of credit. This article is necessarily exploratory because no academic study of this institution or its predecessors appears available since manuscripts left by administrators of the bank decades ago.⁶

This case study is also constrained by the need for broader reassessment of banking in Brazil. Liberalization, privatization, and a wave of direct foreign investment during the 1990s have produced a new Brazilian financial system. However, far from confirming assumptions about the superiority of foreign banks (or similar fears of imperialist domination) both the content and consequences of reforms during the 1990s differed in important respects from neo-liberal theory and policy. In 1995, the Cardoso government reduced protection of domestic finance set in the 1988 Constitution, permitted foreign participation in privatization auctions, and provided new incentives for foreign investment in financial industries. From 1994 through 2002, direct foreign investment in the financial sector summed to US$ 19.8 billion (15% of total DFI). Meanwhile, the adjustment of banks to the end of high inflation required government programs to save and sell failed private banks (PROER, 1995-97), to privatize deeply indebted state government banks (PROES, 1997-99), and to capitalize and reform federal government banks (PROEF, 1999-2001). The Central Bank of Brazil also introduced a variety of new initiatives to improve the supervision and monitoring of banks and credit risk.⁷

Banking itself has also changed dramatically, especially in terms of substantial cost savings through electronic card technology, the introduction of automated teller machines (ATM's), and electronic transfers. Paradoxically, Brazilian banks modernized quickly during the late 1980s and 1990s because of large gains under high inflation. As Goldfajn et al note, strong earnings from investing consumer deposits in indexed instruments: "induced banks to expand, open new branches, offer 'free' bank services and develop a high degree of technological progress, especially aimed at enhancing the speed of transactions".⁸ This trajectory of banking before 1994 helps explain the different impact of financial liberalization thereafter: the modernization of Brazilian banks — private and public — occurred not because of financial liberalization but prior to it during the prolonged period of high inflation and indexation during the 1980s and early 1990s. Timing matters because seemingly minor differences can create enduring comparative advantage.⁹

Bank change sets the context for this study, but the primary focus is on the political economy of savings banks and social inclusion. This article explores the viability of shifting credit and banking policies toward a fundamentally new direction that accelerates social inclusion by basing policies on core concepts of social justice and citizenship as well as market utility and property rights. Recent studies of micro-credit suggest that risks associated with popular credit differ from risks involving middle class groups, large economic interests, and political machines. Commercial banking in Brazil and abroad has found that higher default rates in popular credit are more than compensated by higher profit margins and the greater pulverization of risk across a larger number of clients. From this perspective, new programs of popular credit appear more viable and less prone to default than past experiences with program lending, subsidized middle class housing loans, and large loans to select enterprises that have haunted federal bank balance sheets. These broader concerns about banking are at the center of public policy debates and economics. Since the landmark contribution of Stiglitz & Weiss, economists have attempted to understand the impact of private bank lending practices on social exclusion in advanced economies.¹⁰ In 2001, the US Senate Banking Committee commissioned inquiry into predatory lending practices while, in 2004, the American Political Science Association commissioned a task force to study the causes and consequences of increasing inequality.¹¹

These matters cross the disciplines of political economy, political sociology, political theory, financial economics, and banking studies. However, far from proposing radical reform, this article is based on a gradualist conception of politics that emphasizes the opportunities for building on existing social policies and institutions. Given the creation of new social policies since transition from military rule in Brazil, and new realities of information technology, banking cards, and well organized data bases in government agencies, the gradual expansion of existing programs of popular credit and grants toward broader policies of basic income seems compelling in terms of social justice, sound in terms of economic theory, and essential in terms of political socialization.¹²

This study thereby combines a historical-institutional approach with concerns of empirical democratic theory. Historical institutionalism explores compelling problems about politics and society through analysis of institutions (such as the Caixa) across time.¹³ Empirical democratic theory focuses on how social classes are, or are not, incorporated into political institutions during modernization.¹⁴ Although often perceived as conservative or elite oriented, empirical democratic theory draws attention to the unfinished business of popular incorporation in Brazil. It is precisely the dramatic distribution of Brazilians in terms of IBGE social classes — more simply the persistence of poverty — that suggests the importance of institutions such as the Caixa as agents for social inclusion. These matters converge on questions about democracy in political sociology because of the reality that social classes C, D, and E form a large majority of Brazilian voters. In sum, if statecraft shall continue to deepen the attachment of Brazilians to representative government, democracy, and markets, then more dramatic policies of money, credit, and income are necessary. In this respect, the Caixa (and other federal government banks) appear to provide considerable comparative institutional advantage for deepening financial markets, citizenship, and democracy in Brazil.

GOVERNMENT SAVINGS BANKS IN BRAZILIAN HISTORY

The first Brazilian government savings banks were modeled after European banks, especially the French government savings bank, Caisse d'Epargne.¹⁵ Private savings banks appear to have been created as early as 1831¹⁶ to encourage the gradual accumulation of savings among laborers.¹⁷ Thirteen private and public savings banks appear to have been created during the Empire.¹⁸ The Caixa Econômica e Monte de Socorro was founded in 1861 as part of a series of provisions concerning banks of issue, money, and regulation of financial institutions set in Decree 1083 of August 1860. The impact of this decree has been widely debated, but it set three simple rules for government savings banks: Deposits were limited to Cr$ 4:000 per week; total deposits were limited to Cr$ 50:000; and interest on savings was fixed at 6% per year. In 1861, the government guaranteed Caixa deposits, subordinated these banks to the Finance Ministry, and described their mission as: "( ) responsibility for receiving deposits of popular savings and capital reserves across the Brazilian territory to increase their liquidity, encourage saving habits, and facilitate the development and circulation of wealth".¹⁹ Government refusal to increase the Cr$ 4:000 weekly and Cr$ 50:000 total limits to deposits suggests that popular savings remained the core business of Caixas.²⁰

The fate of Caixas during the Old Republic (18ß89-1930) appears to confirm the limited role of these institutions granted by economic liberalism. Although data is short, the Caixas appear to have "returned to a corner of some public agency" during this period of oligarchic machine politics and orthodox economic policies.²¹ Legislation passed in 1915 apparently increased the limit for weekly deposits and loans (after a period of high inflation) and freed the bank to conduct other types of financial business. In sum, bank change and policy debates appear not to fit cleanly into the standard chronologies of Brazilian history. Debates during the Old Republic led to measures after the 1930 revolution designed to expand public savings banks and credits. Furthermore, these proposals appear to have been developed during the 1920s by Alfredo Rocha and Leopoldo de Bulhões, economists traditionally associated with more orthodox liberal ideas about economic development.

CAIXAS ECONÔMICAS AND NATIONAL POPULISM

After the 1930 revolution, government savings banks were perceived as critical agents for social inclusion, the mobilization of savings, and economic growth. Caixa president Solano Carneiro da Cunha sought to streamline administration and expand services. In 1934, savings banks were brought under new administrative and executive councils, new financial services were launched such as mortgage credits and loans to state and municipal governments (credits to the federal government date from 1915), while the Caixas won both exemption from taxes levied on commercial bank transactions and official monopoly on pawn services. The bank developed guidelines for "independent" and "associated" agencies, set new procedures for administration, staff, capital reserves, and loan approval, and created promotional campaigns to encourage citizens to open savings accounts.²² From 1934 to 1940, Caixas increased share of domestic credit markets from 6.0 % to 10.5 %, with loans increasing from Cr$ 444.912.000 to Cr$ 1.372.698.000. In sum, instead of the more limited and temporary role envisioned by Gide, policies after the 1930 revolution sought to redefine the Caixas as more permanent institutions capable of both encouraging popular savings and providing social assistance.

From the perspective of national-developmentalism, government savings banks were seen as institutions capable of bringing popular savings into circulation to increase the pace of economic growth through credit and investment. De Placido e Silva reports twenty-fold (nominal) increases in the value of savings deposits at Caixas after the 1930 revolution.²³ Furthermore, Caixa balance sheets suggest the focus of these institutions remained on urban development and low-income home loans. In 1936, of the 1623 housing loans reported by the Rio de Janeiro Caixa, 1050 were at the minimum value of 50 Contos, while 306 loans were made between 50-100 Contos, and 112 loans between 100-200 Contos.²⁴

CAIXAS ECONÔMICAS AND DEMOCRACY, 1945-1964

Understanding the trajectory of savings banks after 1945 is critical given the negative perception of government banking in the systems of patronage, clientelism, corporatism, and populism that are associated with the period of competitive electoral politics that ended in military coup on 31 March 1964. The first characteristic of government savings banks from 1945-1964 of note is their sheer size. Although their share of total domestic bank deposits declined from peaks during the 1930s and 1940s, deposits at the Caixas during the 1950s still summed to roughly one third of the paper money circulating in the Brazilian economy. The value of deposits at government savings banks as a percentage of total domestic money supply from 1934 through 1959 reported in Table 1 suggests that these institutions remained at the center of Brazilian political economy during the post-war period of democracy. Deposits at government savings banks peaked during the national-populist regime during the 1930s at almost 60% of the total value of paper money reported by the IBGE.

Caixa market share of bank deposits also remains at significant levels throughout this period. The share of domestic commercial bank deposits held at government savings banks increased from under 20% to over 30% during the early 1930s. After declining during the early 1940s, the market share of domestic commercial bank deposits at government savings banks rose once again during the late 1940s. This second period of increasing market share occurred after transition from the Estado Novo and return to democratic government. This suggests that the evolution of public savings banks in Brazil is not simply correlated with particular political regimes. Indeed, both the increase of deposits after transition from the Vargas regime in 1945 and the steady decline of the domestic share of bank deposits during the 1950s suggest that other factors (such as inflation) are at work. Given that inflation increased from 2.6% in 1947 to between 12 and 25% during the mid-1950s, and peaked at an estimated 91.9% during 1964, the flight of clients from savings accounts with fixed annual returns of 6% is understandable.

The scale of government savings banks can also be seen in their extensive branch network that provides both significant competitive advantages and greater costs than commercial competitors. The study of federal government financial institutions comissioned by the Finance Ministry in 2000 argued that Caixa branches unnecessarily duplicated Banco do Brasil branches.²⁵ However, in historical perspective, the Caixa appears to have brought credit and financial services to regions of the country poorly served by private banking.²⁶ Data from the 1959 Annual Report of Caixas suggests that a total of 483 agencies were operating in Brazil, with 226 standard agencies, 149 postal agencies, 100 affiliated agencies, and 8 correspondent agencies across twenty Brazilian states. Data from the 1969 Annual Report suggests that agencies reached the Brazilian interior, notoriously lacking in banking services. Of the total number of 514 agencies, 163 were located in capital cities, while 351 were located in the interior of states (population breakdown not estimated). Terms for creating Caixa Econômica Federal Postal Agencies were also set in November 1948 (Decree 25.733), echoing initiatives both past initiatives during the 19^th century mentioned above and the government concession in 2001 to place ATM's in postal offices.

This organizational expansion appears to have pressured the balance sheets of government savings banks. Administrative costs increased above inflation during the late 1950s, especially personnel costs. From 1955 to 1959, administrative costs increased from R$ 739.662.000 to over Cr$2.3 billion, while personnel costs increased from Cr$ 613.783.000 to Cr$ 1.9 billion. The latter increase of 284 percent outpaces inflation during this period (228.3%). The dual impact of flight from deposits and rising administrative costs first eroded returns then required cash infusions from the federal government. From 1955 through 1959, returns decreased from 0.8% of deposits to 0.1% of deposits. Upon separation from the Delegacias Fiscais in 1946, the sum of the Contas Patrimoniais or reserves set aside by savings banks totaled Cr$ 274 million. Reserves increased to Cr$ 1.17 billion in 1957 before being consumed by deficits. By 1959, thirteen of twenty-one Caixas Econômicas Federais had consumed their reserves and required cash infusions from the federal government. In 1959, deficits at thirteen state banks totaled Cr$480 million (while eight Caixas supplied a Cr$ 1.39 billion surplus, the São Paulo state branches alone reporting a surplus of Cr$ 116 million for 1959).

GOVERNMENT SAVINGS BANKS UNDER MILITARY RULE AND TRANSITION

The Caixas Econômicas Federais were transformed into a single federal savings bank in March 1970, the Superior Council was eliminated, and the bank was redefined as an agent of federal government credit policy directly responsible to economic ministries. After 1970, the Caixa also became responsible for the implementation of the official savings program (Programa de Integração Social, Program for Social Integration, PIS) and social policy fund (Fundo de Assistência Social, Social Assistance Fund, FAS) under the Social Development Council. The bank was also granted a monopoly over lotteries, became the agent for distribution of educational loans, and administered the concession and execution of a variety of domestic housing, sanitation, and infrastructure credits to state and municipal governments, many based on FGTS savings.

The trend of federal government savings bank lending to the public sector from 1968-2003 appears to reflect the rise and fall of the bank as lender of last resort to government agencies and programs under high inflation and macroeconomic instability (See Table 3). Indeed, Caixa market share peaks at half of all loans to the public sector during the penultimate surge of high inflation in 1989. Thereafter, the market share of the Caixa in loans to the public sector declines to well under 10% of the total loans to the public sector by 2001.

A similar pattern in the market share of the Caixa in loans to the private sector appears from 1968-2003 (See Table 4). From 1968 to 1975, the Caixa's share of domestic loans to the private sector doubles from eight percent to 19%, then rises to a full 50% of loans to the private sector during 1989. Caixa market share declines dramatically thereafter during the 1990s to 6% of total private sector lending in 2001.

In sum, the indexation of savings against inflation and the centralization of government savings banks under military government were followed by the predominance of these financial institutions in domestic credit markets. The prolonged period of high inflation and the profound disorganization of prices by failed economic packages during the late 1980s left the Caixa responsible for half of domestic lending to public and private sector by 1989. Far from crowding out private lending, the erosion of credit and finance appear to have left the Caixa as one of the few lenders of last resort. Since 1990, and especially after price stability in 1994, the Caixa's market share of domestic lending to both the private and public sector declined sharply.

After transition from military rule in 1985, Brazilian government savings banks also acquired new portfolios and new roles. With the extinction of the National Housing Bank (Banco Nacional de Habitação, BNH) in 1986, the Caixa acquired the BNH's considerable (and considerably problematic) portfolio of housing loans. The Caixa also assumed responsibility for administration of the social security fund Fundo de Garantia de Tempo de Serviço (FGTS). By 1993, the Caixa had unified of over 130 million FGTS accounts once dispersed across 76 banks, eliminating over 72 million inactive accounts. In sum, the mission of government savings banks to serve as agent for federal government social policies presents significantly different costs and opportunities compared to commercial competitors.

2001: CAPITALIZATION AND REORIENTATION OF THE CAIXA

In June 2001, Finance Minister Pedro Malan announced a refinancing scheme for federal government banks that injected R$ 12.5 billion to capitalize these institutions sufficiently to meet both Basel Accord capital guidelines and Central Bank of Brazil resolution 2.682/99 that increased Basle Index levels from 8.0 to 11.0. In 2001, the total government infusion of capital into the Caixa was estimated to reach R$ 86.7 billion, although final costs to treasury depend on a variety of matters involving law, accounting, budgets, and asset transfers:

24.0 billion R$ to exchange government paper

5.7 billion R$ to shed non-performing BNH loans

13.0 billion R$ to write off state government debts

8.0 billion R$ to purchase the Caixa's debt with the FGTS

9.3 billion R$ for capitalization

26.7 billion R$ for transferring risk from housing loans

86.7 billion R$ total estimate

Was it worth it? Has the Caixa pursued sound banking strategies and effective methods of policy implementation? Alternatively, are critics of government banking correct to suggest that subsidized credit and poor supervision crowd out more efficient private lending?

In 2001, four reforms attempted to guarantee the continuity of the Caixa as agent of government policies while ensuring against the recurrence of large liabilities. First, a new administrative model was introduced to avert abuses of credit and ensure banking prudence. Second, greater transparency was adopted for distribution of government mandated services, grants, and social programs. Third, a series of measures exchanged assets and restructured bank capital. Finally, non-performing assets and housing loans inherited from the BNH were sold to a specially created independent financial agency of the federal government. Since 2001, Caixa executives have pursued a strategic reorientation designed to reinforce its core business and role as agent for government social policies while promoting new programs of popular credit and investment banking. Persistently high interest rates since 2001 have provided further time to modernize because of the large spread between low interest payed on Caixa savings deposits and high interest bearing government bonds. While it remains difficult to assess the longer-term prospects for the new business strategies adopted by the Caixa, several observations are in order.

Review of balance sheets and policies since 2001 suggest that the Caixa retains its core business in urban development, home loans, real estate, transfer of funds for government programs and services, and lottery administration. However, investment banking, management of third party funds, and the creation of of new products and services have both increased profits and, apparently, increased popular access to banking and social services. The bank recorded net profits of R$ 1.08 billion during 2002, R$ 1.6 billion during 2003, and R$ 1.419 during 2004. The Caixa grew from the seventh to fourth largest domestic investment bank from 2001-2003, while gaining dealer status from the Central Bank of Brazil in primary and secondary markets for government securities. At year-end 2003, the bank retained an estimated R$ 76.7 billion of government paper (over 10% of government paper) in its portfolio, assets earning over R$ 13.5 billion that year.

Unlike practices under military rule, financial statements since 2001 suggest that the bank lends very little to the public sector. Total credits to government agencies — including government industries and services — summed to a modest 5.2% and 6.0% of total loans at the end of first semester 2003 and 2004 respectively. Meanwhile, loans to the private sector sum to 93.9 percent of total Caixa credits. A full 56.9% of loans as of June 2004 were for home construction or purchase, 21.6% were to private firms, and 11.7% to individuals.

Comparison of the Caixa with the largest ten financial institutions in Brazil at year-end 2003 reveals its second largest share of bank deposits with R$81.0 billion, sixth largest share of credits at R$ 21.8 billion, and fourth largest share of assets at R$ 150.5 billion. Caixa returns on assets (28%) and leverage (25.0) during 2003 were higher that any other top-ten bank. The efficiency rating of the Caixa (0.59) was lower than Banco do Brasil (0.73), Bradesco (0.64), and ABN Amro (0.67), slightly higher than Santander (0.56), Unibanco (0.54), and Nossa Caixa (0.54), and significantly higher that Itaú (0.48) and Safra (0.39). These measures suggest that the Caixa appears to allocate resources as effectively and efficiently, on average, as private commercial banks operating in Brazil. This belies theories that see government banks as grabbing hands that cause financial repression.

The second dimension of strategic reorientation of the Caixa involved the development of new programs of popular and micro-credit. In 2002, the bank launched a new account for popular savings and banking-services, CAIXA Aqui, that requires neither minimum deposit, nor proof of income, nor proof of residence. After three months, clients become eligible for loans up to R$ 200.00 at interest rates of 2% per month. By February 2004, 1.27 million new accounts had been opened; by June 2004 accounts surpassed two million.

The Caixa has also continued to distribute federal government benefits since 2001. The number and value of benefits granted by the federal government through the Caixa is reported in Table 6. Income transfers, social security payments, official savings programs (PIS), bonuses, and unemployment insurance payments total over 243 million transactions and R$ 19.4 million during 2002, and 299 million transactions and R$ 24.5 million during 2003.

The Caixa also remains a central agent for payment of federal government grants, especially since a variety of social policies and transactions were consolidated in Citizenship Cards in July 2003. During 2003, over 172 million transactions involving R$ 2.1 million were distributed by the Caixa for programs such as Young Agent, student allowances, gas assistance, food allowances, family and income allowances, crop guarantees, and programs to eradicate child labor (down from 225.9 million transactions valued at R$ 3.1 million during 2002). Financial statements suggests that an additional 4.3 million cards were issued during the first six months of 2004, 3.5 million of which are proprietary Caixa-Citizenship Cards, with the additional cards being either federal government Citizenship Cards or school grant Citizenship Cards.

A consequence of the Caixa's role as agent for federal government social policy is that the bank retains perhaps the largest data base on social policy in Brazil. During 2003 and 1st semester 2004, the Caixa registered 3.6 million households, increasing the total number of households in the data base to 9.2 million. A separate registry of 4.3 million households from the school grants program was also added to the Caixa's social services data set, increasing the grand total of households to over 13 million in June 2004. While President Silva created two new ministerial level posts and a variety of programs to expand social policies, the Caixa remains at the center of new social policies because it retains important data bases and an extenstive national network.

Data from the Central Bank of Brazil and Caixa annual reports suggest that the federal government savings bank retains one of the largest national networks of branches and ATM's, while its lottery monopoly appears to have provided a comparative advantage for expanding banking services to the interior and less developed regions of Brazil. At year-end 2003, the Caixa reported 2126 branches, 1.080 electronic outlets, 8.922 lottery stores, 2.108 corresponding establishments, and 1,966 ATM's. While the Caixa retains the goal of creating agencies in all 5,561 Brazilian municipalities through its existing agreement with the federal lottery system, it should be noted that the Brazilian private bank Bradesco won a government concession to provide banking services inside all existing 5,500 agencies of the postal system.

The organizational structure of the Caixas today reflects broader trends in Brazilian banking away from branches and min-branches toward ATM's and the use of correspondent institutions to provide services (See Table 8). The total number of bank branches in Brazil from 1994-2004 has declined from 17.400 to 17.049, while the number of mini-branches has declined from 10.125 to 7.108. In comparison, the total number of ATM's increased from 3.446 in 1994 to 22.428 in 2002. Furthermore, instead of four different types of outpost arrangements in place during the 1990s, the creation (and judicial challenge) of correspondent banking through shops and lottery outlets increased dramatically the number of banking service points after 2000. In sum, the Caixa retains a large network reaching across the Brazilian territory that appears to provide significant comparative advantage, but also threatens to erode earnings because of higher administrative costs and unnecessary duplication with other government bank networks.

Before concluding, a summary of the historical evolution of the Caixas Econômicas Federais is in order. From 1861-1889, a variety of government incentives led to the growth of the public savings bank under monarchy. From 1890-1930, more orthodox policies of economic liberalism led to the decline of government savings banks. From 1930 to 1945, the national populist regime of Vargas adopted new economic policies designed to mobilize domestic capital and industrialize through the substitution of imports. Government banks were central to these strategies of popular inclusion and state-led development. During the 1950s, rising inflation eroded deposits while increasing costs reduced earnings and reserves. After military intervention in 1964, both the indexation of savings (1965) and the reorientation of the Caixa in 1970 led to an expansion of the bank as agent for public and private credit. During the 1980s, the perverse combination of high inflation and political vacuum during the prolonged transition from military rule appears to have left government savings banks as one of the few sources of domestic credit. Price stability in 1994 forced the Caixa, along with other banks, to adjust to the end of high inflation, while new regulations against capital risk set by the Central Bank of Brazil decreased the Caixa's share of domestic credit markets. Since capitalization in June 2001, the Caixa has pursued a dual reorientation toward investment banking and the expansion of popular credit and savings accounts.

This historical summary begets a central question: does capital flow to or from popular classes through government savings banks? Since Gide, the liberal tradition of political economy suggests that popular savings banks permit the accumulation of wealth among those most needing it. More critical perspectives and the marxist tradition would suggest that savings banks tend to extract value from popular classes. Although further analysis is needed, the initial conclusion from this historical overview is that the fortunes of popular (and other social classes) did indeed change dramatically across Brazilian history. Large gains and losses under the more volatile conditions of dependent development mark the Brazilian experience with markets and government intervention. But the causal process is not uniform. Sometimes value appears to be added in accord with political-economic reforms that favor popular inclusion; sometimes value appears to be dramatically eroded by inflation; and sometimes savings appear to have expanded simply because economic growth creates virtuous cycles.

CONCLUSION

Social scientists and policy makers differ fundamentally about government savings banks. Theories of financial repression see government presence in banking and credit markets as responsible for reproducing underdevelopment, and privatization and liberalization as policies necessary to free market forces and better allocate resources. Theories of comparative institutional advantage, relational banking, and the commanding heights approach suggest that long-term relations between banks, communities, political forces, and firms large and small are necessary to realize the gains of investments and social policies, especially in late development.²⁷ That these relations be sheltered from market forces is at odds with core ideas about optimal equilibria in financial economics. Advocates of markets suggest that liquidity, transparency, competition, and market pricing produce higher levels of welfare. In opposition, advocates of bank-centered development, government coordination, and late development argue that patient capital, face-to-face networks, social policies, government intervention, and institutions, in the long-term, ensure higher rates of growth and welfare.

The sheer size of Brazilian government banks and the diametrically opposed views of these credit institutions suggest that careful reassessment of assumptions about state, society, financial markets, politics, banking, and development is needed. Similar reassessments are underway abroad. For example, instead of liberalization and privatization, large emerging economies such as China, India, and Russia have retained both government and domestic control over banking. National experiences in Europe during financial liberalization and monetary union also provide a rich variety of new comparative references. Indeed, local, regional, non-profit, and government owned credit institutions appear to have maintained or increased their substantial market shares in many European countries, contrary to expectations that big private banks would predominate. Once again, Europeans appear able to craft varieties of capitalism that retain public banks and social policies at their center. Although Brazil may lack the features and policies of other national and regional experiences, this case study suggests that government banks such as the Caixa Econômica Federal may provide significant comparative advantages for new policies of economic development, social inclusion, and democratization.

The severe fiscal constraints on Brazilian government spending and the shallowness of domestic bond markets make this question even more critical. Analysts often compare government bank performance unfavorably with private commercial banks. However, in terms of public policy, government banks can do more for less: Almost ten times more if one compares cash used as capital reserves by banks to other policies that require budgetary outflows. Central Bank of Brazil regulations require 11% (weighted) capital reserves against credit risk. This implies that the Caixa (and other government banks) can loan almost ten times whatever profits are retained or funds may be allocated by congress. From this perspective, and given the evidence explored in this case study, the Caixa appears uniquely positioned to provide social services and extend credit to those left behind during Brazilian development. Tapping the popular credit channel may accelerate social inclusion and economic development, deepen the Brazilian financial system, and provide substance to citizenship and democracy.

Submetido: dezembro 2004, aceito: Maio 2005.

• 3 The reader will note the title of this paper paraphrases: Schmitter, Philippe."Still the Century of Corporatism?" Review of Politics. Vol. 36, (1974), pp. 85-131[
• STANDARDIZEDENDPARAG]4 La Porta, Rafael (et al). "Government Ownership of Banks." Journal of Finance, Vol. 57, no. 1 (2002), pp. 265-301,
• J. Gurley & and E. Shaw "Financial aspects of economic development", American Economic Review, Sept. 1955, pp. 515-38[
• 6 Souza, Henrique A. O Crédito Imobiliário da CEF. Masters Dissertation, Escola de Pós-Graduação em Economia, Fundação Getulio Vargas, 1992,
• Lyra Filho, O Estado Monetário Rio de Janeiro, Irmãos Pinotti, 1948,
• Henrique, João. Estrutura e Conjuntura das Caixas Econômicas Federais Rio de Janeiro: Conselho Superior das Caixas Econômicas Federais, 1960
• and De Placido e Silva, As Caixas Economicas Federais: Sua História, Seu Conceito Jurídico, Sua Organização, Sua Administração e Operações Autorisadas Curitiba: Empresa Gráfica Paranaense, 1937.
• 7 Study of the Central Bank of Brazil is beyond the scope of this study. See: Whitehead, Laurence and Lourdes Sola, eds. Statecrafting Monetary Authority: Democracy and Financial Order in Brazil. Oxford: Center for Brazilian Studies (Forthcoming, 2005).
• 8 Goldfajn, Ilan (et al). "Brazil's Financial System: Resilience to Shocks, no Currency Substitution, but Struggling to Promote Growth", Central Bank of Brazil Working Paper, 75, June, 2003, p. 5.
• 9 Zysman, John. "How Institutions Create Historically-Rooted Trajectories of Growth",Industrial and Corporate Change, Vol. 3, (1994), no. 1, pp. 243-83[
• STANDARDIZEDENDPARAG]10 Stiglitz, Joseph and Andrew Weiss, "Credit Rationing in Markets with Imperfect Information", American Economic Review, (1981), Vol. 71, pp. 353-76.
• 11 Jacobs, Laurence et al. American Democracy in an Age of Rising Inequality. Washington, DC: American Political Science Association, 2004, available at www.apsanet.org.
• 12 Van Parijs, Philippe. "Basic Income: A Simple and Powerful Idea for the Twenty-First Century". Politics & Society, (March 2004), Vol. 32, pp. 7-39,
• Pateman, Carole. "Freedom and Democratization: Why Basic Income is to be Preferred to Basic Capital". In: Dowding, Kieth, et.al (eds.) The Ethics of Stakeholding London: Palgrove, 2003
• and Barry, Brian. "Real Freedom and Basic Income." The Journal of Political Philosophy Vol. 4, (1996), pp. 242-76,
• Suplicy, Eduardo M. "Renda Básica: A Resposta está sendo soprado pelo vento," Brazilian Journal of Political Economy, (2003), Vol. 23, no. 2/90, pp. 47-62.
• 13 Mahoney, James and Dieter Reuschemeyer, eds. Comparative Historical Analysis in the Social Sciences. Cambridge: Cambridge University Press, 2003.
• 14 Dahl, Robert Alan. Polyarchy. New Haven; London: Yale University Press, 1971,
• Huntington, Samuel P. Political Order in Changing Societies New Haven, CT: Yale University Press, 1968.
• 15 On savings banks in Europe, see: Butzbach, Olivier. "Varieties within Capitalism? A Comparison of French and Italian Savings Banks, 1980-2000", Paper presented to the Society for the Advancement of Socio-Economics, Aix-em-Provence, June 2003.
• 16 Founded by José Florindo de Figueiredo Rocha, according to Rocha, Alfredo. As Caixas Econômicas e o Crédito Agrícola, Rio de Janeiro, 1905.
• 17 José Francisco Sigaud, O Homem Benfazejo ou das Vantagens que Resultam da Fundação da Caixa Econômica dos Povos Civilizados. Rio de Janeiro: Tipografia Imperial e Constitucional de Seignot-Plancher & Cia. Vol. XII, Biblioteca Constitucional do Cidadão Brasileiro, 1832.
• 25 Booz-Allen & Hamilton - Fipe (Consortium). Instituições Financeiras Públicas Federais - Alternativas para a Reorientação Estratégica - Public Hearing. Brasília, DF: Ministry of Finance, June, 2000[
• 27 Hall, Peter A. and David W. Soskice, eds. Varieties of capitalism: The Institutional Foundations of Comparative Advantage. Oxford: Oxford University Press, 2001.

1

The author thanks Lourdes Sola, Maria Rita Loureiro Durand, Eduardo Kugelmas, Maria Antonieta del Tedesco Lins, Laurence Whitehead, Leslie Bethell, Valpy Fitzgerald, Peter Evans, Mariano Laplane, Jenny Corbett, and faculty and students of both the Escola de Administração de Empresas de São Paulo - Fundação Getulio Vargas and University of Oxford Centre for Brazilian Studies for comments. Suggestions by an anonymous reviewer of the REP were especially helpful. Financial support for research was provided by the Núcleo de Pesquisas e Publicações of the EAESP-FGV and the University of Oxford Centre for Brazilian Studies. André Carvalho and Tatiana Benítez provided research assistance.

2

Caixa Econômica Federal, "Demonstrações Contábeis", June 2004, p. 2

3

The reader will note the title of this paper paraphrases: Schmitter, Philippe."Still the Century of Corporatism?"

Review of Politics. Vol. 36, (1974), pp. 85-131[ STANDARDIZEDENDPARAG]

4

La Porta, Rafael (et al). "Government Ownership of Banks."

Journal of Finance, Vol. 57, no. 1 (2002), pp. 265-301, J. Gurley & and E. Shaw "Financial aspects of economic development",

American Economic Review, Sept. 1955, pp. 515-38[ STANDARDIZEDENDPARAG]

5

I thank a REP reviewer for this reading of Caixa balance sheets.

6

Souza, Henrique A.

O Crédito Imobiliário da CEF. Masters Dissertation, Escola de Pós-Graduação em Economia, Fundação Getulio Vargas, 1992, Lyra Filho,

O Estado Monetário. Rio de Janeiro, Irmãos Pinotti, 1948, Henrique, João.

Estrutura e Conjuntura das Caixas Econômicas Federais. Rio de Janeiro: Conselho Superior das Caixas Econômicas Federais, 1960 and De Placido e Silva,

As Caixas Economicas Federais: Sua História, Seu Conceito Jurídico, Sua Organização, Sua Administração e Operações Autorisadas. Curitiba: Empresa Gráfica Paranaense, 1937. Analysis of primary sources such as the

Revista das Caixas Econômicas Federais (1949-?), Annual Reports (1935), and the

Boletim Estatístico, (published by the Consultoria Técnica das Caixas Econômicas Federais) will be needed to better understand the evolution of government savings banks in Brazil.

7

Study of the Central Bank of Brazil is beyond the scope of this study. See: Whitehead, Laurence and Lourdes Sola, eds.

Statecrafting Monetary Authority: Democracy and Financial Order in Brazil. Oxford: Center for Brazilian Studies (Forthcoming, 2005).

8

Goldfajn, Ilan (

et al). "Brazil's Financial System: Resilience to Shocks, no Currency Substitution, but Struggling to Promote Growth", Central Bank of Brazil Working Paper, 75, June, 2003, p. 5.

9

Zysman, John. "How Institutions Create Historically-Rooted Trajectories of Growth",

Industrial and Corporate Change, Vol. 3, (1994), no. 1, pp. 243-83[ STANDARDIZEDENDPARAG]

10

Stiglitz, Joseph and Andrew Weiss, "Credit Rationing in Markets with Imperfect Information",

American Economic Review, (1981), Vol. 71, pp. 353-76.

11

Jacobs, Laurence

et al. American Democracy in an Age of Rising Inequality. Washington, DC: American Political Science Association, 2004, available at

www.apsanet.org.

12

Van Parijs, Philippe. "Basic Income: A Simple and Powerful Idea for the Twenty-First Century".

Politics & Society, (March 2004), Vol. 32, pp. 7-39, Pateman, Carole. "Freedom and Democratization: Why Basic Income is to be Preferred to Basic Capital". In: Dowding, Kieth,

et.al. (eds.)

The Ethics of Stakeholding London: Palgrove, 2003 and Barry, Brian. "Real Freedom and Basic Income."

The Journal of Political Philosophy. Vol. 4, (1996), pp. 242-76, Suplicy, Eduardo M. "Renda Básica: A Resposta está sendo soprado pelo vento,"

Brazilian Journal of Political Economy, (2003), Vol. 23, no. 2/90, pp. 47-62.

13

Mahoney, James and Dieter Reuschemeyer, eds.

Comparative Historical Analysis in the Social Sciences. Cambridge: Cambridge University Press, 2003.

14

Dahl, Robert Alan.

Polyarchy. New Haven; London: Yale University Press, 1971, Huntington, Samuel P.

Political Order in Changing Societies. New Haven, CT: Yale University Press, 1968.

15

On savings banks in Europe, see: Butzbach, Olivier. "Varieties within Capitalism? A Comparison of French and Italian Savings Banks, 1980-2000", Paper presented to the Society for the Advancement of Socio-Economics, Aix-em-Provence, June 2003.

16

Founded by José Florindo de Figueiredo Rocha, according to Rocha, Alfredo.

As Caixas Econômicas e o Crédito Agrícola, Rio de Janeiro, 1905.

17

José Francisco Sigaud,

O Homem Benfazejo ou das Vantagens que Resultam da Fundação da Caixa Econômica dos Povos Civilizados. Rio de Janeiro: Tipografia Imperial e Constitucional de Seignot-Plancher & Cia. Vol. XII, Biblioteca Constitucional do Cidadão Brasileiro, 1832.

18

De Placido e Silva cites the following dates and institutions: Caixa Econômica de Campos (1834), Caixa Econômica da Capital da Província da Bahia (1834), Caixa Econômica de Ouro Preto (1838), Caixa Econômica ou de Socorro da Província (Pernambuco, 1847), Caixa Econômica de Valença (1860), Caixa Econômica e Monte de Socorro (Rio de Janeiro, 1861), Caixa de Economias, (Salvador, 1853?), Caixa Econômica da Cidade de Nazaré (1854), Caixa Econômica da Cidade de Maceió (1856), Caixa Econômica de Santos (1857), Associação Econômica Auxiliar, (Rio de Janeiro, 1872), Caixa Econômica Auxiliar da Associação Mútua de Seguros Sobre Vidas Perserverança Brasileira, (Rio de Janeiro, 1874), Caixa Auxiliar da Sociedade Anônima Garantia do Futuro, (Rio de Janeiro, 1874).

19

Decree 2723 January 1861, cited in De Placido e Silva,

op. cit.

20

Caixas expanded from the center to the provinces during the Empire. In 1867, Finance Minister Visconde do Rio Branco authorized Provincial Presidents to retain 1% of lottery receipts to offset administrative costs of savings bank agencies, while reserving 1% of deposits for the treasury. Legislation in 1874 encouraged pawn services at agencies (reversed after 1892 under republican government in several states). In 1886, an initiative sought to place government savings bank agencies within or alongside already existing government bureaus and post offices, while 1887 legislation called for creation of Caixas at all

Mesas de Rendas Coletas and/or postal agencies.

21

De Placido e Silva,

Caixas Econômicas Federais, p. 70.

22

ibid. p. 232.

23

ibid. p. 80.

24

ibid. p. 82.

25

Booz-Allen & Hamilton - Fipe (Consortium). Instituições Financeiras Públicas Federais - Alternativas para a Reorientação Estratégica - Public Hearing. Brasília, DF: Ministry of Finance, June, 2000[ STANDARDIZEDENDPARAG]

26

In 1934, Caixas Econômica Federal branch offices existed in the states of Rio de Janeiro, São Paulo, Rio Grande do Sul, Paraná, Pernambuco, Bahia and Minas Gerais. During 1945, new branches were opened in Amazon, Ceará, Espirito Santo, Maranhão, Mato Grosso, Para, and Santa Catarina states. During 1946, new branches were opened in the states of Alagoas, Goiás, Paraíba, Piauí, Rio Grande do Norte, and Sergipe, followed by December 1956 in the territories of Acre, Amapá, Rondônia and Rio Branco (from Henrique,

Estrutura e Conjuntura das Caixas Econômicas Federais).

27

Hall, Peter A. and David W. Soskice, eds.

Varieties of capitalism: The Institutional Foundations of Comparative Advantage. Oxford: Oxford University Press, 2001.

Publication Dates

• Publication in this collection
  10 Apr 2006
• Date of issue
  Mar 2006

History

• Received
  Dec 2004
• Accepted
  May 2005