“The budget is the skeleton of the state, stripped of all misleading ideologies.”1 This simple but provocative declaration from 1917, one of the first principles of the field of fiscal sociology, introduced the idea that the structure of state finances revealed the history of its economic development, social values, and cultural biases. The notion of who should shoulder the fiscal burden and who should benefit from state spending—and determination of the proper or tolerable balance between the two—was a powerful lens into political, social, and cultural priorities in the relationship between state and citizen. This assumes, however, that budgets were reasonable approximations of social spending. While budgets may reveal the essential values of the state, they also hold the power to perpetuate myths of state priorities. In municipalities in nineteenth-century Brazil, they did the latter. Budgets forecasted surplus after surplus that comfortably provided the necessary resources to invest in the public good, while municipalities reported deficit after deficit, struggling to pay their bills. How was this possible?

This question is crucial to our understanding of the history of Brazil’s socioeconomic development because municipalities bore the legal responsibility for providing the public services that enhanced the quality of life and standard of living of all Brazilians, such as dependable roads and bridges for the circulation of goods and services, basic public health programs like mandatory vaccinations and protection of the potable water supply, and public safety through municipal codes of ordinance regarding proper behaviors and courts and jails when propriety was violated.2 These services were paid for out of local funds generated from taxes, fees, and fines levied on the local transactions of municipal dwellers. We found that the provincial regulatory oversight bureaucracy, put in place to monitor municipal finance under the highly centralized imperial regime that governed Brazil from 1822 to 1889, regularly declared municipalities fiscally fit even as financial results across time and space demonstrated financial instability and insecurity. This article examines the chasm between these two opposing representations of Brazilian fiscal health at the first point of contact between state and citizen, and their ramifications for our understanding of Brazilian socioeconomic development. It focuses on municipalities from the province of São Paulo, which began the century at a moment of transition into a dynamic commercial economy that by the turn of the twentieth century was the economic leader of the nation.

The tale of municipal poverty is not a new one. Scholars have drawn attention to the inefficient colonial tax system inherited by the newly independent nation, which depended on opportunistic or specialized taxes for specific or targeted uses. In this state of limited public investment, private beneficence by local notables filled the role the state would eventually assume (Fleiuss 1922; Boxer 1965; Kuznesof 1980; Tessitore 1995). The declaration of independence and the new constitution treated the fiscal question superficially, offering little guidance on the discrimination of public finances between national, provincial, and municipal authorities. The Additional Act of 1834, a major revision to the Brazilian constitution that acknowledged the difficulties of governing a continental nation from the center, struck a deal with provincial elites that gave them fiscal authority at home in exchange for political support in the capital by creating provincial legislative assemblies with the power to fix and review municipal finances (Dolhnikoff 2005).

With the reapportionment of administrative authority came the first formalized apportionment of revenues. The 1835 budget law divided up revenue streams between national and provincial governments in a manner that disadvantaged municipalities and persisted, almost entirely unchanged, for the remainder of the Empire (Montoro 1974; Nozoe 1984 and 2004; Tessitore 1995).3 It established taxes on the external sector as the principal source of revenue for the central government, including taxes on imports, exports, goods in transit, business conducted in the Custom House, and rates on anchorage and warehouses. These revenues, 80 percent of all public funds, paid for the business of running a nation, particularly the Ministries of Justice, State, Navy, War, Treasury, and so on, as well as for the capital city, Rio de Janeiro.4 To the provinces went the levies on the internal sector, such as taxes on wealth transfers and on intra- and inter-provincial trade, to pay for provincial transportation infrastructure, provide education, subsidize religious worship, and maintain law and order. These amounted to 17 percent of all public funds, reflecting the modesty of the domestic economy relative to the external sector. The municipalities were an afterthought funded by remainders, mainly leases on municipal property and taxes, fees, and fines on local commerce that amounted to just 3 percent of public funds yet were responsible for the physical, social, and economic infrastructure of Brazilian daily life. Modest municipal finances from fees on taverns, slaughterhouses, local businesses, rent from municipal properties, and fines for infractions of local ordinances paid for local administration (employee salaries and municipal council operating costs), law and order (jury costs and jail repairs and operating costs), and modest public services like removing anthills and killing stray dogs, cleaning and repairing streets, vaccinating against endemic and epidemic disease, maintaining the canals and fountains that supplied water, providing rudimentary lighting, and, most importantly, maintaining local roads and bridges for the circulation of people, goods, and services. The revenue apportionment placed municipalities in a financially precarious position (Carvalho 1988; Bieber 1999; Villela 2007; Hanley 2013). Their ability to pay for the provision of basic public goods, beyond private beneficence, directly depended on the ability of the local economy to generate revenues. The ability to generate revenues, paradoxically, depended on municipal investment in public goods that ensured a sound physical infrastructure. This placed municipalities in a negative feedback loop that adversely affected the health of the local economy.

Set against the backdrop of national political unrest and uncertainty, where political schisms in the capital and armed revolts in the provinces revealed tensions between national and provincial governments over the exercise of power, the municipality was engaged in the mundane task of learning by doing: learning to govern, learning to budget, learning to tax, and learning to advocate for the public goods underpinning economic well-being. The foundational laws of independent Brazil had introduced aspirational goals of what the state should provide to the citizens, most of which resided at the local level, with little understanding of the resources required to deliver those goods and services. The 1835 allocation of revenues between the levels of government was probably intended to provide revenue sources adequate to attend to the three levels of responsibility, but it did not do so, leaving municipal and provincial governments to negotiate their revenues and responsibilities. We argue that in this time of learning by doing, municipal poverty was whitewashed in a budgetary process that projected the image of provincial authority resulting in municipal inability to expand basic infrastructure and develop local economies. This, in turn, affected future revenue generation and the ability to fund quality-of-life services, perpetuating a pattern of underinvestment in the public good.

The São Paulo Municipal Economy

São Paulo in 1835 was home to 327,000 souls, a little more than 73 percent free and the balance enslaved, living in forty-six municipalities.5 Their major occupation was in what a contemporary observer termed “agricultural industry.” Among the most valuable products cultivated in the province were coffee, sugar, and a strong sugarcane-derived alcohol, aguardente. The most common products were rice, beans, manioc, and corn, but other products were also cultivated for domestic consumption, among them salt pork, tea, potatoes, cotton, wheat, brown sugar, and marmalade. Manufactures included cotton textiles, charcoal, and tiles. The techniques used in agriculture were simple, based on swidden agriculture inherited from the native population. The use of plows was unusual. The collective provincial economy was modest, with total production placing it ninth among Brazil’s eighteen provinces and thirteenth for per capita production, but it was poised for expansion due to increasing ties to the international economy through sugar, cotton, and coffee.6 By the end of the century, it was among the most dynamic economies in the nation.

This generalized portrait was reflected in the cases selected for this study. We chose six municipalities for investigation into municipal public financial health from 1836 through 1850, one case from each settled zone in the province.7 The municipalities, located in Figure 1, were drawn from the forty-six to fifty-six municipalities in São Paulo at this time. (The capital city, also named São Paulo, is included on the map for reference.) The wealthiest were Bananal and Campinas, which had the strongest ties to regional and export commodity markets. These two municipalities alone contributed 12 percent of the value of the provincial economy, producing on a per capita basis more than twice the provincial average. (See Table 1 for population and wealth statistics.) Araraquara and Franca, located in the Paulista “West” region of the province, had mixed economies based on production of foodstuffs for local consumption, some commodity crops, and substantial presence in livestock rearing for regional markets. Araraquara’s much smaller population produced per capita wealth on par with Bananal and Campinas, while Franca’s economy supported a much larger population at much lower product values per capita. Iguape and Apiaí, located toward the southern border of the province, engaged in domestic food production. Iguape was the most important rice producer of the cases, yielding per capita wealth similar to Franca. Apiaí was poor in every way, contributing little to the provincial economy and generating little wealth for its inhabitants. The six cases, then, represent profiles of communities both rich and poor in nineteenth-century São Paulo, approximating the experiences of communities across the newly independent nation.

Figure 1 

Map of the Province of São Paulo indicating selected municipalities. Authors’ elaboration from the map “Itinerário da 1a viagem de Auguste de Saint-Hilaire pela Província de São Paulo,” in Saint-Hilaire, Viagem à Província de São Paulo e Resumo das viagens ao Brasil, Província Cisplatina e Missões do Paraguai (São Paulo: Livraria Martins, 1940), frontspiece.

Table 1

Population and production data for the selected locales: Selected variables, 1836.

Variables Bananal Campinas Araraquara Franca Iguape Apiaí Cases total Province total

Total population 6,708 6,689 2,764 10,664 9,396 2,453 38,674 326,902
Slave population 1,679 3,917 937 1,515 2,917 490 11,455 86,933
% Slave 25.0% 58.6% 33.9% 14.2% 31.0% 20.0% 29.6% 26.6%
Production value (milréis) 259,426 308,325 91,882 111,883 94,639 1,407 867,562 4,766,918
Product per capita (milréis) 38.674 46.094 33.242 10.492 10.072 .574 22.433 14.582
Muni product% Provincial product 5.4% 6.5% 1.9% 2.0% 1.9% 0.3% 18.0%

Sources: Authors’ elaboration from Müller (1978, 124–129, 154–169). Exchange rates and relative values are from Leff (1982, 246), and measuring,

Note: The currency in Brazil during the nineteenth century and the first part of the twentieth was the milréis, or 1,000 réis. One milréis was worth around US$0.50 in the period under study. In terms of purchasing power parity for 2012, this was approximately US$13.00.

Published Municipal Budgets: The Picture of Health

Published budgets for all municipalities in the province of São Paulo, those “skeletons of the state,” gave the distinct impression that townships and their hinterlands were comfortably provided for.8 Anticipated revenues far outweighed planned expenditures, annual surpluses were the norm, and essential public works were the most important spending priority of municipal administrators. These budgets, prepared annually by municipal councils and vetted by the Committee on Municipal Budgets and Finance of the Provincial Legislative Assembly, were published by the provincial legislative assembly in bound volumes of law that carried the weight of authority.

Taken together, our six municipalities contributed between 10 percent and 15 percent of total budgeted municipal revenues in the province of São Paulo between 1836 and 1850 with the wealthiest of the cases, Campinas and Bananal, projected to produce the most revenues (Figure 2). Bananal’s share of total projected revenues grew beginning in 1839, a trend driven by the spread of commercial coffee production into the Valley of Paraíba from the coffee lands of neighboring Rio de Janeiro. Campinas’s wealth was primarily based on large-scale production of sugar for export to the rest of Brazil and abroad, although it also experienced cultivation of coffee for export before 1850, contributing to economic prosperity that made it the second largest and wealthiest city in the province by the end of the nineteenth century. While coffee was not taxed by the municipalities, the wealth it generated brought general stores, taverns, liberal professionals, apothecaries, and beasts of burdens and their cargo into the municipal centers. These were all subject to licensing and operations fees, as well as to fines for noncompliance with municipal ordinances, making them central sources of municipal revenues. Refined sugar, like coffee, was not taxed by the municipality this early in the nineteenth century, but its valuable by-product, the potent liquor aguardente, was. At least three separate taxes were levied on aguardente and the establishments that sold it, making it one of the most important sources of municipal revenue for most of the nineteenth century. The poorer municipalities, meanwhile, were turned toward supply of the internal market, typically rice, beans, corn, manioc, and pork, all lower-value products than coffee and sugar. (See Table 2 for production patterns.) The modesty of their commerce resulted in lower levels of projected municipal revenue. Figure 2 shows the growing divergence of fortunes between Bananal and Campinas, on the one hand, and the four other municipalities, on the other.

Figure 2 

Revenues as % of all São Paulo municipalities, 1836/37–1849/50. (The fiscal year ran from October 1 to September 30.) Authors’ elaboration from São Paulo Budget Laws, Leis da Província (1836–1849), Acervo Histórico da Assembleia Legislativa do Estado de São Paulo.

Table 2

Principal products by selected locales, 1836.

Product Bananal Campinas Araraquara Franca Iguape Apiaí Cases total Cases % province

Coffeea 64,822 8,081 210 73,113 12.4
Sugara 510 40,520 272 41,742 7.4
Tobaccoa 358 157 311 826 7.0
Cottona 84 84 0.9
Aguardenteb 7,399 70 337 40 7,846 16.8
Ricec 16,931 3,672 1,735 2,111 118,296 1,009 143,753 42.1
Beansc 9,674 21,015 3,654 34,343 14.5
Cornc 34,534 96,786 61,131 113,632 331,083 8.6
Manioc mealc 952 3,893 4,845 6.1
Pigsd 2,615 5,883 3,137 5,613 2,003 19,251 27.8
Sheepd 314 134 171 632 1,251 21.6
Cattled 139 687 2,667 1,817 5,310 14.9
Horsesd 221 287 353 861 7.6
Mulesd 29 4 33 66 2.9

Source: Authors’ elaboration from Müller (1978, 124–129).

a = 15 kilos; b = gallons; c = 72 liters; d = head.

While the varied revenue bases produced abundant projected revenues in some municipalities and modest revenues in others, they were sufficient to generate surpluses in all the economies according to published budgets (Figure 3). Surpluses that remained after the year’s ordinary expenditures had been made and payments due to suppliers, employees, and lenders from previous years (accounts payable) were satisfied suggest that budgeted revenues were more than sufficient to cover expenditures. The budgets for most all of the locales in all of the years we studied contained balances carried forward from prior years. Every municipality but Iguape—even tiny, poor Apiaí—projected surpluses to account for 50 percent of revenues at one point or another.

Figure 3 

Projected surpluses as a percentage of budgeted revenues, 1836/37–1849/50. Authors’ elaboration from São Paulo Budget Laws, Leis da Província (1836–1849), Acervo Histórico da Assembleia Legislativa do Estado de São Paulo.

As further evidence of apparent fiscal health, a look at the expenditures side of the budgets finds that these municipalities projected very low levels of indebtedness. Planned debt payments, the accounts payable for public works projects (materials and wages), unpaid salaries of municipal employees, or loans from local notables, were localized and sporadic, appearing in the budgets of just four of the six municipalities. As we see in Figure 4, Bananal carried the most persistent debt, perhaps the reason why it also projected the lowest surpluses. Franca planned for debt repayments in just five budget years, although three of those were quite sizeable. Campinas and Iguape were rarely in debt and those debts did not exceed 20 percent of planned expenditures.

Figure 4 

Projected accounts payable as a percentage of budgeted expenditures, 1836/37–1849/50. Authors’ elaboration from São Paulo Budget Laws, Leis da Província (1836–1849), Acervo Histórico da Assembleia Legislativa do Estado de São Paulo.

Based on budgetary evidence, then, the broad financial picture shared by all six cases, from wealthy municipalities with substantial export commodity production to relatively poor municipalities based on local food provision, was of adequate or abundant revenues. The regular projected surpluses signaled that municipalities had more financial resources than they required. The limited presence of debt among planned expenditures indicates they were not borrowing to meet their financial obligations. The budgets that were reviewed, approved, and published by the São Paulo provincial legislature led to the conclusion that the sources of municipal revenue were proportional to municipal needs.

Municipal Financial Statements: The Fiction of Fiscal Health

While budgets reflected fiscal abundance, manuscript collections of financial documents laid bare the shaky fiscal health of all municipalities. Financial statements and subvention requests submitted by the same municipal councils to the same legislative assembly committee that approved and published their budgets reveal that revenues in any given year fell short of expenditures, and municipalities regularly made appeals to the province for supplemental funds to pay for urgent projects beyond their means.9 Rich or poor, tied to the export economy or producing for the internal market, all municipalities shared the common characteristic that their budgets rarely reflected their financial reality.

To begin to conceptualize the chasm between the two, take Bananal, one of the wealthiest municipalities we studied. Published budgets for the fourteen years from 1836, the first year the provincial legislative assembly first standardized the type and presentation of official financial data, to midcentury show a municipality with ample resources. Its annual reports for the same time period, however, showed that Bananal twice recorded sizeable deficits.10 If we compare ordinary revenues to ordinary expenditures, that is, if we focus on whether a given year’s revenues were sufficient to pay the bills, the number of deficits doubles to four. The financial picture for four of the other five municipalities was more precarious.

The magnitude of the gap between abundance and uncertainty is revealed through a series of special requests from municipal authorities to provincial authorities for subventions to supplement local revenues. These requests were usually to finance major capital projects that cost far more than local revenues could underwrite.11 In January 1836 Bananal asked the provincial legislature to authorize a lottery to raise money to pay for a jail and sidewalks, a sum that was two hundred times its total revenues for that year.12 Over the years Bananal made an additional six requests for subventions to pay for, or to help pay for, public works projects that were beyond its means. Wealthy as it was relative to the other cases in our study, its annual revenues were insufficient to tackle expensive capital projects. In 1838 Bananal asked for money to complete its main church, explaining that it first used the income from the lease on church properties to pay for construction, but these funds were inadequate to complete the project with the “beauty and decency demanded by the religious cult of the state.”13 While the cost of the project was not included, the quantity requested, “some contos,” indicated that it was looking for a subvention of at least 2,000 milréis, or the equivalent of its total revenues for that year.14 The absence of financing mechanisms in Brazil at this time meant that these expensive projects would drain revenues as quickly as their bills came due. We know from a subsequent request for the same project that this appeal failed. The new appeal, replete with financial weariness, related that the church construction had been paid for through private solicitations and by renting church lands, but these resources had not proved sufficient and the citizenry had grown tired of supporting the expense.15

Often, subvention requests were to get the province to pay for projects that were its responsibility. The province was supposed to provide every municipality with a church and cemetery, a municipal council headquarters building, and a jail; and to subsidize upkeep for indigent prisoners, provide hospitals, and build and maintain provincial roads and bridges for circulation of goods. These were budgeted into provincial expenditures and paid for by revenues that were three to five times greater than all municipal revenues combined.16 What irritated municipal councilors was that their locally generated wealth flowed to the provincial treasury through its control over taxes and fees on provincial and interprovincial commerce, in part to pay for just these types of projects, yet payment requests were routinely delayed, deferred, or denied. In 1843 Bananal asked for more money, for the jail and municipal council headquarters. Clearly frustrated with the lack of support, “the Câmara dispenses with eloquence, hoping that you will decide to attend to our request and that [you will bear in mind] that our municipality is one of the ones that most contributes to the province and that has been the least apportioned in the budgetary laws.” It used the same logic to support a second appeal that year for 2,000 milréis to build a new bridge to improve commerce.17 The council explained that the road that passed through Bananal to the port at Angra dos Reis received a great deal of traffic from both São Paulo and its neighboring state Minas Gerais, traffic that generated taxes in the customs house for the province of São Paulo. This route, however, depended on the ability to cross a branch of a river. Because this branch had a tendency to swell and become impassible in rainy times, travelers from Minas Gerais preferred to take an alternate route to port via the Barra Mansa and Arrozal road in Rio de Janeiro province, meaning a fall in customs house revenues. Customs revenues belonged to the province, not the municipality, but they affected the municipal revenues in that they paid for road repairs and therefore affected the local commerce derived from this provincial trade route.

When properly maintained, provincial expenditures on provincial roads certainly had a direct, positive effect on municipal economies, but roads were rarely maintained even though provincial governments had earmarked revenues specifically for this purpose. Annual reports by provincial presidents in the 1840s bemoaned the heavy responsibility for road construction and maintenance with a litany of woes ranging from frustration with contractors who bid on projects but failed to complete them within budget, to torrential rains that wiped away recent construction and upkeep, to the lack of a scientific method to oversee efficient planning, design, and execution of the road network.18 From the municipal perspective, however, provincial roads were the lifeline that helped them survive. Bananal made much of its living from the pack trains from Minas Gerais that brought foodstuffs and other goods to the municipality in exchange for its products. Taxes on goods and services tied to the local economy were the major source of revenues that paid for basic public services. Therefore, constructing a bridge was essential for municipal economic health. The Camara’s request made this explicit: the province would benefit financially because the bridge would pay for itself out of the increased customs revenues resulting from building the bridge, while the municipality would benefit from improved commerce throughout the year.19

The steady stream of requests for financial subventions in the late 1830s signaled to the legislature that the formal apportionment of revenues between nation, province, and municipality was not meeting local financial requirements. Provincial authorities acknowledged as much in the 1840 presidential report. Low municipal revenues, “and therefore the lack of means [to pay for] material improvements to their municipalities,” had led to the transfer of a provincial tax on urban property to municipal coffers to pay for lighting, churches, and jails.20 Provincial authorities also observed that councilors were reluctant to introduce new municipal taxes because they saw how heavily national and provincial taxes weighed on their population.21

In order to better understand this tension, the provincial president asked that a law be passed to gather “municipal needs” reports from local councils.22 The resulting 1841 law required municipalities to make their needs known to the provincial legislative assembly. This prompted the municipality of Iguape to immediately submit a report asking for help with a new road to improve communication with neighboring towns and villages to the interior.23 Like Bananal’s request for the bridge, this report was a pitch that poor roads had a depressive effect on the local economy and better roads would lead to greater prosperity. Given its natural wealth, the report argued, Iguape should be among the most prosperous towns in the province, but it could not rival others because of the poor communications with its neighboring settlements. Iguape had received provincial funds for an expensive canal to promote rice exports through its coastal port, but internal trade was difficult.24 The existing roads were rutted and only inclined to worsen given that livestock traveled the same roads and were hard on them. This 1841 document did not specify an amount for the road improvement, but the copy of an earlier 1839 report to the provincial authorities included in this appeal placed the cost at 3,000 milréis, more than double Iguape’s revenues. It seems that this project got nowhere with the provincial assembly, however, because Iguape sent a frustrated letter in January 1843 complaining that its economy has suffered because the request for road money went unattended and dampened its exchange with other cities. Arguing for the indirect benefit to the provincial economy from domestic exchange was apparently not as persuasive as requests that supported commodity exports.

Apiaí engaged in a similar exercise in frustration, seeking funds for miscellaneous projects. This municipality was the smallest in our sample in terms of the value of production per capita, and one of the smallest in terms of annual revenues. It proposed a budget for various public works projects in 1838, requesting provincial funds totaling 1,800 milréis that included 300 milréis for work on the jail; 400 milréis for the construction of a new cemetery; 500 milréis to rebuild the roads between Apiaí and other settlements within the municipality; and 600 milréis for a new highway between Apiaí and neighboring Paranapanema, most of which were the province’s responsibility.25 Total budgeted revenues in that year were just 270 milréis, and actual revenues were barely half that.26 In January 1839, the Finance Committee and the Public Works Committee of the provincial legislative assembly commented that it did not think these expenses should come out of the public treasury, which was burdened with “extremely high expenses.” It asked the municipal council to find another means to pay as other municipalities have done, typically through donations of money or labor from a local notable.27 The municipal council, anxious to obtain funding, submitted a new request in January 1839.28 This included an impassioned defense of the request for the road to Paranapanema that was vital to Apiaí’s commerce. It specifically reinforced its argument that this road had been a frequently traveled road until it was denied funding for repairs and was ruined. It was now deserted. This road was badly needed for commerce because it provided the town with the most efficient transportation route to Itapetininga, the major point of reference for regional business.29

The rebuttal also defended the request for a jail, arguing that Apiaí had no jail or municipal council headquarters at all, with a rental house serving in their stead. “This does not seem right,” councilors wrote as they begged the legislative assembly for “some reasonable amount” to build a proper jail. (It bears emphasizing, first, that this was the responsibility of the province and, second, that the prevailing building material of this era was mud.) Finally, the council stressed the need for a proper cemetery to conform to provincial law. As to the provincial assembly’s challenge that Apiaí pay for these projects out of its own revenues, Apiaí explained that it had little revenue because its posturas—the ordinances that specified the fines, fees, and taxes that provided the financial base for the municipalities—were not yet approved. Caught in the bureaucratic labyrinth that feigned efficiency but perpetuated poverty, Apiaí had submitted the posturas to the legislative assembly on two occasions, but they had not yet been approved so the municipality had no ability to collect revenues.30

And so the appeals went. In the beginning of January 1841, Araraquara asked for “some amount” of money to construct a public cemetery to fulfill the requirement first articulated in the Law of October 1, 1828, that elaborated the myriad responsibilities of municipalities and their governing councils, reaffirmed in the provincial edict of March 20, 1838. The municipal council declared that it lacked the money to pay for this public good.31 Campinas, one of the most prosperous of our municipalities, requested elevated sums to complete its main church and to maintain its cemetery. Its request for 2,000 milréis in 1837 and 12,000 milréis in 1839 were double and more than ten times its revenues in those years, respectively.32 The costs of these capital projects, like those in all of the requests reviewed here, far exceeded the local resources.

Contrary to the picture of ample financial resources presented in annual budgets, these requests for supplemental funds indicate that municipalities actually operated under very modest and even precarious financial circumstances. Major capital projects swamped local revenue sources by orders of magnitude. Municipalities were legally prohibited from borrowing without permission from the assembly, and in fact they did not borrow to pay for public works projects until the final decades of the century, so large capital expenditures were met through current accounts, subventions, or private donations. Indebtedness, when it appeared on financial statements, represented unpaid bills for these very works projects.

It is possible that municipalities were simply trying to extract additional resources from the upper echelons of government in a new and evolving institutional setting, but the year-end financial statements confirm what subvention petitions suggested: municipalities experienced financial straits year in and year out.33 We compared the revenues collected in a particular year to the expenditures made that same year, and compared these financial statements to their corresponding budgets. The single most remarkable point of comparison was that while budgets—projections of next year’s revenues and expenditures—were always balanced and reflected a degree of financial stability, annual financial statements revealed violent swings in ordinary revenues and expenses. A year of generous surplus fueled by plentiful revenues and manageable expenditures might be followed by a year of dramatic deficit. Deficits in turn prompted conservative spending in the following year to return to a surplus.

The pattern was anything but predictable. Figure 5 shows the year-to-year volatility of municipal finances. Most municipalities experienced expressive, even alarming deficits in the 1830s as they adjusted to the new administrative and fiscal regimes that asked them to arrange for local tax collection to pay for public administration and services. Bananal’s wealth from ties to the export sector probably helped it avoid the deep deficits that plagued the others but did not spare it from fluctuating between surpluses and deficits every two or three years. The other five had similarly varying experiences with surpluses and deficits, depending on the management of their relative wealth or lack of wealth. Prosperous Campinas, also tied to the export sector, swung between years of surplus (two) and deficit (six), though its deficits were almost always relatively modest. Franca, of mid-tier prosperity with no ties to external commerce, recorded more years of deficit (seven) than of surplus (four) with big swings between the two and with the greatest single-year deficit, 330 percent of revenues. Iguape and Apiaí, the two least well-endowed economies, were perpetually in the red. Only Araraquara recorded more surpluses than deficits. It had one spectacularly bad year when expenditures were 250 percent of revenues, an imbalance possible only because of conservative spending the year before that gave it a large cushion. But it also managed to record eight consecutive years of surpluses, four of which were above 50 percent of the value of annual revenues. This was managed by spending very little on its citizens. Araraquara opted for fiscal safety over investment, based as it was on the domestic economy, which generated low tax revenues. As Table 3 shows, it had among the lowest per capita spending of any municipality.

Figure 5 

Surplus or deficit % annual revenues,1835/36–1849/50. This graph reflects the ability of annual revenues to meet annual expenditures. Taken together, municipalities recorded surpluses in twenty-one years and deficits in twenty-seven years. Authors’ elaboration from São Paulo, “Império” Manuscript Collections, various municipalities, 1837–1850. Acervo Histórico da Assembleia Legislativa do Estado de São Paulo.

Table 3

Per capita spending (nominal reis).

Rank order Total spending 1837/38 Per capita spending 1837/38 Proportion of highest per capita spending Total spending 1849/50 Per capita spending 1849/1850 Proportion of highest per capita spending Pop 1836 Pop 1854

Campinas 1,584,799 237 2,789,619 196 .68 6,689 14,201
Bananal 1,189,549 177 .75 3,272,398 285 6,708 11,482
Iguape 975,219 104 .44 967,201 64 .22 9,396 15,211
Apiaí 200,000 83 .35 201,000 105 .37 2,423 1,907
Franca 616,978 58 .24 986,414 104 .36 10,644 9,500
Araraquara 117,522 43 .18 412,178 83 .29 2,764 4,965

Source: Authors’ elaboration from São Paulo, “Império” Manuscript Collections, various municipalities, 1837–1850, Acervo Histórico da Assembleia Legislativa do Estado de São Paulo,; São Paulo, Leis da Província, Lei 28, May 15, 1849.

Note: All figures are actual except Apiaí 1849/50, which is budget. Campinas 1849/50 is 1850/51. We lack population data for Franca in 1854; we know that population fell to 8,248 in 1872, according to the 1872 census. We averaged the decline for the 1854 estimate. The 1849/50 figure, then is an estimate.

Budgets, then, were terrible at projecting revenues.34 Of the forty-eight sets of data for which we have budgets and financial statements, only six projections were within 10 percent of actual revenues.35 If we expand the margin of error to 20 percent plus or minus budgeted revenues, the number rises to just fourteen instances. The remaining thirty-four budgeted revenues missed their mark by more than 20 percent.36

Budgets, Financial Statements, and Municipal Financial Health

Of course, it was not possible for the municipalities to survive such volatility and sustained deficit production over the long run without adjusting to the uncertainty. They adjusted in two ways. First, municipalities immediately curbed spending by more than they needed to. A shortfall in projected revenues almost always resulted in an even greater reduction in expenditures in either that year or the following year. In nineteen of the twenty-three instances when revenues fell short of budget projections, expenditures were held to a level far below their budgeted amounts to compensate. Second, municipalities relied on years of surplus to buffer years with deficits. These surpluses came in years when improved revenues were used to build up a financial cushion rather than make budgeted expenditures. In more than half of the instances when revenues were greater than anticipated (sixteen of twenty-eight), actual expenditures fell short of budgeted expenditures. In only twelve of the twenty-eight years of unexpected revenues did expenditures match or exceed budgeted amounts, but not to the higher level of the revenues generated.

This reliance on surpluses was built into published budgets; it is the answer to how forecasts presented one history of economic well-being while year-end results presented another of poverty. Problematically, this surplus was not the product of adequate revenues from taxes, fees, and fines nor of modest expenditures. Budgeted surpluses were based on the assumption by budgetary authorities that municipalities would successfully collect all accounts owed to them. In Bananal, for example, the sum of projected surpluses plus accounts receivable—collection of payments owed to the municipality—amounted to anywhere from 30 percent to 80 percent of projected revenues in every published budget but one.37 For Franca, budgeted surpluses and accounts payable amounted to 30 percent or more of revenues in nine of thirteen budgets, while for Campinas and Araraquara, they made up more than 30 percent of budgeted revenues in eight of thirteen budget years. Our two poorest cases, Iguape and Apiaí, also hoped for surpluses and accounts payable in their budgets, but not in such great proportions.

These accounts were the key to, and the fiction behind, municipal fiscal well-being. The provincial budgetary authorities projected full payment of accounts receivable in every year, even though the revenues from these payments rarely matched the amount owed. These illusory revenues, in turn, generated the projected surpluses that were carried forward to the following budget year. There was no adjustment in this pattern to bring budgets in line with practical experience, even though the very provincial committee that approved and published the budgets was the same committee that very carefully reviewed the year-end financial statements. This accounting represented a tacit acceptance of a fiscal arrangement that failed to produce adequate revenue and encouraged conservative investment in public goods.

This fiction had a direct, negative effect on municipal investment in public goods, particularly the works that provided infrastructure for economic exchange. If we look at the difference between budgeted and actual revenues and expenditures for our municipalities, we see which accounts suffered when revenues were insufficient to meet expenses. The major categories of expenditures for all municipalities were administrators’ salaries, law and order expenses, public works, and municipal council operations costs. Administrator salaries went to patronage appointees filling the municipal secretary, inspector, and doorkeeper positions. Some towns also paid for a jail keeper. The stipends for these administrators were established in provincial law were nonnegotiable. Therefore, they were the most predictable values in the annual financial picture. Law and order expenses included the costs of trials, of housing and transporting prisoners, and of cleaning and lighting the jail. Occasional capital outlays for new shackles also counted among these costs. These were the most volatile accounts, because they depended on the number of crimes committed and trials held in a given year and could not be predicted in advance. Public works projects included road construction and repair, bridge repairs, excavation and filling operations, and building expenses, maintenance, and repair. The costs associated with the municipal council were variable but included the costs of books to record minutes, tax collections and fines, furniture, and rental of a meeting place while the regular hall was under construction.

Because year-end municipal financial statements were required to indicate how authorities spent money and why, we learn how the administrators managed their resources from year to year, including which bills were paid in full and how much was owed to or by the municipality. This rich detail reveals that the procurador, the town official charged with collecting revenues and paying the council’s bills, was the one municipal employee most likely to be paid on time. Because his salary was a percentage of the revenues he collected for the municipality, his payment was set aside immediately and therefore guaranteed. The political appointees also were fairly likely to receive their salaries. Accounts for all six municipalities show regular payments to the secretary, inspector, doorkeeper, and jailer. The two municipalities with the most volatility and biggest deficits however, Bananal and Franca, did not pay these administrators in full. The regularly budgeted salaries from year to year were paid at a modest discount in many years, with arrears paid in later years when times were less lean. Bananal’s financial reports almost always failed to pay the full salary owed to the administrators for the fiscal year, but added to the discounted salary in one year the portion they owed from the previous year. Thus, administrators’ salaries approximated budgeted sums in reported values but in fact were perpetually in arrears.

Law and order costs were consistently paid, even as they rarely matched budget projections. Because of the variability in behaviors that drove these accounts, municipal administrators had no good way to economize when revenues fell short. Prisoners had to be detained, juries had to be convened, and court costs had to be paid whether revenues were strong or weak. For three of the municipalities (Apiaí, Campinas, and Iguape), the weight of law and order in actual results was similar to its importance in the budget. For the other three, however, this account was a wild card in municipal finances. In 1841 Franca, just one of many possible examples, law and order was budgeted for 15 percent of expenditures but consumed 60 percent. Araraquara’s law and order also cost more than projected: it ranged from 4 percent to 15 percent of the budget but in actuality consumed 32 to 63 percent of all spending. In Bananal in 1838/39, law and order was allocated 21 percent of the budget but used up just 3 percent. Two years later the inverse occurred: law and order was projected at 4 percent of total spending and then accounted for 25 percent.

Salaries had to be paid and law and order maintained, so the category of accounts that absorbed the shortfall in lean years were operating costs of the municipal councils and costs associated with public works. Both demonstrated flexibility across time and municipalities, but managing council operations in hard times was tinkering at the margins. These operational costs did not account for very much in the overall budgets nor did they fluctuate much except for the odd spike in an election year.38 That meant the brunt of the burden to manage municipal finances fell on public works. No matter the resources budgeted for works, spending lagged far behind. Between 1836/37 and 1849/50, money spent on public works lagged the budgeted amount in thirty-nine of forty-four instances.39 This on its own is not remarkable; actual revenues and expenditures often fell below budgeted amounts. It is when we look at the relative weight of public works in total expenditures that we see how seriously it was discounted: public works as a percentage of total spending was compromised more than any other category 81 percent of the time. Franca provides us with a visual representation of financial juggling in Figure 6. Here we see that public works received generous funding in budgets but rarely saw much of that funding in reality. To return to 1841 Franca, when law and order was allocated 15 percent of expenditures but consumed 60 percent, we find that public works were supposed to get 55 percent but received just 8 percent of all expenditures. In a setting of financial penury, it was public works spending that went unfulfilled or deferred.

Figure 6 

Percent share of major expenditures, budgeted and actual, Franca, 1836/37–1842/43. Authors’ elaboration from São Paulo Budget Laws, Leis da Província 1836–1849; and ALESP Império Manuscript Collection, “Franca,” 1837–1844,

Campinas was the only municipality where public works spending was discounted at about the same rate as the other major categories when revenues fell short and expenditures had to be cut. For all others, public works were disproportionately shortchanged relative to the other categories when revenues failed to achieve budgeted values. The most egregious case was Araraquara, the one municipality that generated the most surpluses. On paper, Araraquara was the picture of fiscal health, but this apparent good fortune was almost entirely the result of municipal neglect of basic infrastructure. Its spending on public works was the least significant of any municipality studied here. Although Araraquara budgeted a hefty percentage of expenses for public works, actual spending barely registered in year-end financial statements.40 It paid salaries, the tax collector, legal fees, and little else. Its administrators, undoubtedly like administrators throughout the province and the Empire, managed their precarious finances in an exceedingly cautious manner that resulted in far lower levels of investment in local public services than they planned for or needed in order to build up surpluses to protect against future shortfalls.


Brazilian budgets, far from providing “the skeleton of the state, stripped of all misleading ideologies,” were fictionalized accounts of municipal financial health that obscured the volatility and insufficiency of municipal revenue sources.41 As a result, they contributed to compromised investment in public services that bore directly on the socioeconomic development of Brazil at the first point of contact between the Brazilian state and citizen. Published budgets were built on the unrealistic assumption that debts to municipal treasuries would be paid in full and in a timely fashion. This assumption, in turn, projected the necessary revenues to cover budgeted expenditures and produce a surplus that carried over to subsequent years. In fact, the financial fortunes of the municipalities rarely came close to the budgets. The budgets generated by the municipalities and approved by the provincial oversight committee were optimistic guesses at the coming years’ revenues and the likely uses for those funds. The administrators adjusted the latter to conform to the former in every fiscal year, almost always selecting a conservative spending profile in order to generate a surplus when they knew revenues would fall short of needs. That the incidence of substantial deficits diminished over the fifteen years we studied suggests they became increasingly able in adapting spending to revenues. This does not mean, however, that revenues were increasingly sufficient to meet the capital requirements of the municipalities. Requests for subventions by municipalities across this era stretched to the end of the Empire in 1889, cataloguing the myriad needs that went far beyond the revenue streams generated by local taxes, fines, and fees.

Our comparison of published budgets to year-end financial statements offers a revealing look into the fiscal history of the Brazilian municipality in the early independence period. These municipalities were quite small at that time—just a few thousand residents in each—and therefore their financial woes may seem insignificant at first glance, but the patterns established in the fifteen years after the Additional Act remained virtually unchanged for the remainder of the Empire. The underinvestment in the municipality we observe here established the basis for underdevelopment of social and economic resources for the next fifty years. The fact that revenues varied so greatly between projected budgets and reported results, and from year to year, created a financial volatility that influenced public spending patterns. These patterns directly and negatively affected economic development. Municipal administrators spent money on the one category they were required to by law—the salaries of patronage appointees to the positions of secretary, fiscal agent, and doorkeeper—and addressed the balance of their needs as they saw fit. Law and order financial requirements were impossible to predict in advance and a major source of volatility in municipal spending. Public works were the easiest projects to cut in times of financial shortfall and therefore the most likely to be shortchanged. But road and bridge construction and repair directly affected the local economy by maintaining infrastructure for the circulation of goods and services. Taxes and fees on these goods and services, in turn, formed the fiscal base of the municipality. Cuts in basic public works projects, then, created a negative feedback loop that compromised future revenue generation. It is this poverty that was the hallmark of the nineteenth-century municipality, not the fiction of abundance resources that flowed from the printed budget laws of the province.