1 Introduction

Terence J. Byres started his classical study on the history of sharecropping with the remarkably simple, but powerful statement that “[s]harecropping is as old as recorded history.”¹² Extremely malleable, this labour-rental arrangement has persisted throughout human history and has been used around the globe to organise the production of the most diverse agricultural goods under radically different types of technologies and social organisations. Consequently, specific sharecropping contracts have shown extremely diverse features. Analytically, sharecropping can even be seen as a labour contract and/or a land-rental arrangement.³ Such differentiation has important implications for the power relations established between the contracting parties, which usually include patron-client relations in the former case, or the need to set property rights over the land being cultivated in the latter.⁴

Notwithstanding such relevant particularities, the sharing of crops between one entity that holds the rights over the land and another that has the means to produce in it – and, thus, has control over (family) labour and/or agricultural implements – remains a pervasive form of organising rural relations. For instance, nowadays sharecropping tends to be used as a labour arrangement among poor families practicing agriculture in low- and medium low- income countries. Particularly in South Asia, this labour-rental arrangement is frequently interlinked with credit operations leading to modern forms of debt-peonage and to personal dependencies that not infrequently go against some of the basic principles of the Decent Work Agenda.⁵

Interlinkages with credit supply and the power relations that this implies are yet another historically ubiquitous feature of sharecropping.⁶ The current chapter will look at a similar arrangement that prevailed in the southwestern Brazilian province of São Paulo from the 1840s, showing how its mechanisms of incentives and controls persisted over time and outlived the original contract to become a cornerstone of Brazilian immigration policy.

In spite of the profound socio-economic and political changes that took place in Brazil between 1840 and 1940 – including the abolition of slavery in 1888 and the substitution of a centralist empire by a federal republic in 1889 – the design and enforcement of various labour-rental contracts remained deeply influenced by the first experiments with sharecropping in the mid-nineteenth century. In this context, the chapter will highlight some institutional features associated with the design of the contract and its enforcement. The effective application of a labour-rental arrangement depends on the presence of formal institutions to enforce it and on the cultural disposition to accept the constraints it imposes on the discretionary will of the contracting parties.⁷ Both were missing for the most part in the plantations examined in this chapter: sharecropping constituted the first form of non-captive labour for conducting ordinary agricultural tasks during the Brazilian transition from slavery to non-captive labour arrangements. Thus, it is not surprising that even landowners interested in this new way of organising agricultural production leaned towards patronage relations, reacting with scepticism – and sometimes with rage – to the opposition of labourers, who not infrequently expressed their dissatisfactions and burst into riots that farmers could not repress in the same way they had been allowed to do with the slaves.

Hardly any other labour-rental arrangement has received more attention in the historiography of Brazilian rural relations than sharecropping. As it was the first attempt to begin the transition from slavery to free labour on economically booming plantations, there is a large body of literature – consisting of studies that have become classics in the field – that has evaluated the evolution of sharecropping and its impacts on other types of contracts.⁸ In his preface to the translation of the memoirs of the leader of a famous sharecroppers’ riot, Sérgio Buarque de Holanda inaugurated various branches of academic research.⁹ This now-classic text embedded the experiments with bonded labourers of the 1840s–1850s in the fields of the history of immigration and the transition from slavery to free labour in Brazil. Studies by Warren Dean, Verena Stolcke and Michael M. Hall, and Emília Viotti da Costa deepened this approach.¹⁰ Luiz A. Corrêa do Lago extended it to show how different economic conditions led to various forms of labour arrangements across southern and central Brazil.¹¹ Studies of specific farms that employed sharecropping complemented these approaches, especially for those plantations that first experimented with sharecroppers.¹² Finally, Maria L. Lamounier provided the approach most closely related to this chapter, studying the institutional framework that supported non-captive labour relations in the coffee economy and how the actual enforcement of contracts led to modifications in Brazilian legislation in the nineteenth century.¹³

This chapter will revisit some of the themes of this rich literature and deepen the discussions on how the early experiments with sharecropping in São Paulo had consequences that made their effects felt at least until the beginning of the twentieth century. The goal of this chapter is to study some of the “inertial” aspects laid down by the first adoption of sharecropping. We thus advance the study of two classical hypotheses: (1) the sharecropping system proposed in the 1840s had long-standing consequences for contractual design; and (2) it influenced other labour-rental arrangements.¹⁴ Even if completely new clauses, contracts and institutions to enforce them emerged between 1840 and 1940, various contractual mechanisms kept reappearing over time, as is demonstrated with some concrete cases. Moreover, the interlinkage of labour-rental arrangements with credit operations remained a cornerstone of rural organisation in southwestern Brazil for this entire period.¹⁵

The chapter uses two sets of contracts studied by the authors in previous research. We will start with those contracts enforced in the central-western plateau of São Paulo in the 1840s–1880s.¹⁶ In this part of the chapter, we will focus on sharecropping and its relation to contracts of fixed payments per time worked and per piece rate (jornal and locação de serviços). The second set of labour arrangements refers mainly to sharecropping and its relation to contracts for the formation of new plantations (a specific form of empreitada) enforced in the 1910s–1930s.¹⁷ This set of contracts has a broader geographical coverage, including older plantations on the central-western plateau and the new agricultural frontiers in the state of São Paulo at the beginning of the twentieth century.

Before proceeding, we would like to reflect briefly on our approach. Through writing a history of contractual changes we adopt a formalist approach to understanding the evolution of actual labour relations. Non-written features that dictated the expected and actual behaviour of labourers and farmers can be retrieved partly from sources that we will use only marginally in this chapter – such as memoirs, letters and interviews, reports of governmental inspections, and the narratives of foreign travellers who visited the farms. Notwithstanding, this chapter is based on the recognition that contracts that survived in the archives framed the legally acceptable range of action of the involved parties. As such, changes in their clauses provide us with an overview – even if incomplete – of an endogenous process of learning and adaptation in daily labour relations.

The chapter is organised as follows. Section 2 provides an overview of the interconnected histories of immigration and the transition from slavery to non-captive labour arrangements in Brazil, particularly in the coffee plantations of São Paulo. Section 3 studies the structure of the sharecropping contracts proposed in the 1840s–1850s, focusing on their multidimensionality – that is, stressing that a contract that was primarily seen as a labour arrangement also had a rental dimension, a mechanism of cost sharing, and an explicit interlinkage with a credit dimension. The study of the persistence and change of each contractual dimension at the beginning of the twentieth century is the object of Section 4, which complements the analysis of the contracts with information about institutional changes in Brazilian labour markets.

2 Slavery, Immigration and Non-captive Labour Arrangements in São Paulo

If the study of contracts is to serve the purpose of understanding labour relations, then it is necessary to have at least an overview of the relevant labour markets. In nineteenth-century Brazil, the history of labour was set by two intertwined processes: the sluggish abolition of slavery and the promotion of immigration – both having enormous impacts on the employment alternatives available to non-enslaved Brazilians. In 1888, slavery was abolished, unconditionally, but without any further support for the formerly enslaved population. As of 1886, mass immigration – subsidised by the provincial government of São Paulo – inserted Brazil into the global circuits of the age of mass migration. These were landmarks for a historical process that had been going on for almost a century.

A legalist approach to the history of abolition allows us to distinguish three phases that overlap with the history of immigration. The first refers to the abolition of the transatlantic slave trade and various experiments with migratory policies in the 1810s–1850s. The rise of bonded labour immigration under sharecropping contracts in the 1840s–1870s and its fade-out into new arrangements constituted an intermediary period. Finally, the abolition of slavery itself and the promotion of state-subsidised mass immigration in the 1880s–1920s crowned the process – and was also a reason for the fall of the Brazilian Crown in 1889.

The enforcement of the legislation against the transatlantic slave trade in 1850 disrupted an international circuit of labour supply that had fuelled the Brazilian plantation economy since colonial times. Since the transfer of the Portuguese Crown to Brazil in 1808 and Brazilian political independence in 1822, the Portuguese and Brazilian governments attempted to keep the slave traffic as active as possible.¹⁸ This put Brazil at odds with the British Empire – leading to severe diplomatic tensions and maritime threats –, which opposed the traffic because its interests were now aligned with the consolidation of industrial capitalism.¹⁹

In this period, Brazilian immigration policy remained experimental. The prevailing practice was to recruit European households for settlement in public lands; nonetheless, experiments with other forms of immigration had already taken place, such as the recruitment of specialised labourers for public works.²⁰ It was only by the mid-1830s that a class of plantation owners established in the agricultural frontiers of São Paulo’s central plateau started to link the category of “immigrant” to that of “agricultural labour” more directly, as an alternative to the government-led promotion of an independent class of foreign peasants in smallholdings.²¹

The responses given to the “labour question” by the coffee planters in the next period, roughly between the late 1840s and the 1870s, can, in hindsight, be seen as an expression of myopic behaviour²² First, with the international supply of slaves cut and facing the possibility of negative natural rates of reproduction among the enslaved population, the end of slavery became a matter of time – an issue that was self-evident to the landowners. Nonetheless, the most common response of plantation owners in São Paulo was to import slaves from less economically booming provinces and regions.²³ This self-illusionary denial of an evident endpoint of slavery was, nonetheless, an individually rational response, informed by short-term concerns. After all, experimenting with non-captive labour involved many uncertainties; these landowners had made significant investments in buying slaves; and a great part of this elite’s social distinction and political power came from the wealth generated by the very institution of slavery. Notwithstanding, more than a hundred landowners started experimenting with different forms of contract labour as an alternative to slavery in the 1850s–1860s.²⁴ These experiments were short-lived for the most part and gave way to a recrudescence of slavery in the 1870s.²⁵ Experimenting with contract labour in the belief that the end of slavery would come about soon was a rational response on the part of the landowners; its abandonment later on – when the scarcity of captive labour had increased even more – was, once again, individually rational in the face of the high transaction costs related to the necessary adaptations to non-captive labour.²⁶

Members of the rural elite of São Paulo, whose plantations were expanding in the 1850s, made some attempts to invest in immigrant labour more systematically. However, Brazil was not a particularly attractive destination to foreign labourers, especially considering the combined prevalence of slavery, the lack of religious freedom in the officially Catholic empire, and the difficulties for the average labourer to attain landownership. Moreover, there was a general lack of experience with non-captive labour: certainly, various forms of patron-client relations with freemen executing sporadic tasks had always been present in the plantations, but not for organising ordinary agricultural labour.

A proposed solution to these problems was to interlink a labour-rental contract to a credit arrangement for poor potential immigrants, who lacked the wealth and income to cover the costs of international migration on their own. In São Paulo, the first landowner to propose this arrangement was the Portuguese-born, Brazilian senator Nicolau Pereira de Campos Vergueiro.²⁷ According to his contract, an immigrant household received credit from a landowner in Brazil to cover the migratory costs and subsistence expenses in the new country; the immigrant household was then obliged to repay the debt by working on the farm and sharing the annual profits of rural production under a sharecropping arrangement.²⁸ Known for many years as the “Vergueiro system,” this contract was adapted over time and across plantations, but the aspect of credit interlinkage became a cornerstone of the Brazilian immigration policy and was crucial for the path taken in Brazil for the transition from slavery to free labour.

This specific form of peonage provided some security to landowners as a mechanism of control over labour mobility and turnover. However, such controls also caused much discontent among immigrants. Aligned with foreigners’ frequently exaggerated expectations, the different readings of the sharecropping clauses by labourers and farmers led to a number of official complaints to diplomatic authorities and even to strikes. The most famous, the so-called Sharecropper’s Riot, led by Swiss immigrants, took place on the very farm of Senator Vergueiro in 1856–1857 and gave rise to a series of inspections from Brazilian and international authorities. This event and other complaints by German immigrants in the province of Minas Gerais triggered the cancellation of various hiring licenses to Brazil and led to the prohibition of pro-emigration propaganda in Prussia under the Von der Heydt Rescript, which was extended to the entire German Empire later on.²⁹

The riots, the alleged interference of diplomatic authorities in the labour discipline of the farms, and farmers’ discontent with all these novelties were symptoms of a deeply rooted problem: the unwillingness to enforce impersonal arrangements between nominally equal contracting parties in a country whose economy, society and politics were primarily based on slavery.³⁰ Thus, in spite of many individual initiatives and various non-materialised political projects – including plans to promote the immigration of Chinese coolies – the rural economy of São Paulo in the 1870s mainly advanced through captive labour bought at high prices in other provinces and regions.

The foundation of the Associação Auxiliadora da Colonisação (Supportive Association for Colonisation) in the 1870s brought about some important new developments. The actual initiatives of that association had rather limited impacts, but its conceptualisation implied an important change in the migratory strategy pursued by the rural elites. The system that had been designed in the 1840s–1850s depended on a farmer who supported the hiring of immigrants in Europe; although public funds usually backed up the credit advances made by landowners to immigrants, recruitments in Europe were decentralised. The Associação Auxiliadora changed that by acting on behalf of its members, increasing the potential scale and the organisational capacity of the recruitments.³¹ The posterior Sociedade Promotora da Imigração (Society for the Promotion of Immigration) scaled up these features. Founded by leading coffee planters in 1885 and taken over by the government of São Paulo in 1892, this organisation successfully accomplished its goals of promoting pro-immigration propaganda, hiring immigrants, lodging them upon arrival and creating a centralised labour market that allowed farmers to hire labourers recently arrived in the city of São Paulo.³²

The pressures of abolitionist activism – which included organised civil movements and mass escapes from the farms in the second half of the 1880s – and the increased demand for a secure source of cheap labour in the booming rural economy explain the foundation of these associations as the somewhat desperate answer of the planters to the final phase of abolition. Furthermore, the new elite of plantation owners was now politically better established. This led to the implementation of an official programme to subsidise immigration, which led to a massive inflow of foreigners from 1886.³³ This policy spread the costs of immigration and relieved individual planters from the insecurity of immigrants who defaulted on their obligations.

In terms of migratory inflows, the policy was a great success: between 1885 and 1919, São Paulo received a gross inflow of approximately 1.72 million foreigners, corresponding to 58.1 per cent of all entries in Brazil.³⁴ Nonetheless, similarly to the 1850s, planters in the 1900s kept complaining about labourers’ turnover and voicing quasi-apocalyptical predictions about the consequences of the “labour crisis” (crise de braços) for the Brazilian economy. These complaints reflected the fact that coffee kept devouring land and labour in its march to the west: the agricultural frontiers had greatly expanded in the first decades of the twentieth century, as public policies maintained coffee prices paid to producers artificially high and the western portions of the state were reached by railroad infrastructure.³⁵

Contractually, an important novelty of the 1880s was the implementation of the so-called colonato system, which constituted the basis for rural labour relations in São Paulo well into the twentieth century. A family under colonato received fixed and predetermined payments for taking care of the coffee trees (or of other crops), a share of the household’s annual harvest, and was permitted to practise subsistence agriculture in between the rows of new coffee trees. A study of the exact origins of each of its clauses goes beyond the scope of this chapter, but, here, one notices a substantial continuity vis-à-vis the old sharecropping arrangements, which included side payments for tasks in the lean season that became more clearly defined over time.³⁶

Mass immigration also required improved institutions and organisations to regulate labour arrangements and deal with conflicts. Public responsiveness to problems on the farms was crucial for maintaining the high inflow of immigrants, as the Prinetti Decree demonstrated in 1902. This Italian decree forbade subsidised emigration from Italy after reports of abuses on the farms of São Paulo, including a riot triggered by the assassination of Diogo Salles, brother of the Brazilian president, by the Italian colonist Angelo Lungaretti, after Diogo’s son attempted to “seduce” Lungaretti’s sisters – which were euphemistically reported, most likely referring to verbal or physical sexual assaults.³⁷ Although mass immigration continued, with Portuguese and Spanish immigrants substituting the primacy of the Italians – who could still immigrate spontaneously –, the Prinetti Decree triggered debates similar to those of the Von der Heydt Rescript.

In 1906, new public initiatives included the foundation of an official agency to match labourers and farmers and to supervise the signing of contracts at the lodgings of the Sociedade Promotora, as well as the foundation of an inspectorate to evaluate the travel conditions of immigrants.³⁸ In 1911, an umbrella organisation – the Labour Department of the State [of São Paulo] – took over these responsibilities. To carry out the task of monitoring rural labour relations, the government of São Paulo established the Patronato Agrícola (Agricultural Office) in 1911, whose first mandate was to “support the execution of federal and state laws concerning the defence of the rights and interests of the agricultural proletarians.”³⁹ For this chapter, it is impotant to remark that the legislation referred to here formalised a system of “current account booklets” (cadernetas). According to this legislation, all accounting related to a household should be noted down not only in the farm’s administrative books but also in booklets given to labourers – in which the labour contract was also written down – allowing for clearer mutual monitoring.

The timeline sketched in this section stresses the overlapping histories of immigration and the abolition of slavery in Brazil. In terms of labour arrangements, we highlighted the learning processes that took place in the half-century after the elaboration of the first sharecropping contract. While defending the thesis that sharecropping had a long-standing influence on the elaboration of other contractual arrangements, we nonetheless discussed the profound organisational changes in the recruitment, employment and monitoring of non-captive labourers in the nineteenth and early twentieth centuries. By contrasting continuities and changes in labour markets, we aimed at avoiding a schematic approach to the history of labour relations. Although specific contractual forms were more prevalent in certain periods – such as sharecropping in the 1850s, fixed payments in the 1870s, and colonato in the 1880s – the complex organisation of a plantation required a multitude of different actors employed under various labour regimes. Moreover, it was not uncommon to find the same actors working under different arrangements during the season, across plantations, and throughout their life cycles.⁴⁰

3 The Multidimensionality of the Vergueiro System and its Changes over Time (1840s–1870s)

Système du partage de récoltes (system of sharing crops); métayer brésilien (Brazilian metayage);⁴¹ share [of] profits and expenses of the crop; colons partiaires (partner colonists); systeme d’association (system of association); Halbpacht- or Parcerie-System (system of share tenancy or of partnership); Parceria-Kolonien or Halbpartsystem (“Partnership” colonies or half-part system);⁴² Parcerie-System nach dem Gesetz der Locação de serviços (Vermietungsrecht) or Halbpachtwesen (“Partnership” system following the Law of Locação de Serviços [Rental Law] or share tenancy);⁴³ Parcerie-Vertrag or Teilungsvertrag (“Partnership” contract or distribution contract); trabalhador parceiro na condição de um mero locador de serviços … remunerado pela divisão dos fructos (“partner” worker only leasing services … remunerated by crop division).⁴⁴ This great variety of denominations for “sharecropping” as first employed in the plantations of São Paulo in the 1840s–1850s indicates the novelty of the sharecropping contract and its complexity: the various names given above stress different dimensions of the contract, which were bundled together in a single arrangement.⁴⁵

What modern scholars summarise under the term “sharecropping contracts” (contratos de parceria) is probably better described by the term “Vergueiro system” (sistema Vergueiro), which was used at the time and referred to a contract that included at least four dimensions.⁴⁶

The contract originally proposed by Vergueiro & Co. – the firm headed by Senator Vergueiro – in 1847 included the application of crop-sharing mechanisms to three different stages of rural production. First, landowners were looking for a secure supply of labourers to take care of the coffee trees and to harvest the annual produce. Here, sharecropping acted as a labour contract that determined a fifty-fifty division of profits between the immigrant household and the farmer. Second, marketing the coffee required an initial processing of the beans on the farm, which included washing, drying, pulping, selecting and sacking them. In the 1847 contract, these tasks were divided between farmers and immigrant households via a system of crop-sharing – which implied a fifty-fifty division of the processing costs. Finally, immigrant families received plots of land to cultivate subsistence goods. The contract stipulated that the parcel of those goods sold on the market – that is, not consumed by the family itself – had to be shared with the landowner.⁴⁷ In this case, sharecropping was a form of land renting.

The term “sharecropping” in the singular thus hides the complexity of a contract according to which crop-sharing had a labour dimension, a cost-sharing dimension, and a land-rental dimension. The fourth dimension of the “Vergueiro system” was a credit interlinkage, by which the farmer covered the immigration costs of a household and bonded the labour force of all its members to the repayment of the outstanding debts.⁴⁸

In hindsight, one notices two clear problems with this mechanism. From the labourer’s point of view, the aspect of credit interlinkage implied too many uncertainties, especially for a family unaware of the working conditions on a Brazilian plantation.⁴⁹ From the farmer’s point of view, the negative effect of a long-standing debt on the motivation of the workers – and their resulting productivity or willingness to default or to riot – proved to be a challenge, even if it was not a concern at first.⁵⁰

In spite of these faults, the credit interlinkage became a cornerstone of the immigration policy in São Paulo. We consider it safe to argue that the aspect of interlinkage was the main mechanism that inserted Brazil into the age of mass migration, as it permitted the country to attract labourers who otherwise would not have had the economic conditions to immigrate. Brazil was thus able to attract immigrants without promoting institutional reforms, which would have been extremely costly for its ruling elites. Moreover, credit interlinkage provided landowners with a major control mechanism, securing the presence of labourers on the farm as long as there was an outstanding debt. By creating a certain degree of stability in the supply of labour, the credit dimension thus long outlived the first sharecropping contracts. The historical evidence and a theoretical illustration discussed by Bruno Witzel de Souza elsewhere point to the malleability of the credit dimension of the Vergueiro system. Under circumstances that most closely reflected the perceptions of the landowners about the contracts, it is possible to show, from a theoretical point of view, that the credit dimension could yield exactly the same benefits to the landowners, whether it was interlinked to a sharecropping contract, a fixed rent contract or a wage labour contract.⁵¹ It is thus not surprising that this malleable dimension continued to be used to bond immigrant labour for most of the nineteenth century. Over time, various landowners substituted a fixed indenture (usually five years) for the uncertain time of the varying outstanding debts. This mechanism reduced the uncertainty involved for both landowners and labourers, but the principle of bonding labour via the coverage of immigration costs remained unaltered.

From 1847 to 1851, Vergueiro & Co. maintained a monopoly on recruiting bonded labourers. It is thus reasonable to assume that no great modification in the structure of sharecropping occurred. In line with the concluding remarks of section 2, however, we highlight that sharecropping was not the only non-captive labour arrangement being employed. Before the relatively successful experience of 1847, Senator Vergueiro had already hired Portuguese labourers in 1840. The current evidence – even if still inconclusive – is that these labourers had been employed under regimes of fixed payments and fixed rents; and that their tasks were organised in a gang system considered too similar to that of the slaves.⁵² Moreover, it is reasonable to assume that sporadic free workers and clients had a number of verbal arrangements with the farms’ administrations.⁵³ As such, the period of the early transition to non-captive labour arrangements was as intense in contractual experimentation as any other that we will study in this chapter.

In the early 1850s, some substantial contractual changes appeared. Vergueiro & Co. started to recruit labourers for other landowners and tightened some control mechanisms, in special those related to the bonding of immigrants. A novelty particularly opposed by international authorities was a clause that allowed the transferral of a family with its debt to another landowner, a prerogative that critics saw as a form of trading freemen. Another source of distress was the new “solidarity clause,” by which each individual was responsible for the debt of his/her entire household and, hence, not allowed to stop working on the farm without consent of the farmer or until the debts had been fully repaid – a prerogative particularly harsh for young people who wanted to marry and for underage orphans. Moreover, farmers were now permitted to assign individuals to households they did not belong to by blood or kinship – and to whom the same principles of the “solidarity clause” applied.⁵⁴

Another crucial change occurred in the division of tasks. We have seen that the 1847 contract envisaged that farmers and labourers would contribute jointly to the processing of the coffee beans. In other words, the first arrangement stipulated that the fifty-fifty share was to prevail beyond the agricultural sphere and should be extended to the processing phase. It was probably daily experience on the farm that showed the unfeasibility of that earlier arrangement, with almost one hundred families simultaneously wishing to process their produce in the farm’s storehouses right after the harvest. The contracts of the early 1850s now stipulated that each household would pay a fixed amount per measure of coffee beans handed in to the farm’s administration – which became solely responsible for the processing operations. Beyond a formalisation of the monopsony power of landowners over the harvested produce, this clause had now established a clear functional division of labour between sharecroppers and the farm’s administration: sharecroppers were agricultural labourers, while the processing became a duty of the administration alone, without the supervision or actual participation of non-captives. Moreover, this new clause implied a fixed payment within a share contract. Although all these operations most likely involved only accounting registries rather than actual monetary disbursements, the change shows the malleability of sharecropping and, in particular, its ability to accommodate side payments – a crucial feature for some of the changes we will discuss below.

A number of marginal modifications appeared as more farmers decided to experiment with the Vergueiro system. These included, for instance, variations in interest rates, in the size and obligations of side payments, in charges of extra fees, and in benefits offered to labourers – such as the duration of rent-free housing or the size of pastures for their own use. Moreover, Senator Francisco Antônio de Souza Queiroz (the son-in-law of Nicolau Vergueiro) entered the business of recruiting immigrants in Europe with a sharecropping contract marginally more advantageous to labourers than those of Vergueiro. More importantly, Senator Souza Queiroz substituted the varying nature of the contract and the undetermined period of labour-tying based on households’ outstanding debts by a maximum indenture of five years. This implied that immigrant families could leave the farm upon the repayment of debts or after five years, which was certainly an asset for risk-averse labourers.⁵⁵

In the 1860s, farmers started to remunerate non-captive labourers with simpler arrangements of fixed payments per day or per piece rate. However, in spite of the declining number of farmers employing sharecropping contracts, this form of labour arrangement remained influential. For good or ill, the Vergueiro system was the non-captive arrangement with which farmers had had the most experience so far. Relatedly, some influential families, such as the Souza Queiroz family, maintained classical crop-sharing clauses in their new recruitments in Europe. Finally, defending immigration to Brazil against critics – who were especially vocal in the German-speaking world – was frequently confounded with defending prevalent labour relations. On the other hand, as political debates in the 1860s made Brazil move away from the idea of recruiting Chinese coolies, the aim of bonding European labourers kept its prominence.⁵⁶ In this context, discussions in the press demonstrate that there was much lively debate going on, with ideas coming from all sides on how to solve the “labour question,” that is, the reigning uncertainty on how to find new sources of cheap labour. In the 1870s, clearer propositions for crowding local markets with immigrants gained momentum, but, in practice, farmers only achieved marginally successful results based on individual initiatives to recruit labourers.⁵⁷

Between the experiments with bonded labour (1840s–1870s) and mass immigration (1880s–1920s), the number of contracts with fixed side payments per piece rate increased, especially when it came to taking care of the coffee trees (cleaning, weeding, hoeing, pruning), including on the main farm of the deceased Senator Vergueiro.⁵⁸ These side payments diminished the uncertainties faced by immigrants, whose main source of income under sharecropping in the 1850s had been a shared yearly amount of the proceeds of the coffee harvested – a source of income subjected to many uncertainties, such as variations in coffee prices in international markets and in productivity dictated by purely exogenous agronomical and ecological conditions.⁵⁹ Many of the new contractual arrangements also attempted to diminish the possibilities for disputes and doubts, explicitly writing down the tasks and obligations of the labourers, which the original Vergueiro system did not make explicit. Together with institutional reforms – including the centralised organisations to promote immigration as well as a new regulatory mark for labour legislation enacted in 1876 – the way was paved for the appearance of the colonato system.⁶⁰ However, as the next section shows, the spectre of the Vergueiro system would keep haunting the plantations in many different ways.

4 Not Set in Stone, but Still Path Dependent: Influences of Crop-Sharing in the 1900s–1930s

Between 1910 and 1912, the Inspectorate of Agricultural Defence – an auxiliary body of the Brazilian Ministry for Agriculture, Industry, and Commerce – conducted a survey on the rural conditions prevailing in Brazil. Besides agro-ecological and socio-economic data, the survey also collected information on the types of labour contracts employed in each county and the average remunerations for various labour categories.⁶¹ Unquestionably, this source has many shortcomings, particularly in its lack of clear definitions for the various labour categories. Notwithstanding, its data provide us with a unique overview of the various labour contracts that prevailed at the beginning of the twentieth century.

In table 1, we categorised the labour-rental contracts registered for the state of São Paulo. Please notice that the term “entries” in the notes accompanying the table refers to the number of times that a labour-rental arrangement was listed in the source and not to the number of counties where it prevailed, as most municipalities had more than one single arrangement listed. While they lacked a standardised methodology for listing the contracts, the Inspectore used an erratic combination of three categories to list the agricultural labour systems prevailing in a county. The type of labourer and the frequency of payment were registered in most cases, which allowed us to divide the table into two columns. Column A registers the categories of labourers (camarada, colono, meeiro, etc.), while Column B does not stipulate the type of contract, but the frequency of payment. Moreover, the Inspectore frequently mentioned some specific tasks or features of the contract (for instance, colonos who worked at a piece rate or in the cultivation of specific crops).

Even if unsurprising, a striking feature shown by Table 12.1 is the variability in labour-rental arrangements per category of worker and per frequency of payments, as well as the number of their possible combinations. As we have stressed throughout, contractual mix characterised the rural economy of São Paulo since the very first introduction of sharecropping in the 1840s.⁶² By the early twentieth century – with rapidly expanding agricultural frontiers – this complexity had certainly not diminished. Indeed, one sees no dominant contractual category among those listed in Table 12.1. The most frequently mentioned category was that of the empreiteiros, who executed tasks per piece rate (the so-called empreitadas), most frequently at their own risk. Nonetheless, even this category appeared only in approximately 30 per cent of the entries, with eight cases simply mentioning that their work was regulated by a “contract” and one even indicating that colonists executed tasks under empreitadas.

Somewhat more surprising is the relatively small number of colono arrangements listed in the survey. It is likely that many colonos in this survey were subsumed within the generic category of people working under “contracts” and “yearly contracts.”

The Inspectore also listed two types of contracts that had explicit crop-sharing features. First, sharecropping (parceria) appeared explicitly in thirteen counties, but only made up 2.67 per cent of all arrangements listed. Second, meeiros and meação contracts were present in forty-eight counties of São Paulo and corresponded to 10.21 per cent of all arrangements listed. The term “meação” refers to mechanisms of crop-sharing, but these arrangements could also refer to tenants, as was the case for the municipality of Campinas, for which the source mentions: “a few meeiros take care of some properties, having the rights to two-thirds of the production.”

Before advancing, a word of caution is due. Even a superficial reading of the Inspectorate’s survey reveals that many other labour-rental arrangements were employed on the farms but were not listed in the survey, probably because it was impossible for the Inspectore to cover all the farms in the regions they visited – some of which were extremely vast and difficult to reach. For forty-six out of the 173 counties of São Paulo, the published results mentioned “etc.” after the last labour-rental arrangement listed, that is, the source stopped short of mentioning all “labour systems” being used in those regions (which also calls into question the completeness of the other counties). Moreover, there were internal inconsistencies in the source. For instance, under the rubric “labour systems,” the colono contract was listed for twenty-two counties; however, this type of contract was mentioned in thirty-three counties under the heading “salaries.”⁶³ Thus, the reader should consider the data above with caution: the survey illustrates the complex contractual mix prevailing at the beginning of the 1910s in broad strokes, but it does not provide a final picture of all labour arrangements prevailing in a particular locality, which can be gained only through the study of contracts prevailing in specific regions – the theme of the rest of this section.

Determining why crop-sharing arrangements prevailed in certain localities is a complex issue that this chapter will not deal with. Nonetheless, figure 1 shows the geographic spread of the parceria and meação contracts mentioned above across the counties of São Paulo in the early 1910s.

View Full Size

Sharecropping crop-sharing mechanisms across counties of São Paulo 1910–1912

Notes: (1) Municipal frontiers from 1911; (2) Missing data for Salto Grande do Paranapanema and Ibiquara

source: same as in table 1, compiled per county, instead of by number of entries

• Download Figure
• Download figure as PowerPoint slide

By discussing how contracts changed as part of a learning process that, in many cases, can be traced back to the first experiences with non-captive labour, we will see how the sample of contracts researched for the 1910s–1930s had become increasingly more detailed.⁶⁴ Notwithstanding, we warn against three misinterpretations regarding the concept of “learning” in contractual design. First, “learning” does not necessarily imply linearity. Indeed, many experiences in the twentieth century paralleled problems already faced in the mid-nineteenth century. Second, while the coffee economy was the motor of the Brazilian economy, the state of São Paulo had different regions in terms of rural specialisation.⁶⁵ Technical conditions dictated by different crops influenced contractual adoption. Finally, the contractual mix in the state implies feedback effects between different arrangements.⁶⁶ As such, sharecropping not only influenced other labour arrangements, but was also influenced by them.

With these remarks in mind, the most important feature of sharecropping and colonato contracts signed in the early twentieth century – in many cases under the supervision of state organisations, such as the Patronato Agrícola – was their level of detail when it came to settling disputes. Landowners attempted to stipulate more clearly in the contracts the behaviour they expected from labourers. For most of the nineteenth century, sharecropping contracts had determined that labourers should accept the extra-contractual bylaws of the farms where they lived. Although written regulations did exist on the largest and most important plantations, on the smaller and less well-organised farms written regulations were less important than the will of the landowners or patronage practices.⁶⁷ Many of the stipulations in the bylaws were incorporated into the contracts themselves at the beginning of the twentieth century. Concretely, this implied an increase in the number of punitive clauses against what farmers considered misbehaviour. For instance, a farm in the county of Pirajuí stipulated that “recidivism of abuses and insubordination against the discipline on the farm” – broadly defined – were sufficient for expelling a family without monetary compensation.⁶⁸ In the county of Novo Horizonte, a 1928 contract returned to a question that had been controversial at least since the Sharecropper’s Riot, namely the discretionary power of the landowner, “during week days,” to impede the entrance onto his property of anybody whom he judged improper. The same contract dealt with the recurrent problem of labour mobility and turnover: the meeiro was not allowed to work for anybody else if his/her tasks were not in accordance with the “good rules” of agricultural production.⁶⁹

In a similar vein, contracts started to distinguish their various dimensions more clearly. Following a trend that had begun in the 1870s, farmers attempted to establish clearer regulations regarding the leasing of plots of land for labourers’ subsistence cultivation within the farms.⁷⁰ The labour-related aspects of the contracts also started to be set out more precisely, mainly via clearer stipulations of the remunerations for each task performed by the workers, probably as a form to diminish their bargaining power between agricultural seasons.

As noted in Section 3, adding side payments to sharecropping contracts diminished the uncertainties surrounding labourers’ annual remuneration. The labour dimension of sharecropping combined with peonage was risky, as the amortisation of the debt depended on the annual harvest, which was subjected to unpredictable natural conditions and economic oscillations in a number of markets. Certainly, sharecropping could generate extraordinary gains if coffee prices increased, but the reverse was also true.⁷¹ Using data from Ribeirão Preto and Franca, Rogério Faleiros showed that sharecropping could lead to a negative variation in labour remuneration of about 35.2 per cent, while variations in empreitadas reached a mere 1.7 per cent.⁷² As such, the mixed form of sharecropping with side payments provided a middle ground for the trade-off between expected gains and contractual risks.

A number of sharecropping contracts also started to stipulate shares that varied according to short-term junctures. In 1929, farmer João Pires Monteiro set a remuneration floor for his sharecropper: profit sharing would take place only if the price of coffee was above a certain threshold (20 mil-réis per arroba); otherwise, the labourer would pocket the yields.⁷³ A similar principle was applied on Figueira farm in 1919 following the 1918 frost that had severely damaged the coffee trees. The farmer set a quantity floor: a household would be entitled to one hundred per cent of the profit if the amount harvested was lower than one thousand arrobas; at that exact level, the family would receive two-thirds of the profit; the usual fifty-fifty share applied only if the quantity produced by the household surpassed that threshold. In 1919, farmer Vicente Ferreira applied crop-sharing for two agricultural products, setting a different floor for each. For cotton, the fifty-fifty share would be applied only if the price reached a minimum of 10 mil-réis per arroba. For coffee, the farmer stipulated a complex system of varying shares: given the expected low productivity of new coffee trees, households would earn one hundred per cent of the profit from trees younger than three years; the fifty-fifty share would apply only from the fourth year onwards – and it seems that this contract had a fixed duration.⁷⁴ Naturally, not all farmers were as responsive to labourers’ demands for more security in moments of crisis, as was illustrated by a 1920 empreitada contract that dictated that labourers would receive no compensation for the coffee trees lost in the 1918 frost, putting the entire burden of the risk of natural disasters on the shoulders of the labourers.⁷⁵

In a similar vein, one notices a preoccupation in defining the participation of each contracting party in the costs of processing and transporting the produce; if labourers were expected to participate in the sharing of costs, contracts now meticulously defined such operations and the size of the side payments.⁷⁶ To illustrate, in 1918, the proprietors of Reunidas Irmãos Camargo farm, in the county of São Carlos, still charged a side payment of 400 réis per arroba of harvested coffee as the labourers’ contribution to the processing of the beans – the fact that this amount was identical to that specified in the Vergueiro system is by chance, we assume. A 1935 contract from the county of Jaú also determined that a side payment should be made, but updated the prices and added another 1,000 réis for a general category of expenses.⁷⁷ Interestingly, another farmer in the same municipality felt the need to stipulate contractually that the processing of the beans must occur on the farm where they had been harvested, suggesting that transportation costs in the 1930s had fallen sufficiently to allow labourers to process their harvests in different farms or localities.⁷⁸

Of course, not all forms of clearer contractual specification meant improvements in actual labour relations. Ironically, in one of the contracts found, the tendency to incorporate more detailed clauses was actually used to increase the discretionary power of the landowner. In 1925, in an empreitada for the planting of new coffee trees, a farmer in the municipality of Novo Horizonte determined a weekly period when labourers had to work for the maintenance and repair of his property, without remuneration.⁷⁹ This had been a thorny issue since the 1850s. In the past, due to ingrained patron-client relations, it had been unimaginable to landowners that labourers would dare to refuse to execute a task because it was not established in a contractual relation; but to the outrage of many farmers, this was precisely what happened. To circumvent this problem, the 1852 contract by Senator Souza Queiroz had outlined a list of tasks that labourers should execute – such as maintaining the water pipelines used for washing coffee beans.⁸⁰ The Central Association for Colonisation discussed this contractual response in 1854 and determined that sharecroppers should not be employed in other tasks than those they had been hired for.⁸¹ It is remarkable that seven decades later, with slavery gone and mass immigration settled, one landowner stipulated a clause that, in practice, meant a return to his discretionary power on how to allocate labour within his farm.

5 Concluding Remarks

The ubiquity of sharecropping as a labour-rental arrangement that for long has been considered less efficient than other available options is a theme that has fascinated generations of historians and social scientists. This chapter touched upon some of these debates by revisiting a classic historiography that deals with the adoption and evolution of sharecropping contracts in the plantations of São Paulo, Brazil, where sharecropping was the first relatively successful arrangement to make use of non-captive labour, usually foreign-born, in the early transition from slavery to free labour.

The first part of the chapter studied the multidimensionality of the original sharecropping contract in the coffee plantations of São Paulo. The Vergueiro system – as it was known by contemporaries – included a labour dimension that involved the sharing of profits from coffee harvesting; a cost-sharing mechanism for processing the coffee beans; a land-rental dimension for the crop-sharing of subsistence goods; and a form of credit interlinkage that allowed for the immigration of poor and otherwise credit-constrained Europeans. In its second part, the chapter discussed how problems with the enforcement of free labour relations led to contractual refinements at the beginning of the twentieth century – a process that was not exempt from many drawbacks. Particularly important was the incorporation of side payments into crop-sharing contracts, which lowered the uncertainties of the contracting parties.

By linking the periods of experimentation with non-captive labour (1840s–1870s) to that of consolidated mass immigration (1900s–1930s), this chapter discussed the history of contractual design within an institutionalist framework, highlighting the importance of path dependence for understanding contractual choice and enforcement. Moreover, this study aimed to show the variety of arrangements that have coexisted even in geographically narrow spaces. The evidence shows that rural labour relations are complex now and that there is no reason to hypothesise that they were simpler in the past. Adding new pieces to this already immense puzzle might blur some pictures we currently have of historical contractual design, but it will certainly contribute to a more accurate depiction of past labour-rental relations.