Land reform and agrarian conflict in 1930s Spain

Jordi Domènech (Universidad Carlos III de Madrid) and Francisco Herreros (Institute of Policies and Public Goods, Spanish Higher Scientific Council)

Government intervention in land markets is always fraught with potential problems. Intervention generates clearly demarcated groups of winners and losers as land is the main asset owned by households in predominantly agrarian contexts. Consequently, intervention can lead to large, generally welfare-reducing changes in the behaviour of the main groups affected by reform, and to policies being poorly targeted towards potential beneficiaries.

In this paper (available here), we analyse the impact of tenancy reform in the early 1930s on Spanish land markets. Adapting general laws to local and regional variation in land tenure patterns and heterogeneity in rural contracts was one of the problems of agricultural policies in 1930s Spain. In the case of Catalonia in the 1930s, the interest of the case lies in the adaptation of a centralized tenancy reform, aimed at fixed-rent contracts, to sharecropping contracts that were predominant in Catalan agriculture. This was more typically the case of sharecropping contracts on vineyards, the case of customary sharecropping contract (rabassa morta), subject to various legal changes in the late 18th and early 19th centuries. It is considered that the 1930s culminated a period of conflicts between the so called rabassaires (sharecroppers under rabassa morta contracts) and owners of land.

The divisions between owners of land and tenants was one of the central cleavages of Catalonia in the 20th century. This was so even in an area that had seen substantial industrialization. In the early 1920s, work started on a Catalan law of rural contracts, aimed especially at sharecroppers. A law, passed on the 21st March 1934, allowed the re-negotiation of existing rural contracts and prohibited the eviction of tenants who had been less than 6 years under the same contract. More importantly, it opened the door to forced sales of land to long-term tenants. Such legislative changes posed a threat to the status quo and the Spanish Constitutional Court ruled the law was unconstitutional.

The comparative literature on the impacts of land reforms argues that land reform, in this case tenancy reform, can in fact change agrarian structures. When property rights are threatened, landowners react by selling land or interrupting existing tenancy contracts, mechanizing and hiring labourers. Agrarian structure is therefore endogenous to existing threats to property rights. The extent of insecurity in property rights in 1930s Catalonia can be seen in the wave of litigation over sharecropping contracts. Over 30,000 contracts were revised in the courts in late 1931 and 1932 which provoked satirical cartoons (Figure 01).

Figure 1. Revisions and the share of the harvest. Source: L’Esquella de la Torratxa, 2nd August 1932, p. 11.
Translation: The rabaissaire question: Peasant: You sweat by coming here to claim your part of the harvest, you would be sweating more if you were to grow it by yourself.

The first wave of petitions to revise contracts led overwhelmingly to most petitions being nullified by the courts. This was most pronounced in the Spanish Supreme Court which ruled against the sharecropper in most of the around 30,000 petitions of contract revision. Nonetheless, sharecroppers were protected by the Catalan autonomous government. The political context in which the Catalan government operated became even more charged in October 1934. That month, with signs that the Centre-Right government was moving towards more reactionary positions, the Generalitat participated in a rebellion orchestrated by the Spanish Socialist Party (PSOE) and Left Republicans. It is in this context of suspension of civil liberties that landowners now had a freer hand to evict unruly peasants. The fact that some sharecroppers did not surrender their harvest meant they could be evicted straight away according to the new rules set by the new military governor of Catalonia.

We use the number of cases of completed and initiated tenant evictions from October 1934 to around mid -1935 as the main dependent variable in the paper. Data were collected from a report produced by the main Catalan tenant union, Unió de Rabassaires (Rabassaires’ Union), published in late 1935 to publicize and denounce tenant evictions or attempts of evicting tenants.

Combining the spatial analysis of eviction cases with individual information on evictors and evicted, we can be reasonably confident about several facts around evictions and terminated contracts in 1930s Catalonia. Our data show that that rabassa morta legacies were not the main determinant of evictions. About 6 per cent of terminated contracts were open ended rabassa morta contracts (arbitrarily set at 150 years in the graph). About 12 per cent of evictions were linked to contracts longer than 50 years, which were probably oral contracts (since Spanish legislation had given a maximum of 50 years). Figure 2 gives the contracts lengths of terminated and threatened contracts.

Untitled 2
Figure 2. Histogram of contract lengths. Source: Own elaboration from Unió de Rabassaires, Els desnonaments rústics.

The spatial distribution of evictions is also consistent with the lack of historical legacies of conflict. Evictions were not more common in historical rabassa morta areas, nor were they typical of areas with a larger share of land planted with vines.

Our study provides a substantial revision of claims by unions or historians about very high levels of conflict in the Catalan countryside during the Second Republic. In many cases, there had a long process of adaptation and fine-tuning of contractual forms to crops and soil and climatic conditions which increased the costs of altering existing institutional arrangements.

To contact the authors:

EHS 2018 special: Wine prices in Anglo-Gascon trade, c.1337-c.1460

Robert Blackmore (University of Southampton)

Tacuina sanitatis (XIV century)
9-autunno,Taccuino Sanitatis, Casanatense 4182. Available at <,Taccuino_Sanitatis,_Casanatense_4182..jpg>

Episodes of major market volatility are rarely out of the headlines today. Their ramifications, though considerable, are discussed as if these were somehow new, and that they are the result of how economies are structured in our globalised world. Yet prices in international markets in the late middle ages could be just as volatile and have just as far-reaching consequences.

The wine trade between Gascony and England is one key example. Gascony, in modern southwestern France, was part of the medieval duchy of Aquitaine: a territory ruled by the English crown almost without interruption from 1154 to 1453.

Geography and geology permitted the production of just one commodity, wine, and as a result the region was dependent, like so many modern states specialised in fossil fuels or mining, on export earnings to pay for the purchase and import of food and all other goods from distant markets.

My research provides a better understanding of the possible factors that influenced fluctuations in prices, and their knock-on effects. To achieve this, I use wholesale prices in Bordeaux and Libourne from between 1337 and 1466, largely sourced from surviving original documents stored both in the Archives départementales de la Gironde in Bordeaux and the National Archives in London.

As today, extreme climactic events, as well as disruption by war, or demographic catastrophes such as disease or famines, can be understood to cause sudden shifts in supply. Likewise there were abrupt changes in local demand, for example, in 1356 the arrival of a victorious Edward, the Black Prince, with his army laden with ransoms and plunder after the battle of Poitiers, can be observed in the data.

Volatility was exacerbated by government intervention: particularly a 1353 English law that had constrained certain merchants from buying up stock in advance at pre-agreed prices, as would be done in modern markets. Likewise, ill-considered price controls at retail in England probably caused suppressed trade.

Critically, wine was a luxury in northern European ale-drinking societies, where only the rich would tolerate high prices, so any brief disruptions in supply or local demand disproportionately affected the level of exports.

Such characteristics also meant that wine prices were responsive to wider economic shocks in ways that would be well understood today. Monetary policy mattered. England and Gascony used different currencies with a changing exchange rate. As the Gascon livre appreciated against sterling in the two decades after the Black Death (1348-9), prices rose for foreign buyers, then later devaluations, such as in c.1370, 1413-4 and c.1440, made purchases suddenly cheaper, and triggered noticeable increases in English wine imports.

Yet, for Gascony, as in Venezuela today, an over-dependence on foreign imports meant such surges or falls in the value of one single exported commodity resulted in sudden strong trade surpluses or deficits. Foreign currency, then in the form of precious metals, poured in and out of the economy with fluctuations in the wine trade.

This made prices, and by extension, the duchy of Aquitaine’s whole economy, even more unstable. In the end inflation set in as production declined and later years of English Gascony were mired in an economic depression that contributed to the region’s loss to the French crown in 1453 at the end of the Hundred Years’ War.

Legacies of inequality: the case of Brazil

by Evan Wigton-Jones (University of California, Riverside)

The Rio Team. In Kidder, D.P., Brazil and the Brazilians : portrayed in historical and descriptive sketches, Philadelphia 1857. Available at


 Recent years have witnessed a renewed interest in issues of economic inequality. This research offers a contribution to this discussion by analysing the effects of inequality within Brazil.

Firstly, it shows that the climate is a key determinant of long-run inequality in Brazilian context. It uses data from a national census conducted in 1920 to show that warmer regions with high rainfall were characterised by plantation economies, with a wealthy agricultural elite and a large underclass of poor labourers. In contrast, cooler and drier areas were conducive to smaller family farms, and hence resulted in a more equitable society.

The study then uses information from the 2000 census to show that this local inequality has persisted for generations: areas that were historically unequal in 1920 are generally unequal today as well.

Finally, the research shows that greater long-term inequality inhibits regional development. It also shows evidence that inequality affects local governance, as municipal spending on health, education and welfare is significantly lower in more economically unequal areas.

To show the climate’s influence on local inequality, the study created an index that quantifies the relative suitability of land for plantation agricultural production. The metric is based on the temperature and precipitation requirements of different crops that are uniquely plantation or smallholder in their method of production. For example, sugarcane has historically been produced on large plantations, while wheat was often cultivated on small farms.

The research then shows that localities with a favourable climate for plantation agriculture contained a more unequal distribution of land. To measure the concentration of land ownership, it calculates a Gini index – a standard measure of inequality that ranges from 0 (perfect equality) to 100 (one individual holds all land).

As Brazil’s economy was predominantly agrarian in 1920, this distribution of land is a good proxy for that of income and wealth. The research combines this with data on municipal spending in the 1920s to show that local governments with higher land inequality spent less on education, health, public goods and public electricity. For example, a one unit increase in the Gini index is associated with a .76 percentage point decline in such spending.

The effects of this inequality have ramifications for contemporary socio-economic welfare in Brazil. Not only has local inequality persisted throughout the twentieth century, but it has also hindered present-day municipal development. Here it measures local development using the municipal-level human development index (HDI) – a metric that accounts for education, public health and income – for the year 2000.

It shows that historically unequal areas score much lower on the HDI: a one unit increase in 1920 land inequality is associated with a reduction of .38 points in this index (which, like the Gini index, is measured on a scale from 0 to 100, with a higher score indicating greater development).

Furthermore, the legacies of historical inequality are still manifest in contemporary local governance: a one unit increase in historical inequality is associated with a .49 percentage point decrease in municipal-level welfare spending for the year 2000.

These findings suggest several important conclusions:

  • First, the environment may play an important role in determining inequality and long-term development, even within countries.
  • Second, economic disparities can persist for generations.
  • Lastly, inequality can have a corrosive effect on welfare and governance, even at a local level.

It should be noted, however, that this study has focused on inequality within Brazil. The extent to which these findings can be generalised to other settings requires further study.

Political Institutions Shaping Economic Outcomes: Land Tenures in Colonial Sind 1843-1920

by Tehreem Husain

Interactions of political and economic institutions and their ramifications on development outcomes have been recognised by academics and policymakers alike. This blog analyzes the principal-agent relation between the British coloniser and local landlords and peasants in British Sind during 1843-1920. It argues that changes in political institutions during the period affected economic institutions, which through path dependence persists today.


Research in the area of comparative institutions and economic development points to the fact that political institutions once in place, persist and shape the political-economic interactions between different groups and agents. Moreover, past institutional frameworks also have a degree of influence on the direction that institutional change takes place (Acemoglu and Robinson, 2008). This has been the case in British Sind (part of Bombay Presidency till 1935) as well. Sind’s primarily agrarian societal structure was based in powerful landlords who held large tracts of land with peasants having little or negligible ownership rights. The institution of land tenures – a term which encompasses rights of land occupancy, land revenue collection and land ownership – did not only impact agricultural productivity and welfare outcomes during the period under study but can still be felt today.

Acemoglu, Johnson and Robinson’s seminal paper on institutions (2005) has highlighted the instrumental role that land tenures and property rights play in determining social outcomes and trajectory of economic growth of a group, region or a nation. This is applicable to British Sind too. During 1843-1920 system and laws regarding land revenue and tenures were taking roots in the region. Overall, three different forms of land tenure systems were introduced in India; landlord-based system (zamindari), an individual cultivator based system (ryotwari), and a village-based system (mahalwari). Selection of a system was mainly defined by the actual or prospective land revenue from an area. The importance of choosing a specific land tenure system and hence extraction of land revenue from a region can be gauged from the fact that by the mid-nineteenth century land revenue contributed more than 50 percent and even seventy years later by 1920 more than 40 percent to the total revenue of British India.

In ensuring smooth collection of land revenue, the governance structure adopted by the British aligned itself to the indigenous societal structure using the local landlords as ‘intermediaries’, usually those who had ‘traditional’ or ‘customary’ authority. The colonial state maintained de jure ‘direct rule’ over the territory however in reality coercion was enforced by intermediate local political elites who operated outside the bureaucratic-rational apparatus of the state (Naseemullah and Staniland, 2014). The local landlord was made powerful firstly by granting them revenue-free lands which were heritable, and secondly, by giving them powers to collect revenue. This was done in exchange of curtailing any political and social resistance against the British. On the other hand, unlike in other parts of India where agricultural tenants had occupancy rights (Swamy, 2011), tenants in Sind would till the land and meet conditions that the landlords may impose on them from time to time without any land-ownership (Hughes, 1876).

Granting local landlords rent free lands and special privileges of land revenue collection fortified the extant hierarchical societal structure and was broadly aimed at establishing and perpetuating British rule through the institution of land tenures. Land revenue and administration records exhibit that 19.5% of land amongst large tracts of land (500 acres or more) had rent-free status in Sind-the highest in the entire Bombay Presidency. Moreover, legislative acts were also passed during this period to ensure that the power and influence the landlords wielded was not undermined.

Interestingly, upon Sind’s annexation to the empire in 1843, the British desired to deal directly with the cultivator and implemented the ryotwari system of land tenures. However, they soon realized the local landlords were wielding enormous authority over the peasants living with little or no land rights. Hence ryotwari system converged closely to a zamindari system; though official records continued to recognize it as the former. Consequently, the utilitarian nature of ryotwari system was destroyed by trading rights of the peasants for achieving political gains.

The approach of granting rent free lands and closely following the zamindari system of land tenures was at odds with the colonial power’s fiscal target of improving public finances from this area for the larger aim of achieving political expediency. The fact that a significant portion (87 percent) of revenue was alienated in Sind relative to rest of Bombay Presidency is evidence of this claim. Moreover, it also impacted agricultural productivity. This can be ascertained from the fact that alienated land had the lowest yield. Colonial records also give evidence to the claim that revenue per acre from alienated land was quite low in Sind and was falling. More importantly, incidence of revenue in non-alienated land was highest in Sind. This shows that the incidence of revenue was primarily on tenants and on small landlords. As much as the analysis of historical land tenure systems in Sind gives insight on how economic institutions were influenced and shaped it also serves as a social premise on Sindi society which to this day has largely been unchanged relative to what it was more than 150 years ago.

Furthermore, comparison of land revenue and administration records from Sind to rest of the Bombay Presidency suggests that the land grants to the landlords in Sind were very pronounced relative to other districts in the Presidency. Overall, analysis of land revenue and administration records from 1843-1920 highlight three aspects of economic history of Bombay Presidency in British India.

First, it argues that the, like in other parts of the colony, British colonizers exploited agency relation to govern Sind and used local landlords as their agents. The interesting part is that they built principle-agent relation by first confiscating the whole area and then making grants of large tracts of heritable rent­-free land to the old rulers and landlords. These land grants were the highest in Sind compared to rest of the Bombay Presidency and reinforced the existing social order through land tenure system. Primary purpose of this approach was to ensure smooth governance in the province.

Second, using land tenures as an instrument to achieve political expediency, they implemented ryotwari land tenure system on paper which, however, in spirit, was more like the zamindari system. This has implications for earlier work on historical land tenures in India (for instance Banerjee, 1985) wherein official records on land tenures have been considered as the practiced one. This needs to be tested for other parts of British India. This impacted agricultural productivity and revenue outcomes.

Third, although land granted as jagirs, inam etc. was done elsewhere in the Bombay Presidency (of which Sind was a part till 1935), this article argues that this was carried out in the harshest form in Sind where the tenant had no occupancy rights unlike elsewhere in the Bombay Presidency. The impact of this was reflected in low agricultural yields from alienated land in Sind relative to the same category elsewhere in the Bombay Presidency. Moreover, these effects are still being felt today and these institutions have permeated through time and shown path dependence. Researchers using data from the latest agricultural census of Pakistan 2010 have shown that inequality in terms of land ownership has increased through time in Sind.