Analysis of the political consequences of economic development has a long history. Modernizing economic transformations, such as industrialization, create both losers and winners, raise popular expectations, and can result in conflict and political instability.
Late Imperial Russia, which underwent rapid industrialization, exemplifies these dynamics. An important part of the revolutionary events of 1917 was turmoil in rural areas, which ended up in a large-scale occupation and redistribution of privately-owned land, be it the nobility’s landed estates or the recently privatized parcels of former commoners, in favor of peasant communes. Better understanding of the roots of the Russian Revolution is important not only from a purely historical perspective; it enables us to shed more light on the causes of conflict in other developing societies.
Peasant unrest in 1917 did not come out of the blue. It was preceded by an extraordinary upsurge in agrarian conflict during the 1905–1907 revolution, including destruction and burning of landlords’ estates by peasants wishing to expel their owners. An earlier major wave of peasant disturbances occurred in 1861-1862, after emancipation of the serfs, and was likely provoked by peasants’ dissatisfaction with implementation of this reform: see Scott Gehlbach and Evgeny Finkel’s post in this blog. In between these major events, from 1863 to 1904 and again from 1908 to 1916, simmering discontent and periodic spurts of peasant unrest targeting private landowners, state officials, or other peasants remained an endemic feature of rural life.
My dissertation, which won the Ronald H. Coase Award for best dissertation in institutional and organization economics, explores the role of industrialization in producing a “conflictual environment” in the Russian countryside. My thesis can be illustrated by a case study of a prominent episode of unrest.
In March 1902, a large-scale peasant revolt suddenly erupted in Poltava and Kharkov provinces (now part of Ukraine), situated in the fertile black-soil belt. The disturbances, which affected 174 communes with 38,000 peasants, lasted about two weeks and were suppressed only with the help of several army detachments. This revolt and the subsequent court trial of its participants attracted much public attention; it was a harbinger of the 1905–1907 rural havoc. The details can be found in published archival documents.
The revolt started on the territory of the giant Karlovka estate occupying more than 150,000 acres – about one tenth of the area of the Konstantinograd district of Poltava province in which it was located. The estate was owned by descendants of the Great Duchess Yelena Pavlovna, one of the originators of the 1861 emancipation reform, and could boast advanced eight-field farming, several industrial facilities processing its agricultural products, a railroad station, and fifteen schools. The revolt was preceded by “fermentation” among peasants, the spread of anti-regime propaganda, threats to landowners, and occasional acts of arson and theft, but the actual event seems to have started spontaneously. Estate managers were distributing food assistance to peasants stricken by a local potato harvest failure, and disaffected recipients, growing in numbers, proceeded to open plunder of estate barns.
The disturbances spread across district and provincial borders. Groups comprising several hundred peasants broke into private farmsteads, seized grain, potato, hay, cattle, agricultural tools, utensils and furniture, and occasionally damaged farmstead buildings and landlord houses. Rebellious peasants openly challenged local landowning elites and claimed that they “now owned all the land and everything that grew on the land”. Interestingly, peasants also targeted sugar refineries. One of them, in Kharkov province, was plundered and pillaged, apparently with an aim of stopping factory activities. According to the Kharkov governor’s report, “What the looters could not take with them, namely, machines and, in general, more or less bulky things, was destroyed or put into disrepair”. This was not the end of the story: in 1905 a long-planned sugar refinery was built on the Karlovka estate; local peasants immediately attacked it and other industrial facilities, causing a temporary halt of their work.
By the turn of the twentieth century, sugar factories were mainly located in three regions: Ukraine and Bessarabia, the Central Black Earth Region, and Poland. Each factory was supplied by sugar beets grown on nearby “plantations” owned either by the factories themselves or by local landowners, including rich peasants. The total amount of land allocated to sugar beets was nontrivial and continued to expand. According to my calculations, in 1897 it accounted for about 4% of the total cultivable land area in Ukrainian Kiev and Podoliya provinces, and about 2% of it in Kharkov and Kursk; by 1911 these figures had more than doubled.
Sugar-beet plantations competed for land with peasants still engaged in subsistence farming and depending on rented land. According to the testimony of one local official, “with the building within the territory of the Valki and Bogodukhov districts [of Kharkov province] of two sugar factories, the area of the land leased out to the peasants narrowed considerably, and land rent prices went up”. Both of these factories were raided by peasants in 1902. It should not come as a surprise that South-West Ukraine and Kursk province, known for the spread of sugar factories and plantations, were among the leaders in peasant unrest in 1905 and 1906.
We can contrast this situation with recent works finding that the proliferation of cash crop plantations (sugar or cotton) intensifies labor coercion policies in the interests of plantation owners wishing to maintain a supply of cheap labor. In the case considered here, the presence of smallholder subsistence farming as an outside option for peasants was potentially destabilizing because of resistance to market-driven reallocation of land to activities more profitable for landlords (reminiscent of the famous Scott–Popkin and Brenner debates).
This logic can be generalized: industrial development that competes with local peasants for land and other agricultural inputs, but does not provide them with commensurate employment opportunities compensating for adverse effects on subsistence farming, can provoke peasant unrest. On the other hand, if industrialization offers attractive outside options and puts relatively less strain on agricultural resources, its influence on peasant unrest can be negative. The first type of industrialization in the Russian Empire was represented by food processing, which in addition to sugar production included alcohol refining and grain milling. The second type of industrial development was embodied by textile manufacturing, which primarily relied on raw materials imported from abroad or delivered from peripheral regions of the empire. These two industries, together accounting for most industrial output and employment, are central to my analysis.
For my project I collected district-level data on industrialization (measured by the number of workers) and peasant unrest from 1879 to 1904, covering the fifty core provinces of European Russia. Data on industrial enterprises, which I aggregated at the district level, were taken from industrial censuses and surveys (part of these latter data, independently collected by Amanda Gregg, are now available here). Peasant unrest was coded on the basis of a “chronicle of the peasant movement” published in Soviet times (data from this source for other years have been used in various other works). I measure the outcome of interest as the number of conflict events in a district-year; industrialization is expressed as the logarithm of the number of industrial workers per 1000 people in a district.
Analysis of these panel data reveals that the number of textile workers is negatively correlated with peasant unrest, whereas areas with a more developed food industry experienced more intensive disturbances: an increase of one standard deviation in the number of textile (food) workers corresponds to a fall (rise) in the occurrence of peasant unrest by 9-18% (13-20%) of its mean. As expected, this relationship is much stronger in districts with a lower share of peasant-owned land, that is, with a greater potential for conflict between communal smallholders and large private landowners. The results are robust to the inclusion of a broad set of controls–yet considerably weakened when district fixed effects are added, the likely consequence of the fact that cross-district variation in industrialization far exceeds within-district variation.
In order to further address identification issues, I perform a cross-sectional analysis at the sub-district (parish) administrative level for Kursk province in 1905 and 1906, using data from Burton Richard Miller, who in Rural Unrest During the First Russian Revolution carefully compiled a list of settlements affected by peasant disturbances. The 1905 Russian Revolution swept across the whole empire and affected all social strata. Its sheer scale has precluded the systematic and comprehensive collection of fine-grained information on unrest episodes of the sort used for other periods. In-depth regional studies, such as Miller’s monograph, come in particularly handy here.
An important advantage of considering this period is that peasant unrest was launched by (and synchronized with) exogenous revolutionary events (e.g., Bloody Sunday in January 1905, the General October Strike, and the issuance of the October Manifesto granting political representation and broad civil and political rights). Before 1905, in Kursk province there were only sporadic disturbances without any discernible trends for more or less industrialized parishes.
It turns out that regardless of whether we use a binary indicator of unrest or the number of affected villages in a given parish, the results are consistent with the district-level analysis. A stronger presence of food industry (which constituted the lion’s share of industrial employment and output in the province) is positively associated with disturbances, such that an increase of one standard deviation in the number of food workers adds 10 percentage points to the probability of unrest (which is experienced by half of all parishes). This effect is smaller in parishes better endowed with allotment land. The correlation between unrest and presence of non-food industries is negative and insignificant. Moreover, if we focus only on sugar manufacturing, we can observe that proximity of a sugar refinery is positively associated with unrest.
These findings demonstrate that the character of industrialization, given the existing institutional framework and land distribution patterns, was consequential for peasant unrest. The context could have been critical: as Tracy Dennison notes in her blog post, many of the institutional features of serfdom protecting landlords’ privileges persisted long after emancipation. This lasting interdependence could help perpetuate landlord-peasant conflict, further exacerbated by the intrusion of modernizing influences. It echoes research showing that communal land tenure and tensions between private landowners and peasant smallholders affected labor-allocation decisions and the political behavior of peasants up to- and beyond the end of the Russian Empire.
My work also resonates with a recent post by Jennifer Brick and Ilia Murtazashvili, who argue that the political economy of property rights, state capacity, and conflict is a vibrant field of study. By including in the analysis economic and social development, we can enrich our knowledge of the nature of political and social instability. Future research may benefit from the comparison across space and time of industrialization experiences, the state of the agricultural sector (in particular, property-rights regimes and land distribution), and the management of social conflict, with the promise of lessons for contemporary developing countries.