Substitute Solutions to the Agrarian Problem

Whatever their other differences, premodern economies were overwhelmingly agrarian. Because of low productivity in agricultural production, urban concentrations of population and non-agricultural economic activity required food supplied by hinterlands that dwarfed them in population. For this reason, in the early stages of economic development, the agrarian sector potentially constrains growth in three ways (Mundle Reference Mundle1985). The first is simply quantitative: as long as agriculture represents the bulk of production, the rate of productivity growth of agriculture will strongly influence the overall rate of growth of the economy. The second is a constraint on the supply of food: an expanding industrial workforce needs to be fed by a farm population decreasing in relative size. The third potential constraint is on demand: especially if it is relatively uncompetitive internationally, the industrial sector needs the rural population to be sufficiently productive and market-oriented to purchase its products. It is frequently argued that, historically, these barriers to industrialization were only overcome in economies that underwent “agrarian modernization,” which entailed the “decomposition of feudal relationships of production” (Senghaas Reference Senghaas and Kimmig1985: 53). The analysis of these barriers and of the social transformations necessary to overcome them forms the core of the classic “agrarian question” of economic development (Byres Reference Byres, Breman and Mundle1991).

A broad family of accounts of both Japan and Russia, which for a time were dominant in the historiography of the two countries, argued that they began processes of catch-up industrialization while in effect circumventing the agrarian question. A paradigmatic statement of this line of argument is Gerschenkron's (Reference Gerschenkron1965) analysis of late Imperial Russia. He agrees that “agrarian reform” had been a “prerequisite to industrialization” in Western Europe, but claims that there was no analogous agrarian reform in Tsarist Russia. Thus, he says, to understand late Imperial Russian industrialization it is necessary to analyze the “substitutions” that “materialize in compensation either for the absence of an agrarian reform or for those disabilities, from the point of view of industrial development, with which an imperfect agrarian reform has been burdened” (ibid.: 207). In Gerschenkron's account, the agricultural sector remained socially premodern and economically stagnant, and there was no sufficiently developed domestic bourgeoisie to carry out industrial capital accumulation on its own. Thus, the state fiscal apparatus was employed to redirect what surplus was produced by the peasantry into supporting industrial investment, as both a source of financing and a captive buyer. The entire arrangement was like a hothouse for the cultivation of industrial capital, in the absence of social and economic conditions that were its prerequisites in earlier cases of capitalist development.Footnote ¹

A similarly influential line of argument about Japan's prewar economic development originated in debates among Japanese Marxists (see Hoston Reference Hoston1986). In the terminology prevalent among Japanese academics for several decades after the war, the Meiji Restoration established an “absolutist” state that adopted agrarian policies consolidating “semi-feudal” relations between landlords and subsistence-oriented peasant tenants (Takahashi Reference Takahashi and Tanaka2001 [1953]). In this analysis, the Japanese state “became the indispensable mediating conduit joining together capitalism and the landlord system” and was continuously called on to resolve “the contradiction incessantly generated by the forceful union of what were originally two different” modes of production (Nakamura Reference Nakamura1968: 238). In the phrasing of a different stream of Marxist thought, Japan represents “a near-classic case of ‘combined development’: an amalgamated social structure fusing different ‘stages’ of development—in this case, an intertwining of tributary and capitalist social relations” (Allinson and Anievas Reference Allinson and Anievas2010: 484). Even if by the end of the nineteenth century a substantial industrial capitalist sector had emerged, it was “a hothouse variety, growing under the shelter of state protection and subsidy” (Norman Reference Norman and Dower1975 [1940]: 218). It coexisted uneasily with an immense, poor, and socially backward agrarian sector.

The Fruits and Limits of Cliometrics

For a time, these accounts of Russian and Japanese development were influential and even dominant. Still, they never went unchallenged and they have since been largely supplanted by alternative accounts. The scholars, often quantitative economic historians, who have developed these alternative accounts contend that the earlier interpretations greatly exaggerated the disconnection between agriculture and industrial development. Quantitatively, the pictures of the beginnings of industrialization drawn by these scholars’ findings are remarkably similar for the two countries, at least on an aggregate level. In 1874, soon after the Meiji Restoration, Japan is estimated to have had a per capita income, in 1990 international dollars, of $860 (Saito Reference Saitō and Baten2016: 173). The earliest reliable estimate for Russia is for 1885, at the level of $865 (Bolt and van Zanden Reference Bolt and van Zanden2013, based on Gregory Reference Gregory1982). These were basically undeveloped agrarian economies that were productive enough to support some commerce and industry tied especially to the elite's concentration of surplus, but which had already fallen substantially behind the large Western economies.Footnote ²

The more striking result of these estimates is what they show about the two countries’ subsequent growth, as summarized in Table 1. By 1913, both Japan and Russia had reached per capita incomes of around $1,400, having grown at about 2 percent per year in the intervening decades. This was driven in part by rapid industrial growth, around 6 percent per year in Japan and 4.5 percent in Russia, or about 3 percent and 2 percent per worker, respectively (Gregory Reference Gregory1982: 134; Ohkawa and Shinohara Reference Ohkawa and Shinohara1979: Table A12). Yet, crucially, economic growth was not limited to catch-up development of modern industry. In addition to mining, metallurgy, and the railroads, industrialization in both cases built on existing trends of expanding light industry in rural areas (Francks Reference Francks2006; Gatrell Reference Gatrell1986: 154–55; Stanziani Reference Stanziani2010). Moreover, industrial growth of all kinds was supported by steady growth of agricultural output, at a rate of 1.6 percent per year in Japan and 2.8 percent per year in Russia, which, correcting for the much more rapid growth of the agricultural workforce in Russia, amounted, respectively, to 1.5 percent and 1.4 percent annual increases in agricultural output per worker (Gregory Reference Gregory1982: 134; Ohkawa and Shinohara Reference Ohkawa and Shinohara1979: Tables A-12 and A-18). Thus, by the eve of the First World War, even if neither country could yet be said to be industrialized, they were clearly industrializing, having reached a level comparable to that of the large continental European economies in the middle of the nineteenth century. Moreover, agriculture was overall not being steadily left behind at a stagnant premodern level, but instead was contributing dynamically to economic development by taking advantage of expanding opportunities for export and for selling food and raw materials to the growing domestic urban industrial sector.

Table 1. Economic Growth in Japan and Russia, Annual Real Rates, 1885–1913.

Sources: Ohkawa and Shinohara (Reference Ohkawa and Shinohara1979) for Japan, and Gregory (Reference Gregory1982) for Russia.

Notes: ¹ Gross national expenditure. ² Net national product. ³ Five-year averages.

The upshot of these arguments is the claim that the non-capitalist or premodern character of rural economic relations had previously been exaggerated. Gatrell (Reference Gatrell1986: 232) insists that indicators emphasized by earlier accounts, such as “[t]he persistence of small-scale units of production in agriculture and industry,” do not in fact represent evidence of “the absence of capitalist relations in late tsarist Russia.” In the even less equivocal words of Paul Gregory, research by himself and others has provided “convincing evidence that Russian agriculture was, by and large, operating according to market principles” (Reference Gregory1994: 83). In the case of Japan, revisionist economic historians have focused on arguing for a continuity of market-responsive economic activity, centered on rural areas, reaching back from the Meiji period to at least the eighteenth century (Francks Reference Francks2006; Hayami Reference Hayami, Hayami, Saito and Toby2004; Yamamura Reference Yamamura1974; Reference Yamamura and Craig1979). Small-scale farming combined with commercial and industrial by-employments remained the dominant rural economic pattern, not due to a failed or incomplete transition to capitalism, but instead because these patterns were themselves highly effective adaptations to the growth and development of the industrial economy (Francks Reference Francks2006; Sashinami Reference Sashinami2001).

The key point is that, according to these accounts, the pace of economic change was the result of economic actors’ responses to shifting market opportunities—to sell in different markets or to buy different products—which were determined in part by state policy and in part by conditions beyond anyone's control. Even in these accounts states still had a role to play, but rather than state intervention being needed to mediate between premodern agriculture and hot-house industry, they credit Japanese and Russian states with contributing to opening up the market opportunities that individuals, in a decentralized way, took advantage of in the process of economic development. Especially important here was the construction of institutional and physical infrastructure. In Russia, railroad construction was one of the most striking achievements of the government's modernization program, increasing total track length from 6,800 to 22,400 kilometers in a single decade from 1868 to 1878 (Gatrell Reference Gatrell1986: 150–53). Railways opened up market opportunities to regions that had previously been thinly settled or had no outlets for agricultural commodities they might have been able to produce (Allen Reference Allen2003: 46; Kopsidis, Bruisch, and Bromley Reference Kopsidis, Bruisch and Bromley2013). Government monetary and financial policies also succeeded in opening Russia to international capital markets (Gatrell Reference Gatrell1986: 207–28; Gregory Reference Gregory1994: 62–77). In Japan as well, government policy fostered the construction of telegraph lines and railroads and the emergence of modern shipping and financial industries (Kasuya Reference Kasuya, Ishii, Hara and Takeda2000; Koiwa Reference Koiwa, Ishii, Hara and Takeda2000). More generally, the Meiji state is recognized as having set up a framework that was able to mobilize existing resources, including the social capital of local notables and the human capital of traditional craftsmen and farmers (Hayami and Yamada Reference Hayami, Yamada and Smitka1998 [1975]; Morris-Suzuki Reference Morris-Suzuki1994; Nakamura Reference Nakamura2010).

An important dimension of this alternative interpretation of the role of the state is that it does not challenge the earlier accounts’ assessment that the two states’ reforms to agrarian property relations were very limited. However, the significance placed on this fact is quite different in the two interpretations. Scholars of the older tradition of state-mediated development concluded that the reforms adopted were insufficiently radical to set off capitalist development in the countryside, while the economic historians contend that more revolutionary reforms were unnecessary. For Japan, many scholars have come to argue that whatever the feudal political system's de jure rules, “peasants’ property rights in land approximated those of a fee-simple title by the end of the Tokugawa period [and] were readily converted to the modern private-property system” following the Meiji Restoration (Hayami and Ruttan Reference Hayami, Ruttan, Otsuka and Runge2011: 227). Industrialization occurred after the opening of the ports not because of a dirigiste program led by the state, but rather because Japanese entrepreneurs took advantage of opportunities for trade and technology transfer, the absence of which had prevented Japan from industrializing in the Tokugawa era (Yamamura Reference Yamamura and Craig1979: 320). For Russia, the claim made is that policies that seemed counterproductive were in fact too weak to have strong negative effects. Restrictions “that were supposed to freeze rural labor” were not enforced seriously enough to harmfully prevent “the rapid growth of the industrial labor force and the regional redistribution of labor” (Gregory Reference Gregory1994: 83). As Gatrell puts it, the terms of the emancipation of the serfs were less important than developments, such as rail construction, that helped “to emancipate Russia from the limitations imposed by her geography” (Reference Gatrell, Eklof, Bushnell and Zakharova1994: 99).

In many ways, the evidence of these two cases is consistent with the story told by these economic historians. Economic growth in Japan and Russia in this period did involve the widespread, decentralized response of economic actors to newly opened market opportunities. Although very few of these accounts are explicitly comparative, that there were two cases, in a shared global market, which seemed to be growing in parallel only adds to the plausibility of this perspective. That said, even if these accounts present a plausible description of the processes of economic growth, they fall short in offering an explanation of them. They contend that the new market opportunities of the late nineteenth century—whether the result of state policy or otherwise—represent a sufficient explanation of the two countries’ trajectories of economic development. Crucially—in the context of their rejection of the previously dominant accounts—they see the consequences of state reforms on rural property relations as insignificant, one way or another. However, a closer look at the evidence of the specific patterns of economic growth in the two cases shows that market opportunities, on their own, cannot explain the occurrence of economic development. It is possible to identify two separate sets of serious anomalies. First, in Japan, the trajectory of growth changed qualitatively not only in sectors affected by the opening of international markets. Second, in Russia, there were large areas of the country where producers did not take advantage of the expanded market opportunities of the late nineteenth century by increasing their productivity. The rest of this section will document these anomalies.

In Japanese agriculture it was sericulture—the raising of silkworms to harvest their cocoons—that most benefited from Japan's tighter integration into world markets. Cocoon production was indeed one of the fastest growing segments of agriculture, quadrupling in output from 1880 to 1915 (Hayami and Yamada Reference Hayami and Yamada1991: Table A-1). Yet the value of silk cocoons was never more than a fraction of that of field crops, of which rice was the most important.Footnote ³ Rice was neither an export productFootnote ⁴ nor did it benefit from any imported technology. Early hopes on the part of government officials that the adoption of crops and cultivation technologies from the West would revolutionize Japanese agriculture were quickly deflated by their incompatibility with the deeply entrenched structure of small-scale peasant farming (Ogura Reference Ogura1963: 150–53). Nonetheless, despite there being, if anything, a diversion of labor from rice cultivation into other activities, rice output grew by 1.2 percent annually from 1880 to 1910 (Hayami and Yamada Reference Hayami and Yamada1991: Table A-1). Likewise, many consumer industries flourished that either had no foreign technology to adopt, such as sake and soy sauce brewing, or persisted on the basis of putting-out systems rather than transitioning to factory production, such as textile weaving (Tanimoto Reference Tanimoto, Ishii, Hara and Takeda2000). In these “traditional” sectors, productivity steadily increased through innovation that “was quite widely spread through many companies and craft workshops in many parts of the country,” often not based on “imported technology at all, but incremental improvement of techniques developed in the Tokugawa period” (Morris-Suzuki Reference Morris-Suzuki1994: 85–86). The question, then, is why the development of these technologies for production, oriented predominately to the domestic consumer market, only accelerated following the Meiji Restoration.

The anomaly of the Russian experience was different: why did some agricultural regions that should have been favorably positioned to import improved technology and export more grain fail to do so? Grain, after all, was the Russian Empire's major export, and the apparent dynamism seen in aggregate statistics is a key piece of evidence indicating that Gerschenkron's theory of substitutes missed something important. That the Russian economy was characterized by immense regional variations is well known. As Gatrell (Reference Gatrell1986: 139) candidly notes: “In the centre, progress was peripheral, but on the periphery there was progress.” Yet, neither he nor other economic historians have fully captured the magnitude of this regional divergence or its significance. To put it most starkly, all of the increased agricultural productivity that supported the beginnings of industrialization—itself highly concentrated geographically—occurred on the empire's western and southern frontiers. In the old agrarian core of central European Russia and the expanding frontier of Siberia—together, home to about a quarter of the country's agrarian population—increasing population swamped increases in output.

Table 2 provides select, quantitative evidence on regional grain cultivation in the final decades of Imperial Russia from the 1850s, before emancipation, to 1913. The old agrarian core accounted for over a third of Russia's grain output in the 1850s, but farmers of this region barely took part in the agricultural growth of subsequent decades. Most growth through the 1890s came from the expanding frontiers in New Russia, the Urals, and Siberia. Productivity estimates present an even more striking divergence. A few highly dynamic regions—Congress Poland, the Caucasus, and New Russia—showed by 1913 agricultural labor productivity that was one-third to over one-half higher than the Russian average. In these regions, producers widely took advantage of expanding markets opened by improved transportation and the opportunity to import more productive technology (Friesen Reference Friesen1994 on New Russia; Kieniewicz Reference Kieniewicz1969: 222 on Congress Poland). Elsewhere, despite their shared economic environment, productivity stagnated. Grain production per rural resident in the Central Black Earth and Volga regions peaked in the 1870s and subsequently fell by over 17 percent through the early 1910s (Wheatcroft Reference Wheatcroft, Kingston-Mann and Mixter1991: 138). Even on the eastern frontier, although extensive growth continued to expand overall output, especially in Siberia, this growth in fact lagged behind the growth of the agricultural population, and so productivity fell between 1897 and 1913 to a level comparable to that of the old agrarian core.

Table 2. Regional Variations in Late Imperial Russian Agricultural Development.

The main sources of data for these table are Wheatcroft (Reference Wheatcroft1980), the 1897 census digitized in the RISTAT database (Kessler and Markevich Reference Kessler and Markevich2014), and the 1913 Russian Statistical Yearbook (Tsentral'nago Statisticheskago Komiteta MVD 1914). For a full description of sources, definitions of regions, methods of calculation, and tables of additional measures and other regions, see the Online Appendix (https://osf.io/x4npt/).

Economic development is usually uneven within a country, but the contrast between dynamic and stagnant regions in late Imperial Russia is astonishing. Figure 1 compares the degree of divergence between leading and lagging regions in Japan and Russia. In Japan, too, some regions were more productive than others, but the differences were relatively small: levels of labor productivity ranged about 20 percent below and above the national average, and no region fell below 0.8 percent annual growth. The absence of productivity growth in the Central Black Earth and Middle Volga regions requires an explanation, as does the question of why Siberia experienced only extensive growth, in contrast to New Russia and the Caucasus where extension of the frontier went hand in hand with higher output per capita. Existing accounts do not explain why, in both Japan and Russia, individuals in different times and places pursued increases in productivity and income to such different degrees despite apparently similar opportunities to do so. What is missing from these accounts is an analysis of how different configurations of rural property relations conditioned different responses to market opportunities and thus different outcomes of economic growth.

Figure 1. Leading versus Lagging Regions in Russia and Japan.

The State, Market Society, and Economic Power

Responding to the errors of the older accounts, the revisionist economic historians adopted a theoretical perspective that, in effect, naturalized the ubiquitous market of capitalist society. Macro-level economic outcomes, according to this perspective, were the aggregation of individual economic behaviors, which each followed the path toward maximized income. From this view, the key question for explaining the economic history of a particular case becomes what opportunities were available for individuals to increase their income through more intensive or productive economic activity, especially through market exchange, weighed against the effective barriers that were raised by powerful actors such as governments. This perspective is open to the accusation that it falls into what Polanyi calls the “economistic fallacy,” which “consist[s] in an artificial identification of the economy with its market form” (Reference Polanyi, Polanyi, Arensberg and Pearson1957: 270). The blind spot created by this fallacy is that it overlooks that possibility that economic actors can limit the intensity of their involvement in markets not because they are being prevented by force or for the lack of lucrative opportunities to do so, but instead because doing so conflicts with alternative avenues by which they secure their livelihoods.

Polanyi's key historical insight is that prior to modern capitalism market exchange was almost never the central organizing principle of social life. The production and circulation of economic goods were not regulated by a distinctively economic logic but instead were “submerged in … social relationships” (Reference Polanyi2001: 48). These relationships, including households, communities, and polities, mobilized people and distributed resources in ways that did not necessarily respond to the price signals of markets. They thus conflicted with the ideal of a “market society” that has been promulgated in liberal economic thought and pursued by political and economic elites. In this ideal, all major economic goods and resources, including crucially land, labor, and money, should behave as commodities, and this “implies that all production is for sale on the market and that all incomes derive from such sales” (ibid.: 72). Thus, the drive to fulfill this ideal entailed a concerted political project to dismantle or subordinate the non-market relations that had previously organized economic life in order to commodify, even if only “fictitiously,” every input and product of the economy. I will show below that a crucial axis of variation between Japan and Russia was the extent to which such a political project was undertaken and successfully pushed through. The main direction in which Polanyi takes these points is to argue that the drive of commodification was necessarily contradictory, but for the purposes of this paper's analysis I will go beyond his argument in two ways.

First, while Polanyi focuses on the pursuit of market society as a utopian vision that state policy was designed to realize, this was not the case in either Russia or Japan. The problem that most directly confronted political elites in these two countries was not economic but instead geopolitical. Both states were exposed to the military power of the leading European states, Russia directly in the Crimean War, and Japan primarily by the example of China's fate in the Opium Wars. The level of military capacity achieved by the European powers by the mid-nineteenth century was the result of a long process of interlinked war-making and state-making (Tilly Reference Tilly, Evans, Rueschemeyer and Skocpol1985; Reference Tilly1992). As Tilly shows, two key results of this process were to demonstrate the overwhelming advantages of what he calls “direct rule” and “capitalized coercion” (Tilly Reference Tilly1992: 103–6). On the one hand, states, in order to secure the manpower and material resources needed to be militarily competitive, had to secure much closer control of their populations than had been previously typical.Footnote ⁵ On the other hand, it was exceedingly difficult to extract the necessary resources without the expansion of commercial flows and the urban concentration of wealth. These were the ends pursued by political elites in their reforms to social, political, and economic institutions in Japan and Russia. Policies that commodied land and labor, or did not, were alternative means to these common ends.Footnote ⁶

Second, Polanyi emphasizes the destructive consequences of a market economy for existing social relations, but he pays relatively little attention to the contours of the social relations and the new patterns of social power that emerge as a result of the transformed role of markets. The latter have been most consistently and thoroughly pursued by Marxist scholars. In the words of Ellen Meiksins Wood: “The specificity of capitalism is that the relations of producers to the means of production, and of appropriators to the means of appropriation, as well as their relation to each other, are mediated, indeed constituted, by the market” (Reference Wood2002: 85). The significance of this is that even if, as Polanyi emphasizes, the market economy is the product of a political project, one of its distinguishing features is that it formally separates the economic power of property owners within markets from the overt exercise of coercion by political power-holders over subjects (Wood Reference Wood1981). Both the feudal lord and capitalist landlord are landowners, and both exert immense power over the farmhands or tenants who work their land, but the ways in which they own their land and exert this power are immensely different: in a market economy both property owners and the working majority become dependent on market competitiveness (Brenner Reference Brenner and Wickham2007). New market opportunities to produce more lucrative goods, more efficiently, or for a wider market, are not merely something agents can pursue out of a desire for more income; in a market economy they must pursue them, since it is forced upon them by competition. This compulsion, working separately on each participant in each market throughout the economy, manifests in steady economic growth, and it is precisely what was absent in both Tokugawa-era Japan and the stagnant regions of late Imperial Russia.

To summarize, the accounts just surveyed lacked explanations for conspicuous facts about the beginning of capitalist economic development in Japan and Russia in the late nineteenth century. They pointed to the market opportunities opened by a combination of the global environment and state policy, which producers took advantage of, thereby driving impressive growth rates in rural economies of both countries. Yet economic actors did not always take advantage of such opportunities: specifically, the developments to rice cultivation that occurred in Meiji-era Japan had been available earlier but were not pursued, and in large regions of Russia, especially the old agrarian core and Siberia, cultivators did not take advantage of the opportunities to produce and sell grain that were avidly exploited by their counterparts elsewhere in the empire. These anomalies can be explained, I contend, by taking into account Polanyi's key insight that, historically, other social institutions organized economic life and pushed market exchange into a secondary role. Markets were transformed into a dominant economic institution only where states, as part of projects of political, fiscal, and military consolidation, dismantled those other institutions and thereby established the basis for property owners’ economic power within market relations with workers and cultivators. The next section applies this theoretical lens in a comparative analysis of Japan and Russia, and within Russia between relatively dynamic and stagnant regions.

Meiji Japan: The Land Tax Reform and the Land Rental Market

In 1871, three years after the Meiji Restoration overthrew the Tokugawa Shogunate, Japan's new imperial regime abolished the rule of the feudal lords and took over direct administration of the entire territory of Japan. The problem faced by the leaders of the new state was to consolidate their hold on the material resources needed to undertake the modernization program that they quickly came to see as necessary to reassert Japan's position in the geopolitical order dominated by the Western industrial powers (Shibahara Reference Shibahara1977: 148–51). One problem was the tax system they had inherited from the old regime, which varied immensely in rates among the multitude of feudal jurisdictions and failed to generate a stable cash income. It undoubtedly needed to be replaced, and the Land Tax Reform that the government devised and implemented beginning in 1873 was meant not only to stabilize revenue, but more fundamentally to “establish the foundation of a tax system that was suitable for a modern state standing over a capitalist society” (Sasaki Reference Sasaki1988: 79). It was based on the principle that every plot of land would be assigned an individual owner and assessed a cash value, which would be the basis for calculating the tax burden (Okuda Reference Okuda2007). This section traces how this fiscal policy entailed deep consequences for the definition and enforcement of land ownership, which set in motion a decisive shift in agrarian property relations that made rural areas into an important engine of capitalist development. To make this argument, we must first look in more detail at the structure of rural society in the late Tokugawa period.

The ideology of feudal rule had been premised on a smallholding peasantry that was supposed to be strictly focused on cultivating for its own subsistence and the lords’ taxes. However, from at least the mid-eighteenth century onward, the reality in the countryside was quite different. Between taxes and the scarcity of land, peasants frequently turned to commerce to supplement their incomes, and the precariousness of premodern agriculture often pushed them into mortgaging their land in bad years, often with thin hopes of ever redeeming it (Kwon Reference Kwon2002; Smith Reference Smith1959). The flipside of this process was the emergence of a stratum of small to medium landlords who were legally commoners under the feudal status order (Furushima Reference Furushima1963; Ōta Reference Ōta1981). These landlords rented out small strips of land to poorer peasants, but also often cultivated some land with family and wage labor, engaged in petty commerce and moneylending, and served in various local offices that acted as intermediaries between the feudal rulers and the cultivating peasantry. They exercised significant authority in their localities, but their positions could also be quite contradictory. As officials, the feudal elite expected them to ensure order and the payment of taxes, but in practice it offered them little legal or coercive support (Befu Reference Befu, Silberman and Harootunian1966). Villages were left largely to govern themselves, and even powerful landlord families had to cultivate the support of the wider community (Smith Reference Smith1959: 59–63). Other peasants expected wealthy families to represent their interests to the lords and to act as patrons to their poorer neighbors (Befu Reference Befu, Hall and Jansen1968). At the same time, landlords profited off poorer peasants as tenants, borrowers, and suppliers of cash crops and handicrafts (Kwon Reference Kwon2002; Pratt Reference Pratt1999). When these activities called up opposition from the regular peasantry—increasingly often over time (White Reference White1995)—the lords provided at best lukewarm backing. When feudal officials did intervene in disputes, landlords could not rely on them to categorically defend landlords’ property rights rather than seeking to minimize conflict and maintain the status quo (Ooms Reference Ooms1996).

As a consequence of their ambiguous position, landlords’ economic power was notably constrained by the continued strength of the peasant community, which prioritized the preservation of peasant subsistence (Furushima Reference Furushima1963; Saitō Reference Saitō2009: 172–73; Waswo Reference Waswo1977: 30–31). Due to communal responsibility for taxes, village residents as a whole had a strong interest in maintaining the viability, and thus the tax-paying capacity, of member households, and communal institutions developed to support this goal (Arimoto Reference Arimoto2006; Sakane Reference Sakane2011: 134–35). The regulation of the peasant community was especially strong when landlords were not themselves village residents, being from the next hamlet or a nearby commercial town (Watanabe Reference Watanabe2007: ch. 1). Except when wealthy commoners financed large-scale land reclamation projects, few landlords acquired very large or geographically dispersed holdings.Footnote ⁷ Where absentee landlordism did develop, landlords’ relations with the cultivating peasantry were often mediated by “middle tenants” who took a share of the rent and could represent tenants’ demands for rent reductions (Nakayama Reference Nakayama2001). In sum, under the old regime, landlords “resigned themselves to what was closer to a right to rent payments than to ownership per se,” and these rights were always mediated through the authority of the peasant community (Niwa Reference Niwa1989: 227).

Thus, despite official prohibitions, land and labor markets had emerged in most rural areas in Tokugawa-era Japan, but these market relations were subordinated to an institutional nexus linking the subsistence of peasant households, the regulating authority of administratively autonomous village communities, and the taxation of the feudal elite. The individualization of tax responsibility by the Meiji state in the 1870s broke up this three-way link. Communal regulation of land ownership was declared legally powerless, particularly that of landlord-tenant relations, which often amounted to quasi-property rights of tenants to their plots (ibid.: 221). This was confirmed in 1877 when the Supreme Court ruled that tenants had no legal recourse against the rental terms offered by landlords (Vanoverbeke Reference Vanoverbeke2004: 35). Instead, owners’ rights to their land were absolute, and as was written on the back of the deeds issued to owners as part of the Land Tax Reform: “The Japanese citizen who holds this deed has the right to freely employ or own this land, as well as the right to sell, transfer, mortgage, or pawn it” (in Ono Reference Ono1948: 220). The new land rights policy, at least among influential officials in the Finance Ministry, was linked to a hope of stoking the entrepreneurial activity of wealthy landowners (Niwa Reference Niwa1989: 221–24). Reporting on the results of the Land Tax Reform in 1882, Finance Minister Matsukata Masayoshi insisted, “Without security of land ownership, we cannot expect to encourage cultivation or promote expanding production” (quoted in Kurauchi Reference Kurauchi1990: 15).

The Meiji state's reforms to property rights in land and construction of legal and tax systems based on them decisively shifted the balance of power between landlords and their tenants. Rent collection provides a particularly clear index of this shift. Nominal rents had been extremely high in the Tokugawa era. It was not unusual for contractual rates, including taxes, to amount to over two-thirds of a normal harvest, as reported by case studies in Echigo (Nakayama Reference Nakayama2001: 51) and southern Okayama (Ōta Reference Ōta1981: 238–39). However, landlords often collected only a fraction of the contractual rent. Records for one landlord in nineteenth-century Echigo show payment rates fluctuated between 70 and 90 percent of the assessed rent even in normal harvest years (Nakayama Reference Nakayama2001: 84–85). Likewise, receipts of a landlord in south Okayama in various years from the late 1850s to the early 1870s show his tenants paid only ½ and ¾ of the assessed rent, even including repayments of existing arrears (Ōta Reference Ōta1981: 239, 311–16). In contrast, by the mid-1880s, landlords were successfully tightening the terms of leases, and payment rates, instead of fluctuating widely from year to year, stabilized at 90 percent or more of the contractual rent (Arimoto Reference Arimoto1970; Ōta Reference Ōta1981: 314–82). Based on landlords’ later reports to government surveys, rent levels across the country converged toward 50–60 percent of average yields in the late 1880s and increased along with or slightly faster than productivity into the 1910s (Kurauchi Reference Kurauchi1996). The 1914 Survey on Tenancy Practices reported that landlords in the five years from 1908 to 1912 collected, on average, 54 percent of the harvest on middle-quality paddy fields, amounting to 92 percent of the contractual rent (Nōchi Seido Shiryō Shūsei Hensan Iinkai 1970: 21–23). Landlords’ ability to claim income from tenanted land, and to increase that income along with productivity, had become stable and predictable.

This shift in agrarian property relations also affected the distribution of land ownership. In some cases, the Land Tax Reform brought this about by fiat, cutting through ambiguous cases of mortgaged land by simply recognizing the lender as owner and landlord (Ono Reference Ono1948: 192–94; Waswo Reference Waswo1977: 22). Landlord accumulation of land and peasant loss of same accelerated in the 1880s and continued into the early twentieth century. Table 3 reports the extent of this redistribution of land that is captured in national statistics. Between 1883/1884 and 1912, in those prefectures included in the Japanese government's first survey of tenancy conditions, the area of paddy cultivated by its owners decreased by 10 percent, while the tenanted area increased by 39 percent. The number of peasant households cultivating only land that they owned decreased by 20 percent, and the number of agricultural households relying entirely on tenancy increased by the same proportion. Overall, in 1912, 51 percent of paddy and 40 percent of dry fields were cultivated by tenants, and 73 percent of cultivators farmed some rented land. Like their Tokugawa-era predecessors, Meiji-era landlords accumulated land mainly by financing reclamation and by lending to and purchasing from peasants unable to pay their short-term expenses, including taxes. But now the safety-valve of communal regulation of the latter was gone, and landlords were free to antagonize the rest of the peasantry and to expand their holdings beyond their immediate localities in a way they had not been able to before (ibid.: 1977).

Table 3. The Increase in Tenancy in Meji-era Japan.

1 chō = 2.45 acres. Hokkaido and Okinawa are excluded. Sources: The 1883–1884 survey, missing nine of forty-five prefectures for land area and an additional five for household counts, is from the 1886 Statistical Yearbook of Imperial Japan (Naikaku Tōkei-kyoku 1886: 86–89). The 1912 data is from the 1914 annual statistical report of the Ministry of Agriculture and Commerce (Nōshōmu Daijin Kanbō Tōkeika 1914: 1–2).

The Meiji state's abolition of the Tokugawa-era status order had eliminated wealthy rural commoners’ roles as intermediaries. However, the new property rights established by the Meiji state, enforced by a national legal and administrative system, offered landlords strategies for their economic advancement that had been impossible before. That is to say, wealthy rural residents could mobilize their economic power through credit, land, and labor markets, which were no longer submerged in the village community's protection of peasant households buttressed by the feudal rulers’ communal strategy of fiscal extraction. For the peasantry, the result was that their claim to subsistence from the land became increasingly precarious, due to the cash tax and less flexible demands for rent if they became tenants. Peasants thus were compelled to seek out ways to improve their agricultural productivity and secure additional non-agricultural income. In 1886, 37 percent of tenant households and 30 percent of owner-cultivator households were reported as being dual-employed, and this probably underestimated the temporary non-agricultural employment of young adults (Naikaku Tōkei-kyoku 1888: 77).

The landlords themselves played an important role in providing the peasantry with such options. Landlords tended to farm some portion of their holdings directly, and more so than smallholders, they had the resources to experiment through trial-and-error in the selection of crops, techniques, and seeds, driving the impressive improvements in productivity seen during the era (Waswo Reference Waswo1977: 38–42). This process occurred not just in agriculture but also in sericulture (Smethurst Reference Smethurst1986: 187–231). Moreover, landlords were key investors and brokers for industries carried out in both peasants’ households and rural factories (Ōta Reference Ōta1981). As heavier, more urbanized industries began to solidify after the turn of the century, landlord capital that was accumulated via rent and profits from rural industry again became an important source of investment (Francks Reference Francks2006; Nakamura Reference Nakamura1968).

These developments were not only important for rural areas but also opened up new patterns of economic change that reverberated throughout Japanese society. Tenants and smallholders provided food for a growing non-agricultural population, and indeed a greater share of the food they produced went to the market over time (Shinohara Reference Shinohara1967). Moreover, small-scale rural industry produced both export commodities (especially raw silk) and domestic consumer goods, for instance inexpensive cotton textiles (Saitō and Tanimoto Reference Saitō, Tanimoto, Hayami, Saitō and Toby2004). In the other direction, as peasant households earned an increasing share of money income by marketing agricultural products and by having one or more members working outside agriculture, they also constituted a growing consumer market for domestic industries including processed food and textiles (Francks Reference Francks2002). Home production was gradually replaced by market production. For instance, the total quantity of factory-produced soy sauce is estimated to have first surpassed that brewed in households in 1905 (Shinohara Reference Shinohara1967: 204–5). In all, peasants were increasingly anchored into a flourishing Japanese capitalism, which had been set in motion by the state's reconfiguration of rural property relations in pursuit of a sturdier fiscal base.

Russia: An Imperial Patchwork of Agrarian Property Relations

The so-called Great Reforms enacted in Russia in the 1860s eliminated serfdom and thus required the establishment of new institutions linking the central state to the rural population. Unlike the Meiji state in Japan, the Tsarist regime did not adopt a policy uniformly establishing and enforcing individual property rights in land. Different patterns of social property relations were established in different regions by the emancipation of the serfs—or emerged later in places with no legacy of serfdom—based on variations of policy and of existing institutions. After all, the Russian Empire was a patchwork of regions brought under Tsarist rule at different times and with substantially different economic institutions. Above, I described the stark differences in development between the old agrarian core, Congress Poland in the West, the southern frontier of New Russia and the Caucasus, and the eastern frontier of the Urals and Siberia. What follows will explain these variations through an analysis of the various property relations that emerged in the different regions.

I begin with and consider most closely the old agrarian core, since it provides the most stark contrast with the policies adopted by the Meiji state and the resulting trajectory of Japanese agriculture. From the eighteenth century until the end of serfdom in the Central Black Earth and Middle VolgaFootnote ⁸ regions, the fertile soil and the opportunity to sell grain to the northern cities or to Europe led estate owners to intensify the cultivation of their demesnes with serf labor (Emmons Reference Emmons1968: 20–25; Melton Reference Melton1987: 92–94).Footnote ⁹ Nonetheless, lords commonly delegated a great deal of village-level administration and even the management of the demesne to peasant communities (Emmons Reference Emmons1968: 20; Manning Reference Manning1982: 4). For this reason, although the region produced large quantities of grain (see Table 2), its agriculture remained technically rudimentary and was linked to markets primarily by the serf owners’ capacity to coercively redistribute the product of serfs’ labor.

The emancipation of the serfs, beginning in 1861, preserved by design the broad outlines of agrarian property relations as they had existed under serfdom. The emancipation “was intended to keep the peasants peasants—intended, that is, to avoid the emergence of a proletariat and the revolutionary upheavals associated with that process” (Zakharova Reference Zakharova, Eklof, Bushnell and Zakharova1994: 30). The peasants were given land with their freedom in order to avoid the sort of unrest that followed the landless emancipation of the Baltic provinces’ in 1816–1819 (ibid.: 25–26). The Tsar also insisted that the property rights of the nobility be respected, at least formally. In practice, this meant that the emancipation confirmed nobles’ ownership of their estates, except for the subsistence “allotments” that were transferred to the liberated peasants, who had to compensate the nobles for their lost property in soil and “souls.” Subsequent state policy sought to both preserve the noble landlords created by the emancipation and prevent the widespread peasant proletarianization that the emancipation had been designed to avoid. In practice, the latter aim was more successful,Footnote ¹⁰ and by 1905 less than 35 percent of the land area in the Central Black Earth and Middle Volga Provinces was owned privately.Footnote ¹¹

The other major goal of state officials was to consolidate centralized administrative control over rural areas. This goal they shared with their counterparts in Japan, but different pre-reform institutions led reformers to adopt different strategies in the two cases. In Japan, a key element of the Meiji state's project to consolidate its administrative and fiscal capacity was the abolition of communal governance and the individualization of landownership and the associated tax burden. In central Russia, the outcome was almost the exact reverse. In place of the immediate rule of nobles over the peasantry, the Tsarist regime legally recognized peasant communal institutions as the owners of allotment land, rendering them collectively liable for their members’ taxes and redemption payments and granting them considerable power over their members (Field Reference Field, Eklof, Bushnell and Zakharova1994). Peasants’ rights to sell their allotment land, to divide it for heirs, or even to leave their villages were legally curtailed (Frierson Reference Frierson and Bartlett1990; Gatrell Reference Gatrell1986: 107–8). Thus, while the Japanese state proclaimed the formal freedom of peasants as individuals and landowners and empowered landlords against the formerly quasi-autonomous peasant community, in Russia the emancipation of the serfs was followed by the legal recognition and consolidation of communal peasant institutions.

In the old agrarian core of the empire in the Central Black Earth and Middle Volga regions, the state's reinforcement of the regulatory power of the peasant commune served as a bulwark against competitive, market-oriented agricultural development. This conclusion is not contradicted by the ample evidence of peasant labor mobility and of limited changes in cultivation practices such as cropping patterns (Bideleux Reference Bideleux and Bartlett1990). The tendency of communally organized villages to allow migration or to raise land productivity as a response to increased population was compatible with the peasant community's role in ensuring its members’ non-market access to subsistence. The crucial contrast is not between “tradition” and “dynamism,” but rather between the competitive productivity increases required by market dependence and the prioritization of household security enabled by communal regulation. As Moon points out, “Like many peasants all over the world who lived close to the margins of subsistence, [Russian peasants’] culture emphasized minimizing risk, rather than maximizing profits” (Reference Moon2000: 687). The periodic redistribution of land for which the Russian peasant commune is so notorious was itself a mechanism to maintain the subsistence of households (Atkinson Reference Atkinson and Bartlett1990; Lewin Reference Lewin and Bartlett1990). The value Russian peasants placed on the commune's role in defending their economic security can be seen in both the instances of resistance to the government's program to abolish it after 1905 and the survival of many aspects of communal regulation even in villages where communal ownership was legally terminated (Pallot Reference Pallot1999).

The upshot was that Russian peasants in the old agrarian core were not exposed to a competitive rental market as were an increasing share of Japanese peasants from the 1880s onward. They thus were not similarly dependent on maintaining a competitive rate of productivity growth at the risk of losing their land. In a typical premodern pattern, peasants responded to a decrease in arable land per person by cultivating more intensively and by migrating either temporarily or permanently, but labor productivity in agriculture at home remained low and stagnant. What then of the regions that showed substantially higher and increasing productivity in agriculture? For different reasons, peasants in western and southern Russia were exposed to dependence on markets in labor and land without the protection of strong communal institutions.

In the provinces of Congress Poland—so-called because the Tsar received them as a result of the Congress of Vienna in 1915—full emancipation came only in 1863, but the process had begun decades earlier under the short-lived Duchy of Warsaw (1807–1815). The Napoleonic government of the Duchy had in late 1807 issued a decree guaranteeing peasants’ right of movement but also confirming landlords’ full ownership of serfs’ allotment land (Kieniewicz Reference Kieniewicz1969: 248–49). Thus, the serfs were no longer bound to the land, but landlords used the threat of eviction as a powerful bargaining chip against their tenants. From the 1820s to the 1840s, landowning nobles increasingly negotiated a shift from labor service to money rent among larger tenants and reduced the number of smallholders and cottagers on their estates, hiring farmhands instead (ibid.: 97). Although the Russian government began in the late 1840s to try to constrain landlords and favored the peasantry in the final emancipation of 1863, by then landlords had already had several decades to cut back the allotments of smallholders, who were never granted legal protection before the emancipation. According to official statistics, in 1859 around 40 percent of the agrarian population was effectively landless (ibid.: 145).

In Poland, peasants’ non-market claims to land were attenuated in the decades between 1807 and 1863, and clear signs of capitalist development in the region appear in later statistics. By the turn of the twentieth century, agricultural labor productivity was substantially outstripping the country as a whole (see Table 2), but this dynamism spread beyond agriculture as well. Already in 1897, the urbanization rate was around 25 percent, with an additional 25 percent of rural residents outside the agricultural workforce, comparable to the provinces around Moscow and St. Petersburg and much higher than the national averages (Kessler and Markevich Reference Kessler and Markevich2014). This symbiosis between agricultural dynamism and growth in non-agricultural production and urbanization represents the closest parallel within Russia to Japan's pattern of development. Their shared basis was a thorough privatization of land that forced peasants into dependence on markets, although with a greater emphasis on agricultural wage labor in Poland as opposed to credit and land rental markets in Japan. A similar pattern, set in motion by the landless emancipation of the serfs in 1816–1819, appears in the Baltic provinces, where in 1897 agricultural labor productivity was 57 percent above the national average and subsequently the region began to shift away from agriculture.Footnote ¹²

The most dynamic agricultural region of the empire was undoubtedly the southern steppes. In New Russia and the Caucasus, the institutional legacy of serfdom was faint because the expansion of grain cultivation, following initial Russian settlement or a shift from animal husbandry, only began in earnest in the 1880s (Mixter Reference Mixter1992: 118–19). A relatively large proportion of the land—in 1905, around 40 percent compared to 27 percent in the Central Black Earth region—was owned in large estates of over 200 desiatiny, or around 220 hectares.Footnote ¹³ Moreover, these estates, rather than being rented out to peasants, were typically farmed by seasonal migrant wage workers (ibid.: 121). These export-oriented grain plantations were increasingly capital-intensive, and beginning in the 1890s planters even began to import advanced machinery that enabled them to economize on labor, so that the growing area of cultivation could be worked by the same, or perhaps even a smaller migrant labor force (ibid.: 128, 53). The region's agricultural sector provided demand for not just imported machinery, but also the (relatively simple) products of a growing local implements industry (Friesen Reference Friesen1994: 416; Mixter Reference Mixter1992: 207–8).

The southern steppes were not the only region of new Russian settlement in the final decades of Tsarist rule, but the other major frontier of the empire, to the east in Siberia, followed a very different trajectory. Whereas peasants from central Russia went south as seasonal laborers, peasants migrated eastward as families seeking to set up homesteads (Treadgold Reference Treadgold1957). Land was essentially free, either claimed “by squatter's rights or with a government allotment of 15 desiatiny [40.5 acres] per male,” and although peasants legally did not have title to their land, there were de facto rental and sale markets (Goriushkin Reference Goriushkin, Grantham and Leonard1989: 448). For most of the last two decades of its rule, the Tsarist government provided logistical and financial support to migrants, including start-up loans on easy terms and temporary forgiveness of central taxes (Treadgold Reference Treadgold1957: 203, 16). In practice, taxes were quite low: 10 rubles per households against cash income usually over 100 rubles according to a 1911 survey (ibid.: 217–18).

To use Polanyi's terminology, late imperial Siberia was dominated by householding, the organization of family labor and resources oriented primarily to directly providing for the family's material needs (Polanyi Reference Polanyi2001: 55–56). Communal institutions were generally reported to be on the whole much weaker than in central Russia, although equalizing repartition emerged in some of the most densely settled areas of western Siberia (Channon Reference Channon and Bartlett1990). Peasants could and did sell their surplus grain, dairy, and handicrafts, and in aggregate Siberia exported a large quantity of especially the first two (Treadgold Reference Treadgold1957: 228). Yet, secure in their possession of land and livestock, peasant households were not dependent on markets for their livelihoods, and so they could focus their production on their own subsistence, selling only the surplus that was conveniently left over (ibid.: 172). After all, with taxes so low, cash expenses could be kept at a minimum; English traveler Philips Price noted that Siberian peasant bought very little: “tea, tobacco, sugar, and some Moscow-manufactured cotton prints from which women's clothes were made” (quoted in ibid.: 221). The economic results of these property relations stand in stark contrast with the southern steppes. Siberian peasants were relatively prosperous by Russian standards, but where official statistics report high and rising per capita grain output in the south, Siberian productivity stagnated at about the same absolute level as the center (see Table 2).

Summary

Table 4 summarizes the connection between property relations and economic outcomes in the case studies of this section. Each row—Japan before and after the Meiji Restoration and five separate regions in Russia—identifies the way in which the direct cultivators secured access to the land, the corresponding degree to which landowners were able to exercise economic power through markets, and the results for agricultural productivity and industrial growth. Agricultural productivity was stagnant and industrialization limited at best in the three cases where peasants had direct access to land that circumvented landowners’ economic power, whether because of communal control as in Tokugawa-era Japan or the old agrarian core in post-emancipation Russia, or because of the free availability of land, as in Siberia. Agricultural producers in these times and places were free to prioritize their security within non-market social institutions. In contrast, when and where peasant cultivators were made dependent on markets dominated by landowners—whether to rent land or hire themselves out as agricultural laborers—economic growth in both agriculture and industry was much higher. Forced to compete on markets, peasants could not afford to forego the kinds of opportunities for greater productivity and more lucrative production that drove economic growth.

Table 4. Property Relations and Agrarian Trajectories.