‘Do property rights matter in Russia?
The Stolypin titling reform and agricultural productivity’1
Paul Castañeda Dower and Andrei Markevich2
Abstract
We study the effect of improvements in peasants’ property rights launched by the Stolypin
reform on agricultural productivity in early 20th century Russia. We find that the enclosure
component of the reform caused an increase in land productivity because of better usage rights,
not because of economies of scale. In addition, we find that the titling component of the reform,
which improved transfer rights, contributed to an outflow of surplus labor from the countryside,
decreasing the total productivity of land. We also discuss a criticism of the reform, that it
increased tensions and conflicts in the Russian village, to explore whether conflicts limited the
positive effect of reform.
1 Very preliminary please do not cite without authors permission.
Acknowledgements: Earlier versions of this paper were presented at 2011 ‘Property rights
workshop’ (Moscow, Russia) and seminar at Universidad Carlos III de Madrid. We would like to
thank all the participants of these seminars for advice and comments.
2 Paul Castañeda Dower is from New Economic School (Moscow, Russia) and Centre of
Economic and Financail Research ((Moscow, Russia); Andrei Markevich is from New Economic
School (Moscow, Russia) and University of Warwick (Coventry, UK).
1. Introduction
The 1906 Stolypin reform was one of the largest property rights reforms in Russian history. The
reform introduced a set of policies to address rural poverty and low productivity of agriculture,
affecting almost 13 million peasant households in European part of the empire. The main
components of the reform granted peasants an opportunity to exit the commune, switching from
communal to individual land ownership, and, in addition, an opportunity to consolidate their land
strips into one allotment. Over the nine years of reform implementation (1907-1915) 2,478,000
peasant households exited the repartition commune and 1,221,000 households consolidated their
plots, or about 20 and 10 percent of all households correspondingly.
The Stolypin reform represented the last attempt of the tsarist regime to reform the Russian
economy and the society before the 1917 Revolution. Because of that it received a huge amount
of attention in the historical literature. The conventional view is that the reform failed because
peasants were against the innovation in agrarian organization and, as a result, few peasants took
advantage of the opportunities initiated by the reform. Moreover, the reform intensified tensions
and conflicts in the Russian village, contributing to rather than preventing revolution (Anfimov
1980, Pallot 1999 etc). In contrast, the reform protagonists underline its positive impact on
peasants’ incentives that should have led to the rapid economic development of Russian
agriculture during the years before the First World War. They argue that the low take-up of the
reform was limited by the supply of the reform rather than the demand (Tukavkin 2001,
Williamson 2006, Davydov 2010).
This paper undertakes an econometric approach to this debate. We take advantage of
province level data regularly published by imperial authorities to evaluate the effect of the
reform onto agricultural productivity. We find that the enclosure component of the reform indeed
caused an increase in land productivity because of better usage rights, as supporters of the reform
argued. At the same time, the improvement in transfer rights, produced by the titling component
of the reform, contributed to an outflow of surplus labor from the countryside, decreasing total
2
productivity of land. We find that a net effect of exits from the commune, represented by title
conversions from communal to individual tenure, onto land productivity is negative, as reform
skeptics believed. We also cannot reject the argument that there is a negative impact of tensions
and conflicts associated with the reform onto agricultural productivity.
Beyond the historical interpretation of the Stolypin reform, our paper contributes to the
economics literature on property rights. We take advantage of historical peculiarities of the
Stolypin reform to disentangle the effect of changes in transfer and exclusion rights from the
effect of changes in usage rights. We find that the latter effect associated with consolidation and
enclosure of peasant allotments is more important, at least in the short-run. This finding is in
contrast to the modern view of the effect of other historical examples of enclosures (Allen 1982,
Hoffman 1991).
2. Russian pre-revolutionary agriculture and the Stolyin reform: an overview.
2.1.
Institutional settings of Russian agriculture after the emancipation: the commune.
The peasant commune dominated the institutional landscape of the Russian village after the
abolishment of serfdom in 1861. The reform empowered the commune with broad decisionmaking authority over local issues both political and economic ones; the commune had to
replace the serf-owner as the source of power in the village. In the economic sphere, in
particular, the commune regulated the majority of peasants’ agricultural activities under the
system of open fields. Each household cultivated a number of separate and scattered small strips,
onto which the peasant land was divided, and accordingly had to coordinate their activities with
their neighbors, i.e. the usage rights in land was limited under the commune system.
Peasants’ transfer and exclusion rights in land varied with the commune type. They were
very limited in repartition (peredel’naya) communes which composed about eighty percent of all
communes. Family ownership and private property in land did not exist in repartition communes
because of the periodical redistribution of allotments between commune members. Accordingly,
peasants could not sell, lease, mortgage or transfer legally their strips under communal tenure
3
and periodical redistributions undertaken by the decision of the commune majority limited their
exclusion rights in land. In addition, the commune regulated peasant mobility and accordingly
the supply of labor because of mutual responsibility for tax payments. Seasonal workers and all
peasants who wished to temporarily leave their native areas had to get passports from local
communal authorities. In hereditary (podvornaya) communes, there were no repartitions and
peasants enjoyed full exclusion rights in land. They also had better land transfer rights but still
limited.
With the exception of three Baltic provinces (Lifliandia, Estliandia, Kurliandia), all
peasants in provinces of the European part of the empire had to be registered in a commune. An
exit from a reparation commune was strongly regulated and demanded the consent of the entire
commune with no compensation for leaving one’s allotment. An exit was easier in a commune
with hereditary tenure; it required an individual either inside or outside the commune willing to
take the land allotment and related obligations. The latter might include monetary compensation
for land, i.e. quasi selling of plots was possible in the commune with hereditary tenure.
Reparation communes strongly dominated in the majority of Russian provinces with the
exception of western, former Polish, provinces of the empire. All communes in Vilno, Kovno,
Grodno, Minsk, Podolia and Volin’ provinces were hereditary ones. The figure 1 presents
geography of the communes by types in early 20th century.
Figure 1 somewhere here.
Repartition status did not necessary mean actual land redistribution, however. About one
third of all repartition communes in early 20th century did not have repartitions of land since the
emancipation. Their number was larger in the black soil provinces because of the higher
profitability of agriculture. Rich peasants did not want to lose their land and lobbied against
repartitions which were supposed to produce more equal distribution of land within the
commune (Ziryanov 1992). Accordingly, land inequality among peasants within the commune in
these provinces was higher.
4
There is a long discussion in historical and economic literatures on the economic
consequences of the institution of the commune and the existence of the ‘agrarian crises’ in the
post-emancipation Russian empire (e.g. Anfimov 1980). The standard argument, ascending to
Vladimir Lenin (1899) and Alexander Gershenkron (1962), is that the free-rider problem caused
by collective responsibility and poor property rights (especially in part of exclusion and transfer
rights in repartition communes) created disincentives for peasants to invest in land; the strip
system and poor usage rights locked peasants in backwardness agricultural techniques limiting
the distribution of modern machinery; finally, restrictions on land sales and peasants’ mobility
froze the development of factor markets, forced peasants to overinvest in agriculture, producing
inefficient allocation of land and labor. All together these produced a negative impact on
efficiency, that was almost three times lower in Russian in 1913 than in England (Anfimov et. al.
1995 p. 80), plunging Russian agriculture into crisis. However, estimations of national income of
the Russian empire by Paul Gregory (1982) failed to reveal stagnating agricultural productivity;
on the contrary, he found rapid growth of peasants’ incomes in the late XIX – early XX
centuries. His macro reconstruction suggests that the commune was a more flexible institution
than historians traditionally thought. Micro evidences on communes from Moscow province
recently analyzed by Stephen Nafziger (2007, 2010) support this new complicated picture.
Nafziger found a negative correlation between the number of repartitions and agricultural
productivity but he also demonstrated that repartitions themselves were not exogenous for
peasants and could act as substitution for undeveloped factor markets.
2.2. The Stolypin reform.
2.2.1. Legal Framework.
The Stolypin reform (1906) was the largest institutional reform in Russian agriculture since
the Emancipation. Politically, the reform was a governmental response to the Russian 1905
revolution and a wave of peasant unrests. The reform aimed to construct a class of small
landowners in the Russian countryside who would provide political support of the government.
5
Economically, the reform attacked the problem of poverty and inefficiency of Russian
agriculture, introducing a whole set of prospects available for peasants.
The key innovation of the Stolypin reform was that peasant got an opportunity to chose
their institutions settings, including property rights. In repartition communes peasants could
choose between staying in and exiting from the commune. The consent of the commune was not
necessary anymore. Under the 09.11.1906 decree, individual peasants could appeal for
conversion of their land title from communal to personal property, i.e. hereditary tenure.3 If an
applicant failed to reach a consensus with his commune on the precise conditions of exit, the
local authority were empowered to solve such disputes. The final exit decisions had to be
approved by local peasant courts (uezdnij krestyanskii sezd). An exit from the repartition
commune and title conversion meant a substantial improvement in land exclusion and transfer
right of an individual household. After the exit household land was protected from possible
future repartitions; the household could also sell their strips to other peasants, to lease it legally
or to mortgage in the Peasant bank. In hereditary communes the reform granted peasants rights
to sell, to lease or to get mortgage their plots in the Peasant bank legally.
The Stolypin reform also opened an opportunity for peasants to consolidate their plots in
joint allotments (khutora and otruba). By a two-thirds majority, a commune – both repartition
and hereditary ones - could vote for general redistribution in separated allotments. Alternatively,
in a repartition commune an individual household could demand for consolidation of its strips to
one place. In the latter case, the commune could try to block an individual request if
consolidation was ‘impossible or inconvenient’ and meet the demand with money. Again local
authorities were in charge of resolving disputes. Title conversion was a pre-requisite for
consolidation under the 1906 decree. The consolidation changed greatly the land usage right of
the household; after consolidation the commune could not force an individual household what to
3 To be precise it was hereditary tenure with personal ownership. Standard hereditary tenure
supposed family ownership existed. In hereditary communes heads of households could also
convert titles of their allotments from family to personal ownership.
6
cultivate anymore.
A household consolidated its land could immediately shift to a new
agricultural technique after the event, in particular a consolidated household could take
advantage of economy of scales if it was applicable. At the same time one might argue that the
consolidation improved slightly transfer rights, as it might be easier to sell a larger plot than
separated strips.
Initial conditions of the Stolypin reform were modified by 14.06.1910 and 29.05.1911
laws. The 1910 law simplified exit procedures for households in those communes where there
were no repartitions since the very act of emancipation. Such households could apply for a titling
certificate (udostoveritel’nii akt) for lands in their possession and could automatically get it
without any discussion with the commune on the precise conditions of exit. Next, under the
29.05.1911 law, titling conversion was not anymore a pre-requisite for consolidation, i.e.
consolidation of plots then could become another way to exit the commune.
2.2.2. The Stolypin reform in action.
Figure 2 presents the dynamics of exits and tilting conversion under both the 1906 decree
and 1910 law. Under the 1906 decree about one and a half hundred thousand households left the
commune in an average year with a spike of seven hundred thousand during the first two years
after the reform. In total, by January 1st, 1916 there were 2, 008, 432 exits with privatized land
of 14, 122, 798 desiatinas or 15 429 157 hectares. In addition, 469, 792 households acquired
titling certificates under the 14.06.1910 law, or about 80,000 annually. All in all 22% of
households privatized 14% of communal land over nine years of the implementation of the
reform.
Figure 2 somewhere here.
The share of households exiting the communes and converting land titles was larger in
provinces with better quality of soil, being the highest in Southern (Kherson, Ekaterinoslav and
Tavrida) and Black Earth provinces (Kiev and Kursk), Mogilev province in the West and Samara
province in the East. Historians of the reform (Zyryanov 1992) pointed out that within provinces
7
households possessing ‘extra’ land allotments, which they would likely to lose under the next
repartition, were among those who tried to get out of the commune first of all because the reform
allowed them to keep these plots. Conflicts and tensions with the rest of the commune were
inevitable outcome in such cases. Only a quarter of exiting households managed to get an
agreement with the commune on the precise conditions of their exits during one month after
application as the law prescribed (Williamson 2006 p. 149). Better options outside agriculture
was another factor promoting exits from the commune; in this case peasants took an advantage
of the reform to get additional finance that they would lose under the old rules.
Figure 3 presents the consolidations dynamic. The number of households consolidated
their plots both individually and in groups was steadily increasing by 1910 year and remained
roughly at the same level after. The 1912 year represented a one year decline in consolidations.
Individual consolidations constituted a majority of all consolidations. After 1910 year there were
about 100,000 individual consolidations per year and 60,000 households consolidated their
allotments jointly in groups. Joint consolidation required a two-thirds majority and, accordingly,
less likely occurred in the communes with high level of tensions and conflicts because the
collective action problem would not be solved. In addition historians of the reform (Pallot 1999)
state that individual consolidations under the reform often caused new conflicts between
separators and the commune. Anecdotal evidences suggest that in order to promote the reform
local authorities consolidated the best land for individual separators with the protest of the rest of
the commune. But conflicts and tensions rarely took an open form. Usually communes tried to
sabotage individual consolidations and then prevent their normal operations. Their most
powerful weapon was to block access to commune pasture and forest which were not subject for
consolidation by law. Such blockade would be illegal but many communes organized it in
practice. The other anti-enclosure actions included pasturing livestock on enclosure’s crops,
arson and finally but in rare case open violence.
Figure 3 somewhere here.
8
Despite within commune tensions and conflicts the demand for the reform was much
higher than the supply. Numbers of applications both to exit and to consolidate were much
higher than actual number of applications to exit approved by local courts and number of
consolidations undertaken in practice. There was a shortage of officials able to implement the
reform quickly. The discrepancy was especially big for consolidations; by the late 1915 only a bit
more than two millions household consolidated their plots (either individually or in groups)
about more than six million applied for. Over all years of the reform implementation there was a
shortage of land survey engineers; their number in a region determined actual number of
consolidations (Davydov 2011). There was a lot of bureaucratic red tape in approval of exits
procedure, because court officials were overburden with other responsibilities. In 1908, the
central government acknowledged the slow confirmation of exits by local courts and tried to
reorganize their work, but without great success (see, for example, decree of the Ministry of
Internal Affairs issued on 30.12.1909 and 14.06.1910 Ministry of Internal affairs 1910, 1 p. 15;
1912, 3 p. 106).
3. Hypothesis
Improvements in exclusion rights should improve dramatically peasants’ incentives to invest in
land because of the lack of repartition threat and accordingly to increase land productivity, but
we argue that time was needed to launch this mechanism in full scale. The Stolypin reform was
in operation for less than a decade only, so we can not verify the hypothesis of a long-run effect
of changes in property rights initiated by the reform. In the short-run, it is more difficult to
predict the effect of exiting onto agricultural productivity. However, at first glance, we argue that
the reform should have had little effect on productivity. First, a title conversion without
consolidation meant little from that point of view because strips of the Russian peasant were
often too narrow – sometimes even several meters – and intermixed with other peasants’ plots,
placing restrictions (sometimes even formal rules about use) on usage rights after exiting the
commune. In such situation individual farmers had to continue to coordinate their decisions
9
about what to cultivate, when to sow etc. with the commune. Second, we could see an effect of
individual exits due to a reallocation of land to the most productive individuals; however, land
markets were not that well developed. Some institutional limitations remained even after the
reform, in particular peasants could not sell their allotments to non-peasants and there were
limits how much former commune land one peasant household could accumulate.
Upon closer inspection, one could argue that adverse selection might take place. Negative
mortality shocks, widespread in such a poor country as Russian empire in early 20 C., created
temporal ‘surplus’ of land in particular households. As it is discussed in the historical section,
these households could rush to take advantage of the reform and to privatize land which they
would lose otherwise. If these households were also in lack of labor because of mortality shock,
this could lead to negative selection.
Better land liquidity caused by title conversion and better transfer rights changed
household’s opportunity costs and eased financial constraints; peasants could explore other
economic activities without losing income from land. Indeed, Chernina et al. (2011) find positive
impact of the Stolypin reform onto internal migration. If there was an overinvestment of labor
into agriculture as many historians argue (Allen 2003) and inefficient allocation of inputs, titling
reform could remove surplus of labor from the countryside decreasing average land productivity
but increasing marginal productivity of labor.
Finally, exits and land privatization were costly and could have negatively impacted
agricultural productivity temporarily because of the strain on peasants’ budgets. As it is discussed
in the historical section, the exiting procedure was quite complicated that implied non trivial
transaction costs. In particular, exits and land privatization may have produced conflicts and
these conflicts could affect productivity negatively. One could also expect that conflicts were less
pronounced in repartition communes where no actual repartitions occurred because there were
no temporary winners and losers there.
10
While the commune regulated agricultural activities of individual households and this
regulation persisted even after exiting the commune, this was not the case after consolidation.
Thus, in the short-run, we would expect that changes in land usage rights caused by
consolidation were positive and the most important effects of the reform. Given the discussion
on conflicts we could expect that the usage effect is more pronounced in group consolidations
which occurred in the communes with better collective decision-making and where the reform
less likely generated new conflicts. Consolidations could also increase productivity because of
scale effect; however, this effect probably demanded some time to be pronounced because of
time needed to update input structure, investing into agricultural machines and the like.
4. Data and Methodology
We construct a provincial level dataset on agricultural output, exits and consolidations in the
early 20th C. Russian Empire before and during the Stolypin property rights reform. Using the
variation in the participation in the reform within and across provinces, we are able to estimate
the impact of the two aspects of the reform that should have affected productivity, enclosure and
land certification.
4.1.
Data Description
We combine several official sources published by imperial authorities in the early 20th C. and
figures extracted from the archives by previous generations of historians. First, we use data on
the titling component of the reform from the journal of the Ministry of Internal Affairs (Izvestiya
Zemskogo otdela MVD) and Dubrovskij (1963). Second, we employ data on consolidations from
annual reports of the chief administration of agriculture and land engineering (various titles and
years). Official annual statistical volumes of the Russian Empire are our sources for data on grain
yields, livestock, rural and urban population; official annual volumes of the Ministry of
Agriculture are the source on agricultural wages. We borrow data on railway deliveries of
agricultural machines from Davydov (2010) and data on violent conflicts from Grave and
11
Dubrovskij (1926) and Dubrovskij (1956, 1963). Table A1 of the appendix provides a full list of
our sources.
Data availability determines the number of observations in our dataset. We collect
information for forty-three European provinces of the empire, thirty-four of which repartition
communes united at least 5 percent of rural population (see figure 1). We have average annual
data on these provinces for seven periods, two before and five after the reform: 1905, 1906,
1907, 1908-1909, 1910-1911, 1912, 1913 and 1914, i.e. three hundred and one province-period
observations in total. The availability of exit statistics, published irregularly, determines the
reform periods.4
Table 1 presents summary statistics of our data. We report variables characterizing the
implementation of the Stolypin reform in households per thousand of rural population units in
order to make them comparable over provinces of different size.5 About three and a third
households per thousand of rural population exited repartition communes in an average province
in an average year during the period under consideration, almost three under the 1906 decree and
a half under the 1910 law for communes where there were no actual reparations. If one consider
only year after the reform and the only provinces where there were repartition communes, the
figures are 6.4, 4.7 and 1.7, correspondingly. The number of consolidations was substantially
smaller. Less than one household per thousand of rural population consolidated its allotment in
an average province in an average year between 1905 and 1913. Only about one-third of them
separated their land individually and two-thirds in groups, when all the land in a village was
redistributed into consolidated plots at once. For the reform period only, the figures are larger
4 Due to the same reason of data availability we have to use data on exits from the commune
since November, 6 1906 (the date when the government issued the reform decree) till January, 31
1908 for the 1907 period and since February, 1 1908 till December 31 1909 for the 1908-1909
period.
5 We have to divide number of households participated in the reform by rural population because
we do not have data on households per province. An average rural household in European
Russian was about six persons.
12
1.16, 0.37 and 0.79 households per a thousand of rural population. As one can see, the variation
in reform implementation between provinces was substantial.
Exit confirmation rate (actual exits to applications to exit ratio) and consolidation
implementation rate (actual consolidations to applications to consolidate ratio) demonstrate
reform realization relative to demand for the reform. As described in the historical section, the
supply of the reform was binding because of red tape and shortage of land survey engineers.
Average exits to applications to exit ratio was only twelve percent and consolidations to
applications to consolidate ratio was about twenty-one percent.
The other two variables characterizing implementation of the reform are a within
commune conflicts index and the number of violent conflicts in a province. The former is a
within commune measure constructed as a share of applicants willing to exit who failed to find a
mutual agreement with the commune on the precise terms of exit and had to request local
officials to determine them.6 As one could see, this was the case for more than a half of all
applications. The latter is an aggregate measure of all types of open protests in the countryside
with protests against landlords as the dominant type, i.e. conflicts with peasants with outsiders.
Because of the 1905 Russian Revolution there was a lot of violence in the countryside, almost
fifty cases in an average province in an average year.
Land productivity in Russian European provinces was about eight hundred and forty
three kilograms per hectare. It increased from about seven and a half hundred per hectare in 1905
to almost a ton in 1913. We cannot estimate labor productivity or TFP in the agrarian sector,
since we do not have precise measures of labor or capital in the Russian village. Instead we
employ rural population figures as a proxy for labor and livestock (cows and horses) for capital.
In addition, using data on railway traffic of agricultural machines, we are able to construct a
proxy for this type of capital in agriculture. There were about forty-eight rural citizens, less than
6 Unfortunately, being extracted from secondary works Grave and Dubrovskij 1926, Dubrovskij
1956 and 1963) both figures hare drawbacks. The first index lacks overtime variation being
available only for the whole period under consideration; and the second measure does not vary
over groups of provinces since our sources report it in an aggregate form.
13
one cow and a half horse per hectare of arable land in an average Russian province. In addition,
we use the amount of credit which peasants received under the small loans program launched by
the central government. This variable is a proxy of peasants’ access to credit in general because
they basically did not have any other options (Korelin 1988). As one could see access to credit
was very limited indeed, about one ruble per a thousand of hectares only.
Rural wages during harvest season, share of urban population and migration to Asian part
of the empire characterize three main options available to the peasant at that time: to become a
hired worker either in agriculture or in a city, or migrate to Southern Siberia where the
government provided land. Rural wages during the harvest peak time were about 30 rubles per
month or about a quarter of 1913 GDP per capita (Markevich and Harrison 2011). Urban
settlements were growing very fast but their average share was only about thirteen percent. Level
of migration to Siberia was high (about 3 mln people over ten years) but much less impressive in
relative terms; roughly one household per three thousand rural population migrated to Siberia
annually. Number of sales of peasants’ allotments is an aggregate proxy for number of peasants
quitting self-employment in agriculture and/or could represent the land market’s capacity for
reallocating land to the highest valuing user.
Finally, we have data on two other important characteristics of Russian European
provinces, the share of repartition communes where no actual repartitions occurred since the
Emancipation and the presence of local self-governance (zemstvo). Share of repartition
communes without actual repartitions measures the strength of the repartition commune in a
province at the start of the Stolypin reform. These repartition communes could not solve
collective action problems. By its nature this variable is stable over time but it varies a lot from
zero (in provinces where there were no repartition communes) to almost ninety percent in
Kaluga province where the repartition commune basically did not function. We view the zemstvo
dummy as a very important determinant of agricultural productivity in a province because
zemstvo initiated various programs aimed to develop peasant agriculture. In particular, they
14
invested a lot into the disseminating of advanced knowledge and techniques as well as basic
education among rural citizens. Local self-governments were introduced by the 1864 law but
only in about half of all provinces. There were several expansions of number of provinces with
zemstvo later, including one such expansion in 1911. Accordingly, this variable almost does not
vary over time.
4.2.
Econometric Specification
To test our hypotheses on the impact of the Stolypin reform on peasant agriculture, we will use
econometric techniques. Data availability to a large extent determines our preferred econometric
specification. Our main dependent variable is yield per hectare. We estimate the effect of the
reform onto land productivity and not onto labor productivity or TFP because the former is the
most precisely measured. Our main explanatory variables track the implementation of the tilling
and enclosure components of the Stolypin reform, namely number of exits from the commune
accompanied by title conversions and the number of consolidations, both per thousand peasants.
We use a linear OLS regression model with fixed region (group of provinces) and year effects.
We prefer pooled OLS specification to the specification with provincial fixed effect because one
of our important control variables, namely zemstvo dummy, does not vary over time. We also
believe that the variation across provinces is an important source of identification. We employ
current reform implementation variables because we are interested in a short-term effect of the
reform.
To be precise, we estimate the following equation:
Yieldit = α + β*Exitsit + Ω* (Consolidationsit) + ∆ (Controlsit) +Λ(Regiont)+Ϭ(Yeari)+εit (1)
where subscripts i and t index provinces and years, respectively. Yield is the output of grain per
hectare of arable land; Exits and Consolidations are measures of the reform implementation. In
different specifications, as Exits, we use either total exits per thousand of rural population, or
exits under the 1906 decree and exits under the 1910 law separately. In the latter case, Exits is a
vector of variables rather than a variable. Similarly, in different specifications, as Consolidations,
15
we use either total number of households that consolidated their plots or separate both types of
consolidations, individual and in group.
As controls we employ rural density per hectare of arable land, number of cows and
horses, and agricultural machines per hectare of arable land to account for labor and capital
inputs. We use rural wages to take into account supply and demand for labor in a province. We
also use amount of credit per hectare of arable land and share of urban population to control for
access to credit and to industrial markets, accordingly. We control for zemstvo because of its
discussed role in agriculture promotion. Further, once exits were not possible in provinces
without reparation communes, we add a dummy to control for that. Finally, we control for year
fixed effects, Yeari , and twelve region fixed effects, Regiont. Year effects take into account any
time trends, such as macroeconomic shocks; regional fixed effects account for unobserved
regional characteristics, such as climate or quality of land.
The primary concern with (1) is potential endogeneity because of selection of various
types. We address this problem by taking advantage of the limited supply of the reform because
of red tape and shortage of land survey engineers. We employ 2SLS approach, instrumenting
either for number of exits or consolidations with exit confirmation rate and consolidation
implementation rate, correspondingly.
5. Results and Analysis
Table 2 reports our baseline results which come from running the pooled OLS specification in
(1). While these baseline results suffer from endogeneity concerns, they provide a useful starting
point. We see in columns 1 and 2 of Table 2 that the coefficient on share of exits is negative,
significant at the 5% level, and the coefficient on the share of consolidations is positive,
significant at the 10% level. Thus, at first glance, the results require a more nuanced explanation
than the common assertion that improvements in property rights lead to higher productivity.
They support the hypothesis that in the short-run changes in the usage right were more important
than changes in exclusion and transfer rights.
16
Table 2 somewhere here.
In columns 3 and 4, we separate total consolidations into individual consolidations and
village-wide consolidations. As discussed in section 3, we expect that the effect of better usage
rights should be stronger for village-wide consolidations. Indeed, we see that the coefficient is
larger and more highly significant than for the total effect. The coefficient on individual
consolidations is negative but insignificant. In column 5, we use all three measures of reform
participation together. The effect of exits remains unchanged while the effect on village-wide
consolidations becomes stronger. The overall changes in land productivity associated with the
reform were substantial; according to the last specification, one standard deviation increase in
exits was associated with 3 percent decrease in land productivity and one standard deviation
increase in group consolidations with 4.2 percent increase in land productivity. Because both
mean and variation in exits was larger than in group consolidations, the magnitude of the effects
estimated in percentage points of households exited and consolidated in groups7 is smaller for the
former and larger for the latter: 0.3 and 2.1 percents decrease and increase in land productivity or
two and a half and seventeen kilograms per hectare accordingly.
All our controls have intuitive signs while many of them are often insignificant. The
coefficient of rural density, i.e. proxy for labor input, is always positive and significant at 10%
level in four out of five specifications but relatively small and is in line with the statement of
surplus of labor in the Russian commune village. One percentage point increase in rural density
would lead to only 0.35 percent increase in land productivity. The coefficient on zemstvo dummy
is always positive and significant in three out of five specifications. Land productivity was four
and a half percent higher in provinces with local self-governments. Coefficients on cows, horses
and agricultural machines are positive but always insignificant. Horses and machines as factors
of production are positively correlated with grain output per hectare, while cows negatively,
7 To estimate the magnitude of the effect in household percentage points we multiply coefficients
of reform implementation variables (household exits and consolidations per a thousand rural
citizens) by six (an average household size) and divide by ten.
17
since animal husbandry was more developed in provinces where it had a clear comparative
advantage relative to grain production. All regional dummies except one have significant signs
capturing local advantages and disadvantages in agriculture.
The most counterintuitive result in the baseline results is the negative effect of share of
exits. Thus, we turn first to analyzing why we observe a negative coefficient. In table 3, we
address endogeneity and selection concerns using an instrumental variables approach. We report
the first stage in column 1 and the second stage results in column 2. The F-statistic of 170.4
suggests that there is ample explanatory power to use exit confirmation rate index as an
instrument. The effect of exits remains negative and significant suggesting that it is unlikely that
selection drives the results. Moreover, the magnitude of the effect becomes a bit larger. One
percentage point increase in exits caused 0.35 percent decrease in land productivity.
Table 3 somewhere here.
In columns 3-7 of table 3, we explore the hypothesis that the negative effect is not at the
margin but rather simply reflects the withdrawal of surplus labor from the agricultural sector. We
try to measure this movement in three ways. First, we look at the land sales in column 3 and add
an interaction between land sales (normalized to zero mean) and rural density in column 4. Next,
we add share of migrants to Siberia (normalized to zero mean) in column 5 and interact this with
rural density in column 6. Lastly, we add an interaction term between urban share (normalized to
zero mean) and share of total exits in column 7 to proxy for rural to urban movement. The results
are generally in favor of overinvestment hypothesis. In column 3 the coefficient on number of
sales is insignificantly different from zero, but it becomes negative and significant at 10% level
once we add the interaction of number of sales with rural density in column 4, implying that the
effect of quitting self-employment in agriculture dominated in peasants’ sales. Moreover, the
coefficient on the interaction itself is positive and significant, i.e. reallocation of recourses
increased marginal productivity of labor in agriculture. One standard deviation in number of land
sales in a province increased marginal land productivity almost by fifty percent. The coefficients
18
on internal migration variable, a direct measure of peasants outflow from the countryside in
European Russian, and its interaction with rural density reported in column 5 and 6 demonstrate
similar effect. There were more migrants where land productivity was lower, but internal
migration was associated with increase in marginal productivity by labor. One standard deviation
increase in number of migrants to Siberia increased marginal land productivity almost by one
third. More importantly, once we account for migration, the coefficients on exits variable become
insignificantly different from zero. Finally, the last column demonstrates that exits were more
negatively associated with productivity in the most urbanized provinces, where peasants had
better perspectives outside agriculture.
Exits and land privatization were costly and could have negatively impacted agricultural
productivity temporarily because of the strain on peasants’ budgets. As it is discussed in the
historical section, the exiting procedure was quite complicated that implied non trivial
transaction costs. In particular, exits and land privatization may have produced conflicts and
these conflicts could affect productivity negatively. One could also expect that transaction costs
and conflicts were less pronounced in repartition communes where no actual repartitions
occurred because there were no temporary winners and losers there.
In table 4, we investigate the hypothesis that transaction costs explain the negative
relationship between exits and land productivity. First, we separate our exits measure into two:
exits under the 1906 decree and exits in repartition communes who have not had a repartition
since the emancipation (exits under the 1910 law) both per thousand of rural population.
Transaction costs should be in general lower in the case of exits under 1910 law since there are
fewer land transfers to trace the priority of claims. Column 1 shows that 1906 exits variable has a
negative and significant coefficient and the 1910 exits variable has a positive but insignificant
one generally consistent with the transaction costs explanation. However, this evidence is also
consistent with the conflict hypothesis. We take advantage of historical peculiarities of the
reform trying to distinguish between these two interpretations.
19
Table 4 somewhere here.
The reform has another built in feature that varied transaction costs. After the 1910
decree, transaction costs decreased for all types of exits in the communes with and without actual
repartitions, so we can test the hypothesis that transaction costs explains the negative effect of
exits under 1906 decree by focusing on exits before 1910. Column 2 reports that the exits before
1910 can not explain the negative effect that we observe. In column 3, we include lagged exits to
see if the negative effect is merely temporary. The coefficient on the lagged share of exits is
negative but insignificant. To summarize, there is mixed evidence about transaction costs.
For conflicts, we need to be careful to separate selection due to commune strength and
actual conflicts caused by the reform. Thus, we include our measure of commune strength. In
table 5, we see that in all specifications this variable gives the right sign, stronger communes
have higher productivity but the effect is not statistically significant. Moreover, in column 2,
interacting commune strength with exits, we find no statistical relationship. Next, we include a
measure of internal conflicts in column 3 and a measure of outright violence between peasants
and landlords in column 4. Conflicts are positively related with productivity most likely
reflecting reverse causality. Interacting each of these variables, in columns 5 and 6, gives little
statistical relationship. Thus, the results seem unsupportive of the basic conflict story.
Table 5 somewhere here.
While the evidence clearly supports the labor overinvestment story, both the transaction
costs and conflict stories can not be rejected, and especially the costs explanation. After making a
back of the envelope calculation, we can establish an upper bound on the impact of withdrawing
surplus labor of 2.616 mln tones. We get this figure assuming that 20 percent of those who exited
quitted self-employment in agriculture (Dubrovsky 1963 p. 359) and all of their lands was
withdrawn from cultivation and did not produce any grain. We estimate that total negative effect
of exits is roughly 4.44 mln tones, i.e. there is enough place to other factors be at work even
under our heroic assumption of land withdrawing from the production.
20
Explaining the negative effect using overinvestment and possibly transaction costs or
conflicts gives more solid footing to the observed positive relationship between consolidations
and productivity. However, just as with exits, endogeniety and selection are concerns. Table 6
shows the first stage and second stage results from instrumental variables estimation of the effect
of consolidations, separately for individual and village-wide. We use the specification as in
column 5 of table 2, except that we instrument for consolidations, using the same instrumental
variable for each (so we cannot instrument for both at the same time). Again F-statistics are high
enough (21.7 and 45.5, correspondingly) to suggest that there is enough explanatory power to
use consolidation implementation rate index as an instrument. Comparing the results with table
2, we see evidence of positive selection. Those communes that were worse off had more trouble
consolidating as a village. We also show that there is no scale effect in columns 5 and 6 of table
6. Coefficients on average consolidated plots either individually or in group are both
insignificantly different from zero.
Table 6 somewhere here.
6. Conclusion.
We reestablish a pessimistic view on the impact of the commune onto agricultural
productivity. While the commune may have been flexible in adjusting to economic changes and
peasants’ demands within a particular production activity, our results demonstrate that the
restrictive land rights imposed by the commune severely limited the rural households’ production
function in general, hampering activities in agricultural and non-agricultural sectors.
The currently fashionable view of enclosure is that it did little to improve agricultural
productivity. Our results suggest that village-wide enclosure increased agricultural productivity
by 1.3 percent per year. Few would claim this represents a modest increase. In fact, if these
estimates are taken seriously, they could explain as much as 0.7 percentage point of GDP growth.
Therefore, the historical context matters a great deal in understanding how institutions affect
economic development.
21
Finally, we can speculate about a widespread criticism of the reform that, by increasing the
level of conflicts, it led the Russian countryside on a path towards revolution. Our results suggest
that explanations of the greater willingness of peasants to support the revolution that rely on
reform-induced conflicts worsening peasants’ living conditions by crippling agricultural
productivity seem unlikely. However, the commune and the expectation of equal distribution of
resources in the countryside could have sown the seeds of revolution. Anecdotal evidence
suggests that conflicts induced by the reform played exactly along these lines. Undoubtedly, a
more complete understanding of the reform as a cause of the revolution demands further
research.
22
References.
Allen, Robert C. (2003). Farm to Factory. A reinterpretation of the Soviet Industrial
Revolution. Princeton and Oxford: Princeton University Press.
Allen, Robert C. (1982), Enclosure and the Yeoman: The Agricultural Development of the
South Midlands 1450-1850. NewYork;OxfordUniversity Press
Anfimov Andrei M. (1980).
Krestyanskoe khozyastvo Evropejskoj Rossii, 1881-1904
[Peasants households in European Russia, 1881 - 1904]. Moscow: Nauka.
Central Statistical Committee of the Ministry of Interior Affairs (1905-1916). Statisticheskii
ezhegodnik Rossii v … godu [Statistical yearbook of Russia in … year]. Annual volumes for
1904 - 1915, S-Peterburg,
Central Statistical Committee of the Ministry of Interior Affairs (1902). Statisticheskii
vremennik Rosijskoj imperii [Statistical periodic of Russian Empire]. Vol. 54. Urozhaj 1902 goda
[Crop Yield in 1902], Saint-Petersburg.
Chief administration of agriculture and land engineering (1908, 1909, 1910, 1911, 1912,
1913, 1914). Annual reports of various titles.
Chernina Eugenia, Paul Castaneda Dower and Andrei Markevich (2011). Property Rights,
Land Liquidity and Internal Migration. Mimeo.
Davydov Mikhail A. (2010). Vserossijskij rinok v kotze 19 – nachale 20 vv. I
zheleznodorozhnaya statistika [All-Russian market in the late 19th –early 20th centuries and railway
traffic statistics]. S-Petersburg: Aleteya.
Dubrovskij S.M. (1956). Krest’yanskoe dvizhenie v revolutsii 1905-1907 godov [Peasants’
movement in the 1905-1907 revolution]. Moscow.
Dubrovskij S.M. (1963). Stolypinskaya agrarnaya reforma [The Stolypin agrarian reform].
Moscow.
Gershenkron, Alexander (1962). “Agrarian Policies and Industrialisation: Russia 1861-1917”
in The Cambridge Economic History of Europe” vol. VI.
Grave and Dubrovskij S.M. (1926). Krest’yanskoe dvizhenie nakanune revolutsii 1905 goda
[Peasants’ movement before the 1905 revolution]. In 1905 god [The year 1905]. Vol. 1. Moscow,
Leningrad.
Gregory, Paul (1982). Russian National Income 1885-1913. Cambridge: Cambridge
University Press.
Hoffman, Phillip (1991). “Land Rents and Agricultural Productivity: The Paris Basin, 14501789,” The Journal of Economic History 51, 771-805.
23
Korelin A.P. (1988). Selsko-khozyastvennij kredit v Rossii [Agricultural credit in Russia in
the late 19th – early 20th century. Moscow: Nauka.
Lenin, Vladimir (1899). Razvitie kaitalisma v Rossii [Development of capitalism in Russia].
S-Petersburg.
Markevich Andrei and Mark Harrison (2011), ‘Great War, Civil War, and Recovery: Russia’s
National Income, 1913 to 1928’, The Journal of Economic History, forthcoming
Ministry of Interior Affairs (1908-1915)Novosti Zemskogo Otdela [The News of Zemsky
Department]. Vol. 2, 1908; Vol. 3, 1910; Vol. 3, 1912; Vol. 6, 1914, Vol7, 1915.
Ministry of agriculture. Department of agricultural economics and statistics (19061914).Sbornik po selskomu khozyastvuza … god [Annual agricultural volume for .. year] Annual
volumes for 1906-1914, Saint-Petersburg.
Ministry of internal affairs. (1908-1914). Izvestia zemskogo otdela MVD. [News of provincial
department of the Ministry of internal affairs]. Monthly volumes for 1908-1914, S-Petersburg.
Nafziger, Steven (2007). Land Redistribution and the Russian Peasant Commune in the 19th
century.Mimeo.
Nafziger, Steven (2010) ‘Peasant communes and factor markets in late nineteenth-century
Russia’, Explorations in Economic History, 47: 381–402.
Pallot, Judith (1999). Land Reform in Russia, 1906-1917. Peasant Response to Stolypin’s
Project of Rural Transformation. Oxford: Clarendon press.
Tukavkin V.G. (2001). Velikorusskoe krest’yanstvo I Stolypinskaya agrarnaya reforma. [Great
Russian peasants and the Stolypin agrarian reform] Moscow.
Turchaninov N. (1910). Itogi pereselencheskogo dela [The results of resettlement movement]
vol.1.Saint-Petersburg.
Turchaninov N. (1915). Itogi pereselencheskogo dela [The results of resettlement movement]
vol.2.Saint-Petersburg.
Williams, Stephen F. (2006) Liberal reform in an Illiberal Regime. The Creation of Private
Property in Russia, 1906 - 1915, Hoover Institution Press.
ZyrianovP. (1992). Krest’yanskayaobshchinavEvropeiskoiRossiiv 1907 – 1914 gg.[Peasant
commune in European Russia in 1907-1914], Moscow.
24
Figure 1: A map of provinces in the European part of Imperial Russia
25
Figure 2. Exits and exits in repartition communes without actual repartitions (under 1910
Source. Dubrovksy (1963).
Sources: Annual reports of the chief administration of agriculture and land engineering
26
Table 1. Summary statistics
Variable
Nhhs exited per 1000 peasants (total)
Nhhs exited per 1000 peasants under
1906 decree
Nhhs exited per 1000 peasants under
1910 law (in communes without
actual repartitions)
Nhhs consolidated land per 1000
peasants (total)
Nhhs consolidated land individually
per 1000 peasants
Nhhs consolidated land in groups per
1000 peas
Average size of individually
consolidated plot
Average size of plot consolidated in
groups
Exit confirmation rate (exits to
applications ratio)
Consolidation implementation rate
(consolidations to application ration)
Conflicts index (caused by exits)
N open violence affairs
Yield tones per hectare
Rural density per sq km
Number of cows per hectare
Number of horses per hectare
Agricultural machines in tons per
hectare
Rural monthly wage in harvest
season
Urban share
N migrants per 1000 peasants
Nsales per 1000 peasants
Share of repartition communes
without actual repartitions since
emancipation
Amount of small credit loans per
1000 hectares
Local self-government dummy
(zemstvo)
Repartition province dummy
27
Obs
265
Mean
3.35
Std. Dev.
6.62
Min
0
Max
41
300
2.72
5.52
0
38.86
266
0.45
1.18
0
7.65
301
0.75
1.19
0
8.02
300
0.22
0.43
0
2.9
301
0.54
0.96
0
5.4
300
6.2
16.76
0
213.2
301
6.31
7.66
0
84.77
299
0.12
0.19
0
0.86
301
43
301
301
301
301
301
0.21
0.57
46
0.843
48.07
0.88
0.54
0.27
0.35
74.77
0.26
21.55
0.6
0.25
0
0.95
0
0.13
5.35
0.2
0.14
2.43
43
520
1.59
113.51
3.91
1.5
294
0.00027
0.00028
0.000001
0.002
297
301
300
301
28.53
0.132
0.31
2.18
9.26
0.122
0.31
1.93
13.5
0.026
0 .003
0
70.2
0.741
1.16
8.7
42
40.37
30.09
0
88.4
298
1.04
2.89
0
44.09
301
301
0.77
0.81
0.42
0.39
0
0
1
1
Table 2. OLS estimates of the effect of exits and consolidations on agricultural productivity.
Pooled
OLS
Pooled
OLS
1
sh_totexits
Pooled
OLS
2
Pooled
OLS
3
Pooled
OLS
4
5
-0.004**
-0.004**
[0.002]
[0.002]
sh_cons_ind
0.016*
[0.009]
sh_cons_ind_sing
-0.007
-0.051
[0.025]
[0.033]
sh_cons_ind_coll
rdensity
Ncowspd
Nhorsespd
agrmachpd
rwage
zemstvo
repartition_province
crcooppd
ushare
[0.010]
[0.012]
0.004*
0.003*
0.004*
0.003*
[0.002]
[0.002]
[0.002]
[0.002]
[0.002]
-0.032
-0.034
-0.056
-0.035
-0.028
[0.088]
[0.097]
[0.095]
[0.097]
[0.089]
0.047
0.055
0.082
0.054
0.037
[0.193]
[0.203]
[0.201]
[0.202]
[0.190]
82.082
32.684
41.431
33.018
79.198
[73.035]
[66.264]
[66.625]
[66.392]
[69.675]
0.006**
0.005
0.005*
0.005
0.006**
[0.003]
[0.003]
[0.003]
[0.003]
[0.002]
0.049
0.103*
0.108*
0.099*
0.038
[0.046]
[0.056]
[0.058]
[0.056]
[0.048]
0.104
0.072
0.071
0.071
0.105
[0.076]
[0.074]
[0.075]
[0.074]
[0.077]
0.003
-0.001
-0.000
-0.002
-0.001
[0.011]
[0.002]
[0.002]
[0.002]
[0.010]
-0.065
-0.015
-0.012
-0.027
-0.097
[0.124]
[0.121]
[0.120]
[0.122]
[0.126]
Yes
Yes
Yes
Yes
Yes
0.546***
Yes
0.799***
Yes
Time Effects
Yes
28
0.030**
0.003
Regional Effects
Constant
0.025**
0.830***
Yes
0.552***
Yes
0.341**
[0.144]
Observations
R-squared
[0.189]
254
0.581
[0.157]
296
0.565
295
0.565
Robust standard errors in brackets. *** p<0.01, ** p<0.05, * p<0.1
29
[0.187]
[0.143]
296
0.567
254
0.588
Table 3. IV estimates of the effect of exits, selection and the overinvestment hypothesis.
First Stage
2SLS
Pooled
OLS
1
2
3
sh_totexits
sh_cons_ind_sing
sh_cons_ind_coll
rdensity
Ncowspd
Nhorsespd
agrmachpd
rwage
zemstvo
repartition_province
ivexitsimple
Pooled
OLS
Pooled
OLS
4
Pooled
OLS
5
Pooled
OLS
6
7
-0.005**
-0.004***
-0.004**
-0.002
-0.001
-0.004***
[0.002]
[0.002]
[0.002]
[0.002]
[0.002]
[0. 001]
0.683
-0.049
-0.057
-0.058*
-0.057
-0.057
-0.053
[0.595]
[0.031]
[0.034]
[0.034]
[0.036]
[0.037]
[0.033]
1.234***
0.031***
0.027*
0.032**
0.032**
0.031**
0.033***
[0.427]
[0.011]
[0.014]
[0.014]
[0.014]
[0.014]
[0.011]
0.001
0.003*
0.003*
0.003*
0.003*
0.003**
0.003*
[0.033]
[0.002]
[0.002]
[0.002]
[0.002]
[0.001]
[0.002]
-1.525
-0.024
-0.025
-0.028
-0.024
-0.028
-0.031
[1.338]
[0.083]
[0.089]
[0.087]
[0.085]
[0.079]
[0.090]
1.743
0.027
0.038
0.044
0.027
0.022
0.052
[2.434]
[0.179]
[0.190]
[0.184]
[0.181]
[0.164]
[0.190]
3,247.520**
80.919
75.817
61.878
50.198
48.243
124.173*
[1,268.642]
[64.757]
[68.799]
[65.984]
[66.304]
[62.612]
[70.518]
0.046
0.006***
0.006**
0.006**
0.006**
0.006**
0.005**
[0.067]
[0.002]
[0.003]
[0.003]
[0.003]
[0.003]
[0.002]
0.354
0.032
0.038
0.029
0.014
0.017
0.042
[1.113]
[0.046]
[0.048]
[0.045]
[0.045]
[0.044]
[0.049]
1.002
0.125
0.103
0.096
0.112
0.097
0.090
[0.879]
[0.086]
[0.076]
[0.070]
[0.069]
[0.062]
[0.079]
0.073
-0.001
-0.002
-0.006
-0.007
-0.006
0.001
[0.239]
[0.009]
[0.010]
[0.010]
[0.010]
[0.010]
[0.010]
-4.179*
-0.096
-0.099
-0.074
-0.069
-0.049
0.012
[2.421]
[0.116]
[0.128]
[0.121]
[0.129]
[0.119]
[0.100]
0.007
-0.043*
-0.036
-0.015
[0.014]
[0.024]
[0.025]
[0.027]
0.001***
0.001**
0.001
25.923***
[3.907]
crcooppd
ushare
sh_hhsales
Nsh_hhsalesrdensity
30
[0.000]
sh_migrants
[0.000]
[0.000]
-0.054*
-0.262***
[0.030]
[0.088]
Nsh_migrantsrdensity
0.003**
[0.001]
sh_totexitsNushare
-0.040***
[0.012]
Regional Effects
Yes
Time Effects
Yes
Constant
Yes
Yes
Yes
Yes
-4.837**
Yes
0.409***
Yes
0.790***
Yes
0.834***
Yes
0.629***
[2.287]
[0.112]
[0.142]
[0.142]
[0.167]
Observations
R-squared
253
0.769
253
0.588
254
0.588
Robust standard errors in brackets. *** p<0.01, ** p<0.05, * p<0.1
31
254
0.594
Yes
Yes
Yes
Yes
0.688***
0.787***
[0.165]
[0.141]
253
0.602
253
0.611
254
0.597
Table 4. The effect of transaction costs related to exits on agricultural productivity.
Pooled OLS
Pooled OLS
1
sh_exits
sh_nrep_exit
sh_cons_ind_sing
sh_cons_ind_coll
rdensity
Ncowspd
Nhorsespd
agrmachpd
rwage
zemstvo
repartition_province
crcooppd
ushare
sh_exits19001910
Pooled OLS
2
3
-0.004**
-0.004**
-0.004**
[0.002]
[0.002]
[0.002]
0.003
0.006
0.003
[0.009]
[0.008]
[0.009]
-0.045
-0.034
-0.035
[0.035]
[0.040]
[0.035]
0.028**
0.027**
0.029**
[0.013]
[0.013]
[0.013]
0.003*
0.003*
0.003*
[0.002]
[0.002]
[0.002]
-0.028
-0.032
-0.031
[0.089]
[0.089]
[0.089]
0.040
0.047
0.048
[0.190]
[0.192]
[0.192]
76.158
84.946
76.216
[72.042]
[74.693]
[71.529]
0.006**
0.006**
0.006**
[0.003]
[0.003]
[0.003]
0.034
0.030
0.031
[0.048]
[0.050]
[0.049]
0.092
0.084
0.087
[0.076]
[0.079]
[0.079]
-0.001
-0.001
0.000
[0.010]
[0.010]
[0.010]
-0.094
-0.101
-0.097
[0.133]
[0.134]
[0.134]
-0.005
[0.007]
lagsh_exits
-0.002
[0.003]
32
Regional Effects
Yes
Time Effects
Yes
Constant
Yes
Yes
0.794***
Yes
0.782***
Yes
0.608***
[0.143]
[0.147]
[0.155]
Observations
R-squared
254
0.587
254
0.588
253
0.579
Robust standard errors in brackets. *** p<0.01, ** p<0.05, * p<0.1
33
Table 5. The effect of reform-induced conflict on agricultural productivity.
Pooled
OLS
Pooled
OLS
1
sh_exits
sh_nrep_exit
sh_cons_ind_sing
sh_cons_ind_coll
nrepsh1905
rdensity
Ncowspd
Nhorsespd
agrmachpd
rwage
crcooppd
ushare
zemstvo
repartition_province
sh_exitsnrepsh1905
34
Pooled
OLS
2
Pooled
OLS
3
Pooled
OLS
4
Pooled
OLS
5
6
-0.004**
-0.005
-0.005**
-0.004**
-0.004
-0.006**
[0.002]
[0.004]
[0.002]
[0.002]
[0.002]
[0.003]
0.006
0.006
0.002
0.005
0.005
0.003
[0.007]
[0.007]
[0.008]
[0.008]
[0.008]
[0.008]
-0.039
-0.039
-0.080**
-0.070**
-0.069**
-0.070**
[0.034]
[0.034]
[0.035]
[0.034]
[0.034]
[0.034]
0.028**
0.028**
0.031**
0.029**
0.029**
0.030**
[0.013]
[0.013]
[0.014]
[0.014]
[0.014]
[0.014]
-0.001
-0.001
-0.001
-0.001
-0.001
-0.001
[0.001]
[0.001]
[0.001]
[0.001]
[0.001]
[0.001]
0.002
0.002
0.003*
0.003*
0.003*
0.003*
[0.002]
[0.002]
[0.002]
[0.002]
[0.002]
[0.002]
-0.050
-0.051
-0.062
-0.068
-0.067
-0.070
[0.091]
[0.090]
[0.092]
[0.091]
[0.091]
[0.091]
0.102
0.105
0.165
0.170
0.168
0.171
[0.206]
[0.204]
[0.207]
[0.204]
[0.205]
[0.203]
79.001
78.612
104.905
121.017
123.409
131.192*
[71.094]
[70.973]
[73.927]
[75.598]
[76.558]
[75.814]
0.006**
0.006**
0.005*
0.005**
0.005**
0.005**
[0.002]
[0.002]
[0.003]
[0.002]
[0.002]
[0.002]
-0.003
-0.003
-0.011
-0.012
-0.012
-0.012
[0.009]
[0.009]
[0.009]
[0.009]
[0.009]
[0.009]
-0.110
-0.108
-0.099
-0.105
-0.108
-0.113
[0.133]
[0.132]
[0.143]
[0.141]
[0.142]
[0.140]
0.042
0.042
0.050
0.025
0.028
0.034
[0.050]
[0.050]
[0.049]
[0.054]
[0.057]
[0.058]
0.100
0.105
0.057
0.080
0.077
0.071
[0.078]
[0.081]
[0.081]
[0.091]
[0.095]
[0.094]
0.000
[0.000]
conflict2
0.246***
0.273***
0.281***
0.279***
[0.073]
[0.078]
[0.083]
[0.078]
0.000**
0.000**
0.000**
[0.000]
[0.000]
[0.000]
violenceregion
sh_exitsconflict2
-0.002
[0.006]
sh_exitsviolence
0.000
[0.000]
Regional Effects
Yes
Time Effects
Yes
Constant
Yes
Yes
Yes
0.798***
Yes
0.798***
Yes
0.703***
Yes
0.646***
Yes
0.650***
[0.148]
[0.147]
[0.154]
[0.158]
[0.159]
Observations
R-squared
254
0.589
254
0.589
254
0.607
254
0.614
Robust standard errors in brackets. *** p<0.01, ** p<0.05, * p<0.1
35
Yes
Yes
Yes
0.669***
[0.162]
254
0.615
254
0.615
Table 6. IV estimates of consolidations and scale effects.
sh_totexits
First
Stage
First
Stage
2SLS
2SLS
1
2
3
4
rdensity
Ncowspd
Nhorsespd
agrmachpd
rwage
zemstvo
repartition_province
IVsimple_ind
crcooppd
ushare
av_cons_ind_coll
Pooled
OLS
5
6
0.005
0.017**
-0.005**
-0.005***
-0.004**
-0.004**
[0.004]
[0.007]
[0.002]
[0.002]
[0.002]
[0.002]
0.248
0.063
-0.074**
-0.048
-0.051
[0.243]
[0.116]
[0.035]
[0.033]
[0.033]
0.065
0.014
0.068**
0.028**
0.030**
[0.070]
[0.019]
[0.034]
[0.012]
[0.012]
sh_cons_ind_sing
sh_cons_ind_coll
Pooled
OLS
-0.002
-0.015***
0.003*
0.003**
0.003
0.003*
[0.002]
[0.003]
[0.002]
[0.002]
[0.002]
[0.002]
-0.274**
-0.613**
-0.003
-0.009
-0.030
-0.028
[0.134]
[0.253]
[0.094]
[0.088]
[0.089]
[0.089]
0.227
0.796**
0.025
0.017
0.034
0.037
[0.170]
[0.383]
[0.185]
[0.179]
[0.188]
[0.192]
324.511
751.481
58.865
64.642
115.367
75.484
[242.404]
[508.107]
[75.569]
[67.478]
[80.721]
[70.456]
0.008
0.010
0.005*
0.006**
0.006**
0.006**
[0.005]
[0.011]
[0.003]
[0.002]
[0.003]
[0.002]
-0.160
0.099
0.054
0.029
0.049
0.036
[0.131]
[0.147]
[0.046]
[0.048]
[0.045]
[0.049]
-0.053
-0.163
0.107
0.108
0.097
0.105
[0.056]
[0.135]
[0.075]
[0.075]
[0.078]
[0.078]
0.639**
1.726***
[0.262]
[0.384]
-0.037
0.069
0.004
-0.003
0.002
-0.004
[0.045]
[0.075]
[0.012]
[0.010]
[0.011]
[0.011]
-0.572
0.158
-0.040
-0.118
-0.128
-0.089
[0.370]
[0.497]
[0.136]
[0.118]
[0.137]
[0.123]
-0.002
[0.002]
av_cons_ind_sing
36
0.001
[0.001]
Regional Effects
Yes
Time Effects
Yes
Constant
Yes
Yes
Yes
-0.173
Yes
0.842**
Yes
0.365***
Yes
0.420***
Yes
0.827***
[0.222]
[0.373]
[0.123]
[0.111]
[0.147]
Observations
R-squared
254
0.547
254
0.631
254
0.569
Robust standard errors in brackets. *** p<0.01, ** p<0.05, * p<0.1
37
Yes
Yes
254
0.579
Yes
0.801***
[0.142]
254
0.590
254
0.589
Table A1. Variables definitions and data sources
Variablename
Sh_totexits
Sh_exits
Sh_nrep_exit
Sh_cons_ind
Sh_cons_ind_sing
Sh_cons_ind_coll
Av_cons_ind_sing
Av_cons_ind_coll
Ivexitsimple
IVsimple_ind
Conflicts2
Nrepsh1905
violenceregion
Yield
Rdensity
Ushare
Ncowspd
38
Variable definition
Source
Number of households exited the
commune both under 1906 decree and 1910 Ministry of Internal Affairs (1908,
law per 1000 rural citizens
1915, 1910, 1912, 1914).
Number of households exited the
commune under 1906 decree only per 1000 Ministry of Internal Affairs (1908,
rural citizens
1915, 1910, 1912, 1914).
Number of households exited the
commune under 1910 law (in communes
without actual repartitions) per 1000 rural Ministry of Internal Affairs (1908,
citizens
1915, 1910, 1912, 1914).
Chief administration of agriculture
Number of households consolidated land and land engineering (1908, 1909,
per 1000 rural citizens (total)
1910, 1911, 1912, 1913, 1914).
Chief administration of agriculture
Number of households consolidated land and land engineering (1908, 1909,
individually per 1000 rural citizens
1910, 1911, 1912, 1913, 1914).
Chief administration of agriculture
Number of households consolidated land in and land engineering (1908, 1909,
groups per 1000 rural citizens
1910, 1911, 1912, 1913, 1914).
Chief administration of agriculture
Average size of individually consolidated and land engineering (1908, 1909,
plot
1910, 1911, 1912, 1913, 1914).
Chief administration of agriculture
and land engineering (1908, 1909,
Average size of plot consolidated in groups 1910, 1911, 1912, 1913, 1914).
Exit confirmation rate (ratio of exits
confirmed by local courts to all applications Ministry of Internal Affairs (1908,
to exit)
1915, 1910, 1912, 1914).
Consolidation implementation rate (ratio of Chief administration of agriculture
all implemented consolidations to all and land engineering (1908, 1909,
applications to consolidated)
1910, 1911, 1912, 1913, 1914).
Conflicts index (share of applications to
exit who failed to get an agreement with the
commune and had to appeal to local officials
to organize their exits
Dubrovskij (1963)
Share of repartition communes without
actual repartitions since emancipation
Dubrovskij (1963)
Grave and Dubrovskij (1926);
Number of outright violent conflicts
Dubrovskij (1956, 1963)
Grain yield tons per hectar, calculated as
total grain yield divided by total area under
grain crops (desyatinas are transformed into Central Statistical Committee of
hectares at 1.0925 rate; puds are transformed the Ministry of Interior (1902, 1903,
into kg at 16.38 rate)
1905-1916)
Rural population per 1 sqkilometeron
January 1, corresponding year (sqversta are Central Statistical Committee of
transformed into sq kilometers at 1.138 rate) the Ministry of Interior (1905-1916)
Central Statistical Committee of
Urban share in a province
the Ministry of Interior (1905-1916)
Number of horses per hectare of crops, Central Statistical Committee of
calculated as total number of horses divided the Ministry of Interior (1905-1916)
Nhorsespd
Agrmachpd
Rwage
Hhsales
Crcoopd
Migrants
39
by total area under grain crops (desyatinas are
transformed into hectares at 1.0925 rate)
Number of cows per hectare of crops,
calculated as total number of horses divided
by total area under grain crops (desyatinas are Central Statistical Committee of
transformed into hectares at 1.0925 rate)
the Ministry of Interior (1905-1916)
Agricultural machines in tons per hectare
of crops, calculated as total number of horses
divided by total area under grain crops
(desyatinas are transformed into hectares at
1.0925 rate; puds are transformed into kg at
16.38 rate)
Davydov (2010)
Daily earnings of rural workers in harvest Ministry of agriculture (1906season
1914)
Number of land sales by peasants
Ministry of Justice (1907-1915)
Amount of small credit loans per 1000 Department of small credit (1905hectares
1915)
Number of migrant families passed
through Syzran and Chelyabinsk registration
centers per 1000 rural citizens
Turchaninov N. (1910, 1915)