LAND OWNERSHIP INEQUALITY AND RURAL UNREST: EVIDENCE FROM THE LATIFUNDIA REGIONS OF SPAIN BEFORE THE CIVIL WAR Jordi Domènech Department of Economic History Universidad Carlos III de Madrid Calle Madrid 126 28903 GETAFE (Madrid) SPAIN Email: [email protected]Tel: (+34) 91 624 9809 ***very preliminary draft, not for quotation**** INTRODUCTION The Second Republic in Spain (1931-1936) witnessed one of the fastest and deepest processes of popular mobilization in interwar Europe. The General Workers’ Union (UGT) jumped from around 230’000 members in 1929 to more than a million members in 1932. Similarly, despite losing a large share of peasant workers to the socialists, the anarcho-syndicalist National Confederation of Labour (CNT) managed to unite about another million workers in the second half of 1931. As a result, union density jumped from its previous peak of 15 per cent of the gainfully employed in 1920 to almost 30 per cent in 1932, below the levels of mobilization of the Scandinavian countries but above the union density levels of Britain, France or Belgium in the 1920s (at a lower level of development) (union densities in Mann, 1995). In addition, strike activity also jumped to unprecedented levels, reaching in 1933 or in 1936 levels of strike 1
37
Embed
LAND OWNERSHIP INEQUALITY AND RURAL … · Web viewLAND OWNERSHIP INEQUALITY AND RURAL UNREST: EVIDENCE FROM THE LATIFUNDIA REGIONS OF SPAIN BEFORE THE CIVIL WAR Jordi Domènech Department
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
LAND OWNERSHIP INEQUALITY AND RURAL UNREST: EVIDENCE FROM THE LATIFUNDIA REGIONS OF SPAIN BEFORE THE CIVIL WAR
Jordi DomènechDepartment of Economic HistoryUniversidad Carlos III de MadridCalle Madrid 12628903 GETAFE (Madrid)SPAIN
Information on strikes and land invasions has relied mostly on newspapers
in the case of the province of Sevilla. The classic reference for Sevilla, Pérez
Cevallos (1983), although it gives the total number of conflicts, does not give the
full break down of conflicts at the provincial level. The number of strikes and land
invasions identified with the newspapers is close enough to the total number of
conflicts uncovered by Pascual Cevallos to justify relying exclusively on the
national press. Pérez Yruela (1979) and Cobo Romero (1932) give very detailed
information on towns in Córdoba and Jaén and his information is superior in
coverage to the information provided by the newspapers, with the exception of
general strikes, in which I have relied on the summaries of prefect report of
published in the newspapers to identify the towns participating in the general
strikes. Table 2 gives the breakdown of rural strikes in the provinces according to
Francisco Cobo Romero (1992), which has to be read as a a relatively “generous”
quantification of the total number of strikes (because it also uses in some cases the
socialist press). With the data set I have assembled I am able to track down about
75 per cent of the strikes reported for the period 1931 to 1934.
Table 2. Number of rural strikes in Córdoba, Jaén and Sevilla
Córdoba Jaén Sevilla Totals
1931 69 72 48 189
1932 32 110 75 217
1933 100 195 60 355
1934 53 135 32 220
1935 0 1 1 2
1936 29 19 22 70
Source: Cobo Romero, Revolución campesina y contrarrevolución franquista, p. 85.
Once this information was processed, the objective was linking the strike
and land invasion data to the characteristics of towns in order to uncover:
5
a. which workers in which towns protested more often, the distribution of
strikes.
b. A first attempt to answer why? I exploit in this paper certain institutional
characteristics of labour markets in the 1930s to isolate some of the factors
explaining protest in the 1930s.
EXPLANATIONS OF CONFLICTS AND RESEARCH DESIGN
Historians of Andalusian protest have focussed on two main sources of discontent,
which are in fact related. The most important one is harvest failure in a context in
which most landless workers did not have alternative sources of income. The
second one is the failure of the Republic to re-distribute land. In this setting,
harvest failure would obviously increase the pressure to re-distribute land to the
hungry peasants.
In his classic book on the history of conflicts in Córdoba, Juan Díaz del Moral
(the notary of Bujalance, in Córdoba) in 1928 dismissed the view that harvest
failures were a trigger of conflict and instead posited that in years of good harvests
typically raised protest. Moreover, in his view Córdoba and Andalucía was not the
agricultural and traditional backwater that several foreign journalists had depicted
when describing the events of the Mano Negra in the early 1880s or the Jerez
revolt of 1892, but rather a rapidly changing agricultural economy (Díaz del moral,
1973: 20). In his view, moreover, access to land was changing and the reliance of
large estates on labourers was changing as parts of latifundia in relatively well
functioning land markets or were rented to sharecroppers.
Following this line of reasoning, James Simpson and Juan Carmona shifted
the focus of attention from traditional accounts based on the backwardness of
Andalucía to the analysis of a well-functioning labour market with particular
characteristics (Carmona and Simpson, 2003: chapter 3). In their view, the chief
characteristic of Andalusian labour markets were the large changes in labour
demand. Demand peaked in harvest times for cereals and olives, and was generally
low outside harvest (with some exceptions). Therefore, Andalusian landless
labourers only were employed for about 180-200 days a year. In this context, large
6
employers also relied on temporary workers and in fact there was a very active
pattern of temporary labour migrations looking for work in the main harvests of
cereals, olives and vines. In Simpson and Carmona’s view (and in Díaz del Moral’s),
the main conflict was between local and temporary workers (forasteros) and
between the temporary workers and the workers permanently employed in the
large farms. The area sown grew constantly in the first three decades of the 20th
century and mechanization of harvest work in cereals was only modest, therefore
there was no surplus of workers in the Andalusian provinces. Rather, as unions
wanted to stabilize and negotiate working conditions, it was necessary to restrict
the presence of temporary workers, who undercut wages and working conditions
and broke strikes. Collective bargaining in this case basically had a re-distributive
effect away from the atomised temporary migrants to the local, organized workers.
In their view, it was the ‘exogenous’ arrival of working class ideologies that
disrrupted the normal functioning of relatively efficient and well-functioning
labour market for agricultural labourers.
The 1930s saw a further twist to the evolution of labour markets in
Andalucía. Until 1919, local workers, generally repressed by the police, had been
unable to prevent the movement of temporary migrants. It was only in the wave of
strikes of 1918 and especially 1919 that workers, especially in Córdoba, had been
able to restrict the employment of temporary migrants. The arrival of the Spanish
Socialist party to power in 1931 (in a coalition with the Centre-Left Republicans),
however, meant a very substantial increase in the regulation of labour markets,
and especially of rural labour markets. Firstly, the law of municipal boundaries (ley
de términos municipales), finally derogated in May 1934, forbade the recruitment
of transient migrants and workers from other towns if there were local workers
unemployed. Second, the worker employment law of October 1931 (ley de
ocupación obrera) meant that the local conciliation board -dominated by the
unions and the mayor- organized the list of employable workers and established
who was going to get the next available job (the “turno”). Although both laws
certainly were not fully enforced, the qualitative evidence suggests strongly that
enforcement was high enough to guarantee a virtual closed shop in the main rural
labour markets. This was also the case in anarcho-syndicalist holds, where the
7
unions did not support collective bargaining but were staunch defenders of the
municipal boundaries law and the turno. No doubt, membership exploded. I can
think of better “selective incentives” to join (a death threat), but a closed shop
seems to be enough of an incentive (Lichbach, 1994).
This distributional impact can also be coupled with other triggers of
collective action. Recently, economists and political scientists have explicitly
formulated testable hypothesis on the origins of conflict. Just to mention a recent
one, Hidalgo, Naidu, Nichter and Richardson (2010) tried to explain land invasions
in Brazil as a response to harvest failure or income declines. In their study, they
capture “exogenous” variations in living standards by instrumenting income with
rainfall, assuming that agricultural incomes will be subject to substantial variation
caused by weather conditions. In their study, they also see that the probability of
invading is also higher, the bigger the inequality in the distribution of land
holdings. Unexpected declines in living standards and land ownership inequality
brought about higher levels of rural conflict. In Spain, as dry-farming agricultural
output was quite volatile and land ownership distribution was apparently very
unequal, it is probably worthwhile to test these two hypotheses.
The new institutional economics also offers an alternative view of rural
conflict which will be relevant to the historical context of 1930s Spain. Alston,
Libecap, and Mueller (2000) for example argue that legal inconsistencies in land
reform policies in Brazil led to an increase in violent conflict between landowners
and squatters and to the deforestation of the Amazon forest. Posing a threat to
established property rights, in this view, the land reform pushed by the Republican
authorities potentially had dramatic effects in areas with a highly skewed land
ownership distribution.
My approximation to the determinants of strikes and land invasions
therefore takes into account the insights from the historical and social science
literature on the origins of conflicts. To do so I exploit first the relatively exogenous
variation in the land ownership distribution (determined to a great extent in a
period before the explosion of rural conflict in the 1930s), which one the one hand
8
captures the effect of inequality on conflict and also could be related to the ability
of strikers to detect strikebreakers in large estates as opposed to the same ability
in scattered landholdings.
Secondly, I exploit variation in population densities, which give an
approximation to the land to labour ratio, in turn giving a further approximation to
the local labour demand that the labourers faced after the passing of the law of
municipal boundaries in 1931. In this context, the situation is closed to a quasi
experiment, since area and population (the ingredients of population density)
were determined well before the passing of the law of municipal boundaries.
Wages were determined by provincial collective contracts, negotiated every year
(Anuario Español de Política Social, 1935). To the point that wages reflected the
local labour supply and demand equilibrium of the towns with an average land to
labour ratio, adjustments in the local labour markets of towns with low land to
labour ratios will take the form of unemployment (via quantities rather than
prices). On the other hand, in the long run, population decisions are endogenous,
therefore landless workers will tend to concentrate in large towns with relatively
thick labour markets (develop).
HARVEST FAILURE
Jaén, Córdoba and Sevilla especialised in a combination of wheat and olives. In
Córdoba in 1932, 35 per cent of the agricultural land was used to grow wheat, 37
per cent olive trees, and 16 per cent was left fallow. Only 12 per cent of the land
had alternative uses. In Jaén, in the same year, 26 per cent of the agricultural land
was left fallow, 27 per cent was used to grow cereals and 42 per cent to grow olive
trees. Only 5 per cent of the land was left for alternative crops. In Sevilla, 43 per
cent was cereal, 35 per cent was olive trees and 8 per cent was fallow. 14 per cent
of the land was left for alternative uses (all data from the Anuario Estadístico,
1932-1933). The vulnerability of these provinces to crop failure is obvious.
I first look at aggregate trends in production of cereals and olives (Carreras
and Tafunell, 2005). In both cases, output trends up with substantial fluctuations,
which are especially acute in the olive crops (in this case proxied by the production
9
of olive oil). Graph 1 gives the trends in total production since 1890 (Jaén and
Córdoba were the main producers of olive oil by a large margin).
Graph 2. Olive oil and wheat production (index numbers, 1890=100)01002003004005001890189519001905191019151920192519301935yearindexwheatindexoliveoilIn this case, aggregate olive output is basically tracking the output produced
by Córdoba and Jaén, and to a lesser extent Sevilla. 1926, 1928 and 1930 were
cases of substantial crop failure, whereas the 1933 crop was poor but not
dramatically poor. There were two spectacular spikes in 1927 and 1928. When I
look at the average production of olives per province between 1922 and 1934
(when the Spanish statistical yearbooks gave information on production at the
provincial level), only two years in each province appear to be one standard
deviation below the mean (the calculation takes into account the greater amount of
land used to grow olives). For both Sevilla and Córdoba, between 1922 and 1934,
1928 and 1930 were the only two years in which the harvest of olives was one
standard deviation below the mean (again, calculation takes into account that
more land was available to grow olives), for Jaén, 1930 and 1933 were one
standard deviation below the mean.
Similar inspection of data on cereal harvests shows that for our three
provinces, 1931 and 1933 were particularly bad years, while 1932 and 1934 were
delivered extraordinary harvests. The reason seems to be largely exogenous. The
10
spring temperatures in Andalucía in 1931 and 1933 were particularly high,
compared to the relatively cold spring temperatures of 1932 and 1934, and high
temperatures in Spring have a negative impact on cereal output (temperatures
from Anuario Estadístico 1931, 1932-1933, 1934; Rodríguez,-Puebla, Encinas,
Frías, 2004a,2004b on effects of winter rain and spring temperatures on output).
Also taking into account the amount of land put into production, the harvest of
1931 in Sevilla was well above the mean, while the 1933 harvest was about one
standard deviation below the mean (one could expect a harvest as bad as that one
once in every six years). In Jaén, the 1931 and the 1933 harvests were less than
one standard deviation below the mean. Only in Córdoba was the 1933 harvest
catastrophic, less than 2 standard deviations below the mean, while the 1931
harvest was one standard deviation above the mean. In all three cases, the 1932
harvest was more than 2 standard deviations above the mean.
To what extent do we see largely exogenous crop failures affecting patterns
of conflict? There is no doubt that crop failures must have had serious effects on
the living standards of the workers dependent on paid employment by causing a
collapse in the demand for labour. However, it is far more difficult to link periods
of crop failure to the explosion of rural unrest of the 1930s. For cereals, harvests in
the 1920s were pretty stable and only 1931 and 1933 seem to have been
particularly poor years. In 1933, Córdoba had certainly a very poor harvest, but
strikes and conflict increased also in both Jaén and Sevilla in the same year. On the
other hand, 1932 was an exceptional year, and yet the number of strikes grew from
the previous 1931 level. In 1934, a year of an exceptionally good cereal harvest, the
National Federation of Peasants organized the most serious general strike of its
history. In Jaén, more dependent on olives than the other two provinces,
experienced increases in strikes both in 1932 (a year of a relatively good harvest)
and in 1933 (a year of poor harvest).
LAND INEQUALITY AND POPULATION DENSITIES,
Despite the homogeneous image of being characterized as a latifundia region, the
truth was that the distributions of land ownership varied widely in the three
provinces studied here. Andalucía in this sense was no different than other
11
Southern European regions characterized by dry farming. For example, Julian Pitt
Rivers noted “The agricultural land of Andalucía has been held in latifundia, ever
since the Roman times. The complex of dry-farming, with its uncertain yearly
returns and fluctuating demand for manpower, may be seen through the history of
the Mediterranean to have tended to encourage the agglomeration of property, and
also of habitations, into large units. The vast expanse cultivated by short-term,
town-dwelling labour for a single master is common not only in Betica, la Mancha
and Castile, but to much of Italy as well. This type of landholding is typical only of
the plains, and it is rarely found today among the mountains. The soil of the sierra
tends to be poorer than that of the plain and is more labourious to cultivate and
more subject to erosion. In any case, its unevenness makes it unsuitable for large
scale exploitation. (Pitt Rivers, 1954: 39; see as well Braudel, La Méditerranée, vol.
I: 82).”
Data on the distribution of land ownership at the municipal level are taken
from Pascual Carrión’s classic Los latifundios en España (first published in 1932),
which provides information at the municipal level on the fraction of total area
owned by estates of more than 250 hectares (2,5 million square meters or about
330 football pitches). In graph 3 I show the histogram of the distribution of land
ownership for 198 towns (of a total of 272 towns in the three provinces excluding
the cities over 40000 and the capitals of the province). In many towns, the level of
land ownership concentration was actually low.
Graph 3. Historigram, proportion of town are owned by estate of more than 250
hectares (N=198 towns in Jaén, Córdoba, Sevilla).0.005.01.015.02Density0102030405060708090100landinequality12
Landinequality was higher in the fertile plains than in the hills. Indeed, our
crude measure capturing the presence of large estates is negatively correlated with
altitude. Moreover, altitude is also negatively correlated with the presence of
landless labourers in the total population (data from the Censo de Campesinos,
which are clustered at the judicial district level rather than given for each town).
Finally, land ownership inequality is mildly negatively correlated with population.
Table 3 gives the correlations and graph 4 the relationship between ownership
inequality and population.
Table 3. Correlation matrix of of inequality and geographical and demographic
Graph 4. Land ownership inequality and population. Lowess smoothing.020406080100landinequality050001000015000200002500030000pop1930bandwidth = .8Lowess smootherMoreover, I look at the distribution of labourers, small owners and
sharecroppers among the Andalucia pueblos, using the Peasant Census (Censo de
13
Campesinos) of 1933. The data are aggregated at the level of judicial district and it
is unclear what was the level at which a landowner was considered a “small
proprietor”. Graphs 5, 6 and 7 give the histograms. Again the picture is one of more
heterogeneity than traditionally assumed, everywhere labourers were far more
prevalent than either small proprietors or sharecroppers, but Córdoba and Jaén
had a higher share of small landowners.
Graph 5. Histogram with the fraction of labourers (fraction of all “poor” peasants,
not of total population in each district).0.05.1.15.2Fraction.5.6.7.8.9fraction labourers
Graph 6. Histogram of fraction of sharecroppers 0.1.2.3Fraction0.1.2.3fraction sharecroppers
14
Graph 7. Histogram fraction of small owners.0.05.1.15.2Fraction0.1.2.3.4fraction small owners
In table 5 I present the results from the estimation:
Table 5. the determinants of strikes, 1931-1934
variables Zero inflated
poisson
All towns
Zero inflated
poisson,
Pop1930
excluded
Zero inflated
poisson,
Towns below
400 metres
Zero inflated
poisson, small
towns only
(below sample
mean)
Constant -2.23***
(0.645)
1.39***
(0.25)
-2.03***
(0.97)
1.66***
(0.39)
Ln (land in) 0.03
(0.07)
0.008
(0.065)
0.09
(0.12)
-0.07
(0.101)
Ln (popden) 0.09
(0.07)
0.22**
(0.07)
0.1
(0.1)
0.13
(0.11)
Ln (pop 1930) 0.39***
(0.06)
0.39***
(0.09)
Alt250500 -0.112
(0.13)
-0.18
(0.133)
-0.09
(0.16)
-0.63***
(0.21)
Altover500 -0.47***
(0.134)
-0.47***
(0.14)
-0.84***
(0.21)
Jaen 0.044
(0.12)
-0.03
(0.114)
0.07
(0.21)
0.07
(0.17)
Sevilla -0.73***
(0.134)
-0.8***
(0.134)
-0.86***
(0.16)
-1.04***
(0.2)
N 201 201 88 125
LR chi2 90.59 52.97 48.76 38.46
Prob>chi2 0.0000 0.0000 0.0000 0.0000
The results show how land inequality was irrelevant in the explanation of
strikes. The main predictor of strikes are the population of the town (remember
we are also correcting for the potential effect of population on the reporting of
strikes in newspapers), the altitude and the province (with Sevilla towns
18
organizing significantly less strikes). The negative coefficients on altitude and
Sevilla remain when I exclude the population variable or look only at sub-samples
of towns in my original data set. There are several potential reasons why
population might matter to explain the higher incidence of strikes. However, since
labourers tended to concentrate in large towns, the population effect, among other
potential causes, also reflects the insider power of the local labourers. On the other
hand, because large towns also had higher population densities part of this size
effect also is channelled through the shocks to local demand caused by the
municipal boundaries law.
In order to teasa out the various effects it is useful to consider only those
strikes that were organized demanding work. In this second exercise therefore I
only concentrate on the strikes organized by unemployed workers. Table 6 reports
the results from the regressions:
Table 6. Determinants of regressions demanding work.
Variables Zero inflated
Poisson regression
Zero inflated
Poisson regression
Lpop1930
excluded
Zero inflated
Poisson
regression, towns
below population
average
Constant -3.03***
(1.84)
-1.13***
(0.65)
-1.24
(1.04)
Ln (land in) 0.028
(0.15)
0.08
(0.144)
0.21
(0.24)
Ln (pop density) 0.36
(0.25)
0.46**
(0.23)
0.6**
(0.36)
Ln (pop1930) 0.214
(0.193)
Alt250500 -0.203
(0.4)
-0.32
(0.39)
-0.04
(0.7)
Altover 500 -1.46***
(0.68)
-0.9
(0.39)
-0.8
(0.7)
Jaén 2.35*** 2.34*** 1.79***
19
(0.37) (0.38) (0.56)
Sevilla -1.46***
(0.678)
-1.54***
(0.68)
-1.7***
(0.86)
N 201 201 125
LR chi2 77.14 75.89 30.24
Prob >chi2 0.0000 0.0000 0.0000
In this case, the effect of population and altitutde still persist and Sevilla still
organizes far less strikes than the other two provinces. Jaén on the other hand
organized far more strikes of unemployed workers than any of the provinces. The
reason probably has to do with the harvest failure of olives in 1933. In this
regressions, the effect of population densities and local demand shocks is now an
important determinant of this type of strikes (although the predominance of Jaén
dwarfs any kind of other effect).
LAND INVASIONS
My second approximation to the levels of conflict uses cases of land invasions.
When looking at the qualititative evidence, it is clear land invasions in Andalucía
did not represent a threat to established property rights. Rather than squatting,
most invasions of property reflected the fact that workers started working in the
fields of some landowner or entered a large estate to pick up olives after harvest
(the so called “rebusca” done by old people and children) or to pick up acorns. To a
great extent, illegally entering the fields of a landowner to work responded to the
widespread use of the law of compulsory cultivation, a law passed by the
republican government forcing landowners to employ unemployed workers under
certain circumstances (when a commission could prove the property was not
being worked at the highest level of efficiency, but in fact there were surely many
cases of arbitrarily decided compulsory cultivation). Landowners and employers
bitterly complained about the workings of this law, which in fact left in the hands
of mayors (in many rural towns, socialist and pro-workers) and the so-called
“comisión de policía rural” (also controlled by mayors) the final decisions on the
decrees of compulsory cultivation.
20
My data on land invasions still need substantial revisions, however I do not
think the picture is going to vary systematically. Graph 9 gives the histogram of
land invasions, which gives about 80 per cent of towns not reporting a single event
of workers trespassing a property. Indeed, several contemporary observers argued
that in fact the issue of land invasions was indeed an exageration of the right-wing
press. But even the moderate El Sol often ran a section called “Invasiones de fincas”
in which some cases of true workers’ trespassing were recorded along robberies
and other crimes.
Graph 9. Histogram of land invasions.0.2.4.6.8Fraction012345land invasionsAgain, I perform the same count model to understand the determinants of
invasions. Neither inequality or population density explain the results and as in the
other exercises population and the provincial dummies swallow all the
explanatory power. Quite surprisingly, land invasions did not decline in the hilly
areas, something not altogether surprising considering the type of property
trespassing and its reasons that we are considering. Results are presented in table
7:
21
Table 7. Determinants of land invasions.
Variables Zero inflated
Poisson
Zero inflated
Poisson
Constant -5,93***
(2,13)
0,0145
(0,86)
Ln (land in) -0,11
(0,22)
0,011
(0,235)
Ln (popdensity) -0,33
(0,27)
-0,361
(0,29)
Ln(pop1930) 0,612***
(0,21)
Alt250500 -0,552
(0,367)
-0,42
(0,39)
Altover500 0,006
(0,382)
-0,028
(0,4)
Jaen -0,605
(0,44)
-0,804*
(0,454)
Sevilla 0,98***
(0,37)
0,72**
(0,4)
N 201 201
Chi2 29,66 20,01
Prob>chi2 0,0000 0,000
As in the case of strikes, and although I have to do some more work in my
data set of land invasions, the main determinants of invasions was again the
population of town. In this case, most probably most cases of property trespassing
were caused by workers anticipating a favourable concession from the mayor
forcing some landowner to employ unemployed workers. Provincial dummies
again seem to be crucial in determining land invasions, as workers in Sevilla were
clearly more prone to invade properties than in other provinces. Probably, one will
need to take a look at local politics to understand this effect.
22
CONCLUSIONS
The picture that emerges from my study of rural conflict in Andalucía challenges
the view we have of the determinants of conflict. Firstly, conflict was more
concentrated than otherwise thought into few localities that concentrated a large
number of labourers. Conflict in this sense favoured local insider power at the
expense of workers in other towns, and therefore could not be generated by
widespread harvest failure. When I look at the determinants of strikes demanding
work, I show how probably shocks to the local demand of labour caused by the law
of municipal boundaries were important, although harvest failure in olives in Jaén
also increased the strikes of unemployed workers in winter. My first preliminary
results on land invasions suggests this was a phenomenon that also required local,
union power and favoured the local, permanent labourers in the large towns of the
plains.
Bibliography:
Acemoglu, Daron and James A. Robinson, Economic origins of dictatorship and democracy (Cambridge, 2005).
Alston, Lee J., Gary D. Libecap and Bernardo Mueller, “Land reform policies, the sources of violent conflict, and implications for deforestation in the Brazilian Amazon,” Journal of Environmental Economics and Management, 39 (2000), pp. 162-188.
André, Catherine and Jean Philippe Plateau, “Land relations under unbearable stress: Rwanda caught in the Malthusian trap,” Journal of Economic Behavior and Organization, 34 (1998), pp. 1-47.
Angrist, Joshua D. and Adriana D. Kluger, “Rural windfall or a new resource curse? Coca, income, and civil conflict in Colombia,” Review of Economics and Statistics, XC, 2 (2008), pp. 192-214.
Beck, E. M. and Stewart E. Tolnay, “Analyzing historical count data: Poisson and negative binomial regression models,” Historical Methods, 28, 3 (1995), pp. 125-131.
Berman, Eli, Joseph Felter and Jacob N. Shapiro, “Do working men rebel? Insurgency and unemployment in Iraq and the Philippines,” NBER working paper 15547 (November 2009).
Bernal, Antonio Miguel, La propiedad de la tierra y las luchas agrarias andaluzas (Barcelona, 1974).
23
Bernaldo de Quirós, Constancio, “Informe acerca del paro de los jornaleros del campo de Andalucía durante el otoño de 1930,” in Estudios y documentos: la crisis agraria andaluza de 1930-31 (Madrid, 1931 [1973]).
Béteille, André, Studies in agrarian social structure (Oxford, 1974).
Braudel, Fernand, La Méditerranée, volume I.
Brenan, Gerald, The Spanish labyrinth: the social and political background of the Spanish civil war (Cambridge, 2008 [1943]).
Carmona, Juan and James Simpson, El laberinto de la agricultura española: instituciones, contratos y organización entre 1850 y 1936 (Zaragoza, 2003).
Carrión, Pascual, Los latifundios en España. Su importancia, origen, consecuencia y solución (Barcelona, 1975 [1932]).
Creamer, Cristopher, “Inequality and conflict. A review of an age-old concern,” United Nations Institute for Social Development. Identitities, Conflict and Cohesion working paper number 11 (October 2005).
Díaz del Moral, Juan, Historia de las agitaciones campesinas andaluzas (Madrid, 1973 [1929]).
Espinoza, Luis E., Ricardo Robledo, María Pilar Brel and Julio Villar, “Estructura social del campo: el Censo de Campesinos (1932-1936): primeros resultados,” in Ricardo Robledo and Santiago López (eds.), ¿Interés particular, bienestar público? Grandes patrimonios y reformas agrarias (Zaragoza, 2007), pp. 307-342.
Florencio Puntas, Antonio y Antonio López Martínez, “El trabajo asalariado en la agricultura de la Baja Andalucía. Siglos XVIII y XIX,” Historia Agraria, 21 (2000), pp. 99-126.
Florencio Puntas, Antonio y Antonio López Martínez, “Las migraciones estacionales agrarias en Andalucía anteriores al siglo XIX,” Boletín de la Asociación de Demografía Histórica, 18 (2000), pp. 71-100.
Garrido González, Luis, Historia del olivar y del aceite en Andalucía (Málaga, 2000).
Gerschenkron, Alexander, Income and democracy in Germany (New York, 1966).
Hart, Peter, “The geography of revolution in Ireland, 1917-1923,” Past and Present, 115 (1997), pp. 142-176.
24
Hidalgo, Daniel F., Suresh Naidu, Simeon Nichter, and Neal Richardson, “Economic determinants of land invasions,” Review of Economics and Statistics, 92, 3 (2010), pp. 505-523.
Hobsbawm, E. J. and George Rudé, Captain Swing (London, 1973).
Humphries, Jane, “Enclosures, common rights and women: the proletarianization of families in the late eighteenth and early nineteenth centuries,” Journal of Economic History, L, 1 (1990), pp. 17-42.
King, Gary, “Statistical models for political science event counts: bias in conventional procedures and evidence for the exponential Poisson regression model,” American Journal of Political Science, 32 (1988), pp. 338-363.
Lichbach, Mark I., “What makes rational peasants revolutionary? Dilemma, paradox, and irony in peasant collective action,” World Politics, 46, 3 (1994), pp. 383-418.
Luebbert, Gregory, Liberalism, fascism or social democracy: social classes and the political origins of regimes in interwar Europe (Oxford, 1991).
Malefakis, Edward, Agrarian reform and peasant revolution in Spain. Origins of the Civil War (New Haven, 1970).
Markoff, John, “Context and forms of rural revolt: France in 1789,” Journal of Conflict Resolution, 30, 2 (1986), pp. 253-289.
Markoff, John, “The social geography of rural revolt at the beginning of the French Revolution,” American Sociological Review, 50, 6 (1985), pp. 761-781.
Martinez Alier, Juan, Labourers and landowners in Southern Spain (London, 1971).
Montañés, Enrique, “Crisis agrícola y reformismo republicano: la instauración de la negociación colectiva en la campiña andaluza, 1931-1933,” Estudios Regionales, 49 (1997), pp. 133-160.
Moore, Barrington Jr., The social origins of dictatorship and democracy: lord and peasant in the making of the modern world (Boston, 1966).
Pascual Cevallos, Fernando, Luchas agrarias en Sevilla durante la Segunda República (Sevilla, 1983).
Pérez Yruela, Manuel, La conflictividad campesina en la provincia de Córdoba (1931-1936) (Barcelona, 1979).
Pitt-Rivers, J. A., The people of the Sierra (London, 1954).
Pulido Matos, Manuel, Historia de un pueblo andaluz: Lebrija, de la Revolución Gloriosa a la Democracia, 1868-1979 (Lebrija, ?).
25
Scott, James C., The art of not being governed: an anarchist history of upland southeast Asia (New Haven, 2009).
Sevilla Guzmán, Eduardo, La evolución del campesinado en España. Elementos para una sociología política del campesinado (Barcelona, ?).
Sindicato Vertical del Olivo, Estudios sobre la economía del aceite (Madrid, 1944).
Sindicato Vertical del Olivo, El paro estacional campesino (Madrid, 1946).
Skocpol, Theda, States and social revolutions. A comparative analysis of France, Russia and China (Cambridge, 1979).
Villarreal, Andrés, “The social ecology of rural violence: land scarcity, the organization of agricultural production, and the presence of the state,” American Journal of Sociology, 110, 2 (2004), pp. 313-48.