From Banerjee and Iyer (2005) History, Institutions, and Economic Performance: The Legacy of Colonial Land Tenure Systems in India American Economic Review, Vol. 95, No. 4 (Sep., 2005), pp. 1190-1213
Similar agenda to AJR i.e., effect of history on contemporary outcomes through inherited institutions Differs in focusing on specific institution in one specific country Land revenue systems in India Different places inherited different systems under British Colonial rule Some had revenue collected by landlords, others had direct levying of taxes on cultivators
These differences lead to large differences in outcomes today (controlling for geography) Differences in agricultural yields Agricultural Investments Investments in education and health Education and health outcomes
Advantages of this micro focus: Detailed history of how institutional variation arose One specific institution so can hone in on source of differences and try to identify channels exactly. The institution left with the British No land taxes today So any effects are institutional overhang not direct Suggesting extreme persistence Disadvantage: External validity Just about India?
On the channels of effect: Not exact identification here, but some clues Could have been landlord areas more unequal but no evidence of inequality differences post colonial departure Despite abolition of formal landlordism, class based antagonism might have persisted, with conflictual environment limiting possibility of collective action. This seems more consistent with the paper s findings
Historical Background British arrived as traders through English East India Company receiving permit from Mughal Emperor (Jahangir) in 1613 Empire building started after battles where they displaced leaders (some Hindu, some Mughal) in differing regions of India Winning control of Bengal and Bihar They then formally obtained revenue collection rights in these areas (presumably from previous incumbents) 1765 Then a litany of battles by the EEIC up to the 1850s, taking over large parts of the sub-continent (pretty much the lot) Mutiny of Indian troops in the EEIC forced the crown to take over in 1857 and India became a colony administered by the British Government The British left in 1947.
Land revenue or tax was the major source of revenue for all governments (pre British too) Under Mughal rule it was collected by state officials. With weakening of Mughal rule this broke down Some officials became de facto landlords In other places others grabbed this power So land revenue systems varied, and it was hard for the Colonists to really tell what the original land tenure system was.
Land revenue continued to be the most important source of revenue under British Rule (60% of revenue in 1841) A key source of revenue, heavily debated and regulated Different rules described the focus of liability for delivery of land revenues to the crown Essentially 3 different systems: Zamindari (landlord based) Raiyatwari (cultivator based) Mahalwari (village based)
Landlord areas (zamindari) Village liability for tax to British fell to landlord Landlord free to set tax on peasants and dispossess those who could not pay (or for any other reason) What was left over after paying British was for landlord to keep Revenue collecting rights bequeathable and saleable Varying procedures in place for determining amount of landlord commitment
Peasant liability areas (Raiyatwari) Revenue settlement made with individual farmer (raiyat) Survey done, and land rights determined for differing individuals (giving cultivator title) Cultivator s commitment calculated as money value of share of annual output Output estimated Varied by place, soil quality and adjust periodically for quality changes
Village liability areas (Mahalwari) Village bodies jointly owned village and responsible for cultivating and tax liability Varied from narrow body (one family) to broad (whole village) So spanned something looking like zamindari to something looking like raiyatwari Liabilities determined in ad hoc way depending on productivity and other factors (e.g.caste)
Choice of Land-tenure systems Why did the British choose different ones in different places? Premier motivations: Large and steady form of revenue Not too many disturbances But faced lack of solid information as to what would work best And often driven by theories as to what would work best that were based on abstract principles and little knowledge And grasp of local history was weak
E.g. Madras Thomas Munro argued for establishment of an individual cultivator system in Madras. He Asserted: It would raise agricultural productivity by improving incentives; Cultivators would be less subject to arbitrary expropriation than under a landlord; Cultivators would have a measure of insurance (via government revenue remissions in bad times) Government would be assured of its revenue (since small peasants are less able to resist paying their dues); This was the prevailing mode of land tenure prevailing in South India from ancient time
But Madras Board of Revenue argued for favoring landlords. Large landlords would have the capacity to invest more and therefore productivity would be higher; the peasants' long-term relationship with the landlord would result in less expropriation than the short-term one with a government official; a big landlord would provide insurance for small farmers; a steady revenue would be assured because the landlords would be wealthy and could make up an occasional shortfall from their own resources; this was the mode of tenure prevailing from ancient times
Bottom line on decision making Decisions often taken on basis of: some general principle, the ideology of the individual decision maker Monroe lost in Madras, travelled to London, convinced the board to overturn the local decision in 1820 contemporary economic doctrines the exigencies of the moment
Munro s victory was an important precedent Almost no further landlord administrated areas after 1820 (one exception) So landlord areas were mostly set up first, and then non-landlord areas after landlord systems less costly to administer and thus chosen earlier (once set up costly to change) Influence of Munro then changed things Economists and other leaders in Britain changed views French Revolution 1790s made elites likely to side with landlords In 1820s this has blown over peasant power defeated, more sympathy for utilitarians (James Mill) arguing for direct dealing with peasants NB sometimes presence of a landlord class was used as the reason, but clearly not always, and sometimes wrongly attributed
Post independence Most states with large landholders implemented land-ceiling acts. Breaking up land of large holding families Removing non-resident landlords from ownership
Why Matter Inequality could well be affected. Landlord richer and residual claimants Can squeeze peasants and appropriate productivity gains through time In non-landlord areas, taxes raised equally, landholdings smaller, therefore more equal Some evidence supportive of this
Differences in Inequality
Differences in Political power Right to collect, set and penalize gave landlords power Peasant property insecure and unlikely to be invested in due to risk of expropriation Also created class based resentments that persisted Maoist and Naxalite movements correlate (Bihar, Bengal) Possible these effect capacity of classes to work together Non landlord areas rights to work the land set with colonial administrator Land security may have created better investment incentives And less class conflict
Differences in relation to the state British got residual in non-landlord areas so more incentives for them to invest in agricultural productivity enhancing actions irrigation, railways, schools, infrastructure e.g. almost all british canals constructed in nonlandlord areas
Data District level Compare ag investments and outcomes in landlord and nonlandlord areas Controls: latitude, altitude, soil type, mean annual rainfall, coast dummy length of time under British rule early British rule particularly rapacious or because the best (or the worst) districts fell to the British first
Landlord areas seem historically and inherently better suited to ag production lower altitudes, higher rainfall, fewer areas with black soil, greater depth of topsoil, greater rainfall significantly higher population density they seem to be more fertile areas. greater proportion of minorities, "Scheduled Castes" in the Indian Constitution, more people living in rural areas. greater proportion of workforce in farming, more area to food crops like rice and wheat, less to cash crops like cotton, oilseeds, tobacco, and sugarcane
historical yields
Diff due to colonial state explanation Weak
Differences due to land distribution varying? In 1990, 64 percent of land holdings in landlord areas classified as marginal" (less than one hectare), about eight percentage points higher than in non-landlord areas. 48 percent of holdings are small to medium sized (one to ten hectares) in non-ll areas, 35 percent in landlord areas. No significant difference in the proportion of extremely large holdings, due to land ceiling laws after independence Not plausible explanation: For observed difference of 16 percent in agricultural yields to be due to more marginal holdings in landlord areas, (less productive and underinvest) requires small holding to be 12 percent as productive as larger ones. Also inconsistent with observations in all other developing countries
Differences in political environment? If political environment operates through insecurity of peasant property in landlord areas, should have observed convergence peasant property clearly became less insecure after landlords lost formal authority. So important difference in political environment may have to do with nature of collective action in the two areas.
Sates with a higher proportion of landlord areas spent less on development expenditure. Between 1960 and 1965, 13 rupees per capita on development expenditure, compared to 19 rupees in the non-landlord states. Important in explaining results. Add development expenditure per capita as an explanatory variable in our base regressions, Sharply reduces magnitude of non-landlord coefficient for the measures of HYV adoption (Table 7, column 2). State policy priorities as major channel of influence Estimate the investment and yield equations after including a fixed effect for each state. Reduces coefficient on the non landlord share substantially (by 50 percent or so), also reflected in difference between landlord and non-landlord areas in provision of educational and health: landlord areas had 21 percent fewer villages (15 percentage points) equipped with primary schools, while the gap in middle school and high school availability are 61 percent and 63 percent, respectively. Literacy rates are 5 percentage points higher in non-landlord areas, infant mortality rates 40 percent lower
Political priorities seem to differ in two areas Masses in LL areas may perceive their interests as in conflict with the elite. Whereas in non-ll perceive that all will benefit from development, may work together More conflict in LL areas higher murder, rape, kidnap, armed robbery, riots
Big Picture Conclusion In this context, it has not just been about letting the market work, and prices indicating where resources need to go to create development Here, and perhaps in many developing country contexts, many problems in getting capitalist development off the ground. Need for some sort of centralized (or collective) action. This is, of course, difficult to do when government are corrupt, unresponsive etc But India is a democracy (corruption is tolerated to some extent) ineffective government is too. This paper suggests one reason why there might be such degree of ineffectiveness Where the history of class conflict is greater, the government does less, poverty alleviation and development indicators are worse.