Land, tenancy and sharecropping

Land is the main asset of rural households, and many of the poorest farm land they do not own. There are four basic arrangements for a plot with one owner and one cultivator.

• Owner cultivation: the owner works the land and keeps the output and bears the risk.
• Fixed-rent tenancy: the tenant pays a fixed sum, in money or in grain, and keeps the rest. All risk is borne by the tenant (JNU 2018), and the tenant has full incentive, because the whole of any extra output is his.
• Sharecropping: the tenant pays a fixed share of the crop. Risk is shared between landlord and tenant, and so is the extra output from more effort.
• Wage labour: the landlord hires a worker for a wage and keeps the whole crop. The landlord bears the risk, and the worker has no direct stake in the output.

The four sit on a line from full incentive and full risk (the tenant at fixed rent) to no incentive and no risk (the wage worker), with sharecropping in the middle. The rest of the lesson is about why people choose the middle.

Alfred Marshall (Principles of Economics, 1890, Book VI, chapter X) argued that a tenant who hands over half of what he grows will stop working when the value of the extra output to himself, half of it, equals the cost of the effort. A tenant who kept it all would work until the whole of the extra output equalled the cost. So share tenancy under-supplies effort: it is a tax on the tenant’s marginal product. In the model of this lesson, output is the tenant’s effort e plus a shock from the weather, and effort costs ke²/2. With a share α, the tenant picks effort so that his share of the extra output, α, equals the marginal cost ke, so

At a half share, effort is half the efficient level. The evidence has been mixed: Shaban (1987) compared plots in Indian villages and found lower yields and input use on plots of the same owner that were sharecropped than on those he cultivated himself, which supports Marshall; other studies find little difference.

Steven Cheung (1968, 1969) replied that the argument assumes the contract fixes only the share. A landlord can also write into the contract the inputs the tenant must use and the crops he must grow, and can end the tenancy if the tenant shirks; where these can be enforced at low cost, a share contract gives the same effort as a fixed rent, and the landlord, who earns a share of the output, has an interest in doing so. On this view the choice among contracts reflects the cost of enforcing terms and the wish to share risk, and the observed contracts are not inefficient given those costs.

Joseph Stiglitz (1974) put risk into the account. If the tenant could carry risk at no cost, the landlord would charge a fixed rent and the efficient effort would follow. But a poor tenant has no savings to fall back on, and a failed crop is a disaster, so he is risk averse; the landlord, who has other income, is less so. A fixed rent therefore puts the risk on the party least able to bear it, and a share contract moves part of it to the landlord, at the price of weaker incentives. The right share balances the two costs: the loss from under-supplied effort (which is bigger when the tenant’s share is low) and the cost of risk (which is bigger when his share is high).

Write the total surplus, output less the cost of effort and less the cost of risk, as a function of the tenant’s share α. With risk aversion ρ and variance of the shock σ², the cost of risk to the tenant is ρσ²α²/2, and the surplus is maximised at

The best share is below one whenever there is any risk and risk aversion, and falls as the tenant becomes more risk averse, the weather more variable or effort more costly to give. If the tenant does not mind risk (ρ = 0) the best share is one: a fixed rent. In the calculator’s base case (k = 1, σ² = 0.5, ρ = 2), the best share is one half, and the surplus is 0.25 at that share and zero at a fixed rent, since at a fixed rent the tenant works at the efficient level but the cost of risk absorbs all of the gain. Move the sliders and see which way the best share goes.

In village economies the same two people often trade in several markets at once. A landlord who rents out land also lends to the tenant, hires him as labour in the busy season and buys his surplus crop. These interlinked (interlocked) contracts are the subject of a literature that began with Amit Bhaduri (1973), who argued that landlords lent to tenants at a high interest rate because the debt kept the tenants tied to the land, and so had a reason to resist improvements that would make the tenants better off, a condition he called semi-feudalism. Pranab Bardhan (1980) surveyed the issues: interlinking lets a landlord use one contract to enforce another, to screen tenants, to get information about their effort, and to extract more of the surplus than separate markets would allow.

An example relevant to the exam: tenant farmers often cannot obtain crop loans from banks because the tenancy is oral: with no written record of the lease, the bank has no proof that the borrower is the cultivator, and no collateral (UOH 2017). A tenant therefore turns to the landlord or a moneylender, who can monitor him more easily, which is how credit and tenancy become linked.

Land reform changes who owns or controls land. After independence India tried four kinds:

• Abolition of intermediaries (zamindars, jagirdars, inamdars), in the 1950s, which put the actual cultivators in direct relation with the state.
• Tenancy reform: regulation of rents, security of tenure and, in some states, ownership for tenants.
• Ceilings on holdings, with the surplus land to be distributed to the landless. The laws largely failed: little land was declared surplus (UOH 2017), by the usual accounts because owners had divided their holdings among relatives before the laws came into force and because of exemptions.
• Consolidation of fragmented holdings.

The evidence differs by type. Besley and Burgess (2000), using state laws from 1958 to 1992, found that land reform legislation was associated with lower rural poverty, mainly through tenancy reform and the abolition of intermediaries, and not through ceilings. Banerjee and Iyer (2005) found that districts where the British gave landlords the right to collect revenue had lower farm investment and yields after independence than districts where the right lay with cultivators. West Bengal’s Operation Barga, from 1977, registered sharecroppers and gave them security against eviction; Banerjee, Gertler and Ghatak (2002) attribute a large part of the rise in the state’s farm productivity to it. Outside India, the successful reforms were those carried out by strong governments after the Second World War in Japan, South Korea and Taiwan, in which landlords were paid little in real terms (UOH 2011).

Questions on tenancy contracts, land reform and rural credit for tenants. Agricultural credit more generally is in Lesson 13.