Course Introduction and Overview

• Land markets lecture covering two broad themes:
  • Allocation of land and moral hazard/sharecropping issues
  • Property rights and investment decisions (covered Monday)
• Focus on basic theory first, then applications

Historical Context of Sharecropping

• Very old subject, studied by economists like Smith and Marshall
• Sharecropping is ubiquitous - laborers pay owners share of output
• Recognized inefficiency: workers don't get full returns to labor
• Basic issue: reduced work incentives when sharing output with landlord

Basic Sharecropping Model

• Output: f(I) where I is input
• Tenant pays cost C for labor/inputs
• Owner receives share α of output
• Tenant's first order condition: marginal product = marginal cost/(1-α)
• Key insight: distortion from taxing output, not profits
• If taxed profits instead of output, no distortion would occur
• Problem: output more observable than profits (especially for self-employed farmers)

Rental Contract Solution

• Optimal solution: tenant rents land for fixed rent r, keeps all output
• Tenant maximizes F(I) - C(I) - r
• Fixed rent r doesn't affect input decisions (lump sum)
• Achieves first best efficiency
• Question: why doesn't this happen in practice?

Reasons Rental Contracts Don't Emerge

Risk Aversion and Insurance (Stiglitz Model)

• Trade-off between incentives and risk sharing
• Rental contract: tenant bears all risk, landlord gets fixed fee
• Typically poor tenant vs. rich landlord
• Optimal risk sharing requires both parties to share some risk
• Results in sharecropping-type contracts for insurance purposes

Detailed Risk Aversion Model

• Two-state model: high output (H) and low output (0)
• Effort e determines probability of high state
• Cost of effort: ½ce²
• Landlord sets payments h (high state) and l (low state)
• First best effort: e = H/C
• Rental contract requires h - l = H for first best
• With risk aversion: utility less than certainty equivalent
• Optimal contract involves some insurance (h - l < H)
• Second-order efficiency loss vs. first-order utility gain from insurance

Limited Liability Model

• Constraint: l ≥ 0 (can't force negative payments)
• When first best requires l < 0, constraint binds
• Optimal solution: l = 0, h = H/2
• Results in effort = H/2C (half of first best)
• Generates 50-50 sharecropping contract naturally
• May give tenant rents if utility exceeds outside option

Land Reform Implications

• Risk aversion model: sharecropping might re-emerge after land reform
  • Tenants still want insurance, might sell back to landlords
• Limited liability model: land reform solves problem
  • No need to pay rent, so limited liability constraint disappears
• Different models have very different policy implications

Empirical Challenges

• Difficult to test sharecropping theory with observational data
• Problem: farmers choosing different contracts may differ systematically
• Land quality differences between owned vs. sharecropped plots
• Potential solution: farmers with multiple plots (some owned, some sharecropped)
• Still challenges with unobservable plot characteristics

Bandiera et al. Uganda Experiment

Experimental Design

• Randomized controlled trial with tenant farmers in Uganda
• Three treatments:
  • Baseline: keep 50% of output
  • Treatment 1: keep 75% of output
  • Treatment 2: keep 50% of output plus exogenous cash transfer (25% of expected output)
  • Treatment 3: keep 50% of output plus risky cash transfer

Key Results

• Strong response to 75% vs. 50% contract
• Minimal response to cash transfer treatments
• Confirms incentive effects rather than pure income effects
• Effects on:
  • Intensive margin of capital usage
  • Unpaid family labor
  • Crop choice (shift from safer beans to riskier maize/tomatoes)

Methodological Issues

• Randomization inference used due to small sample
• External validity questions (unusual setting with women's groups)
• Risk treatment not perfectly analogous to production risk
• Crop choice effects puzzling (most observable/contractible margin)

Policy Implications Discussion

• Paper establishes efficiency consequences of sharecropping
• Doesn't resolve why sharecropping emerges endogenously
• Different underlying causes (risk vs. limited liability) suggest different solutions:
  • If risk aversion: agricultural insurance based on weather/exogenous shocks
  • If limited liability: transfers or debt contracts
• Challenge: basis risk in weather insurance (imperfect correlation with actual losses)