Are Indian Farms Too Small. Mechanization, Agency Costs, and Farm Efficiency (Figures and Tables)

Figure 1. Proportion o f Farms with Mechanized Farming Equipment, by Owned Landholdings and Survey Year 0.25 2007 1998 1970 0.2 0.15 0.1 0.05 0 0 2 4 6 8 10 12 14 16 18 20 Figure 2. Fraction of Agricul tural Labor Employed in Three Labor Regi mes, by Landhol ding Size 1 0.9 0.8 0.7 Regime 1: No hiring, family sells labor 0.6 Regime 2: Hiring labor 0.5 Regime 3: Autarchy, no hiring or selling 0.4 0.3 0.2 0.1 0 0 2 4 6 8 10 12 14 16 18 20 Fi gure 3. Measures of Per-Acre Producti vity, by O wned Landholdi ng Size (2007-8) 14000 12000 10000 Output value P rofit, lower-bo und wag e 8000 P rofit, upper-bound wage P rofit, zero-cost family labor 6000 4000 2000 0 0 2 4 6 8 10 12 14 16 18 20 -2000 Figure 4 . Loca lly-weighted FE-IV Estima tes of the Effects of Land Owned on Pro fits per Acre (and one sd Confi dence Bounds), by Landho lding Size 3000 2500 2000 1500 1000 500 0 0 2 4 6 8 10 12 14 16 18 20 Figure 5. Estimated Reservation Rental Rate, by Owned Landholding Size 16000 14000 12000 10000 8000 6000 4000 2000 0 0 2 4 6 8 10 12 14 16 18 20 Figure 6. Within-Vill age Rel ati onship Between the Probability of Leasing In and Leasing out Land, by Ownership Holdings (N=119,349) 0.025 0.02 Lease In Lease Out 0.015 0.01 0.005 0 -5 -3 -1 1 -0.005 -0.01 3 5 7 9 Table 1. Panel Data Estimates (1999-2008): Effects of the Use of Hired and Family Labor on Supervision Time (Adjusted Mandays), by Estimation Procedure Estimation procedure: OLS Farmer Fixed-Effectsa Farmer Fixed-Effects-IV Hired labor mandays, adjusted for sex/age .0906 (2.62) .0977 (3.39) .573 (2.81) Family labor mandays, adjusted for sex/age .170 (2.96) .0844 (2.41) .138 (2.10) Test statistic, H0: supervision costs= χ2(1), p-value 0.89 (.345) 0.02 (.812) 3.76 (.0525) Number of observations 3,280 3,280 3,280 Number of farmers 1,640 1,640 1,640 Kleinberger-Paap underidentification test statistic χ2(df), p-value (7) 18.6, .0095 Hansen J overidentification test statistic χ2(df), p-value (6) 7.41, .285 Absolute value of asymptotic t-ratios in parentheses. aSpecification includes state dummy variables and whether old person in the household aged 60+. Identifying variables include number of head’s brothers, head’s father’s age, whether the head’s father co-resided, the number of boys and the number of girls aged<10 and over 5, the number of boys and the number of girls aged>=10 and <15, and household size in 1999; and total inherited land by 1999. Table 2. Panel Data Estimates (1999-2008): Effects of Own Landholdings and Own Farm Equipment on Profits per Acre (Supervision Costs-adjusted), by Estimation Procedure Village Fixed-Effectsa Estimation procedure: Farmer Fixed-Effects IVb Farmer Fixed-Effects Owned landholdings 44.7 (4.47) 46.6 (4.62) 41.7 (4.02) 42.3 (1.53) 54.5 (1.91) 54.6 (1.91) 577.8 (3.15) 653.1 (2.79) 712.8 (2.63) Fraction of owned landholdings irrigated - 957.8 (3.64) 936.2 (3.55) - 1225.5 (3.11 1222.9 (3.10) - 4254.8 (2.52) 4404.2 (2.25) Value of farm equipment - - .00290 (2.05) - - .000067 (0.16) - - .0189 (1.39) Number of observations 3,967 3,967 3,967 3,967 3,967 3,967 3,524 3,524 3,524 Number of farmers 2,221 2,221 2,221 2,221 2,221 1,749 1,745 1,654 1,654 Kleinberger-Paap underidentification test statistic χ2(df), p-value (5) 11.8, .0372 (8) 19.5, .0122 (8) 15.1, .0575 Hansen J overidentification test statistic χ2(df), p-value (4) 0.63, .960 (7) 6.76, .455 (7) 10.3, .174 Absolute value of asymptotic t-ratios in parentheses. aSpecification includes year=2008 dummy; clustered t-ratios at the household level. b Instruments include land inherited after 1999, assets inherited after 1999, whether the current head has brothers, the standard deviation of the schooling of inheritance claimants, the father’s age in 1999, whether the father is co-resident in 1999, owned asset values in 1999, inherited irrigation assets in 1999. Table 3. Retrospective Panel Data Estimates (2008): Effects of Own Landholdings and Own Farm Equipment on Investment in Farm Equipment and Equipment Rental, by Estimation Procedure Dependent variable Equipment Investment Equipment Hire Expenditure Estimation procedure FE-Farmera FE-Farmer IVb FE-Farmer FE-Farmer IVb Owned landholdings 23.4 (0.08) 916.9 (2.13) 125.5 (1.94) 185.6 (2.55) Value of owned farm equipment -.0874 (1.61) -.740 (4.49) -.0187 (4.36) -.0326 (1.67) Bank within 10 Km 3267 (3.05) 4187.9 (2.68) -328.1 (1.67) -328.6 (1.66) Number of farmers 2,570 2,570 1,822 1,822 Kleinberger-Paap underidentification test statistic χ2(df), p-value (6) 17.2, .0085 (6) 12.8, .0464 Hansen J overidentification test statistic χ2(df), p-value (5) 4.64, .461 (5) 2.08, .838 Absolute value of asymptotic t-ratios in parentheses. aSpecification includes year=2008 dummy; clustered t-ratios at the village level. bInstruments include land inherited after 1999, assets inherited after 1999, whether the current head has brothers, the standard deviation of the schooling of inheritance claimants, the father’s age in 1999, whether the father is co-resident in 1999, owned asset values in 1999, and the presence of a bank within 10 km in 1999. Table 4. Within-Farmer/Season Plot-Level Estimates (2007-8): Effects of Plot Size on Plot-Specific Profits and Labor Costs and Use of Tractor Services Profits per Acre Any Tractor Services Usedb Total Labor Costs per Acre Plot area 536.3 (4.55) 534.9 (4.54) 494.8 (4.20) -610.5 (5.49) -615.2 (5.50) -568.4 (5.21) .303 (2.01) .317 (1.93) .286 (1.85) Whether plot irrigated 1564.6 (2.93) 1514.7 (2.84) 1527.1 (2.78) -105.9 (0.17) -151.9 (0.27) -26.2 (0.04) .962 (2.29) 1.01 (2.25) 1.01 (2.18) Formal equipment rental market in village*area - - 713.5 (1.73) - - -770.4 (2.09) - - .0662 (0.14) Include soil characteristics?a N Y Y N Y Y N Y Y Number of plot observations 16,544 16,544 16,544 16,544 16,544 16,544 16,544 16,544 16,544 Number of farmers 7,845 7,845 7,845 7,845 7,845 7,845 7,845 7,845 7,845 Absolute value of asymptotic t-ratios in parentheses clustered at the village level. aSoil characteristics include depth, salinity, percolation and drainage; five soil colors (red, black, grey, yellow, brown, off-white); five soil types (gravel, sandy, loam, clay, and hard clay). bML conditional logit estimates. Table 5. Within-Plot Estimates Across Three Seasons (2007-8): Effects of Previous-Period Farm-Level Profit Shocks on Plot-Level Current Profits per Acre, by Owned Landholding Size Farm size: Owned Landholdings<4 Owned Landholdings>=4, <10 Owned Landholdings>=10 Farm profits per acre, all other plots, previous season .0848 (2.10) .112 (1.45) -.0426 (0.62) Farm profits per acre, this plot, previous season -.479 (8.25) -.549 (5.01) -.540 (3.01) Fertilizer use, this plot, previous season (value per acre) 1.14 (2.15) .777 (2.11) .363 (2.72) Number of plot observations 5,802 3,060 1,964 Number of farmers 3,595 1,967 1,234 Absolute value of asymptotic t-ratios in parentheses clustered at the farm level. Specifications include season*village dummy variables.