22 CHAPTER 4


where y is the hedonic plot value in logarithm; K is a set of rural infrastructure variables, including travel time to the closest market in logarithm, access to irrigation, and whether the household received a visit from extension agents; X is a set of household characteristics; and D stands for a vector of district fixed effects.1 District fixed effects control for variations across districts, including physical characteristics such as terrain, agroecological potential, and previous public investments, as well as unobservable characteristics of the district such as political influence or public leadership. Equation 4.1 is estimated in a cross-section of data, so we omit time subscripts for simplicity. This equation is first estimated for the 1995/96 NLSS survey round and then for the 2003/04 NLSS 2 survey round. Differences in coefficients across rounds imply changes in the impact of access to rural infrastructure and extension nationally, because the survey rounds are nationally representative.


Hedonic Estimates of the Benefits of Rural Investments Table 4.1 presents the estimation results for 1995/96 and 2003/04, respec- tively, based on the hedonic approach using NLSS 1 and 2 data.2 In the first specification, using the first wave of data, we include district fixed effects. In the second specification, using the second wave of data, we include district fixed effects and a conflict variable that captures the intensity of conflict as represented by the number of conflict-related deaths over that time period. By explicitly including the conflict variable, we ensure that our results are robust to this potentially disruptive shock in public services and household welfare.


Comparing the point estimates across the two surveys illustrates the increas- ing benefits of rural roads. In the 1995/96 period, the elasticity of travel time on plot value was 0.261, implying that a 10 percent reduction in travel time would increase plot value by 2.61 percent. The results for 1995/96 are reasonably similar to those of Jacoby, although we have included extension services as an additional variable in the specification. The coefficient for the road variable greatly increased, to 0.471, in 2003/04. These large impacts were driven in part by the rapid increases in median land values, from NPR 16,000 in 1996 to NPR 50,000 in 2003. They also may overestimate the actual


1 We omit several of the plot quality variables found in Jacoby (2000) because the corresponding survey questions were not included in the 2003/04 round of the NLSS. In Table 4.1, we demon- strate that the modifications proposed earlier and the omission of the plot quality variables still yield coefficient estimates of the impact of the reduction in travel time that are quite close to


those estimated by Jacoby (2000). 2 Note that the point estimates in Table 4.1 with respect to travel time are not substantially dif- ferent from those reported in Jacoby (2000). Hence, our minor modifications to the econometric specification do not alter our estimates of the benefits.


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50