CHAPTER 5
Panel Data Approaches to Estimating the Impact of Access to Public Infrastructure and Extension: Methodology and Results
he two rounds of NLSS household survey data have a panel component of 962 households, of which 784 are rural households. Therefore, we can use this panel dataset to estimate the impact of public infrastruc- ture on consumption growth, an approach similar to that used by Dercon et al. (2009). Our basic model builds on the growth literature to estimate the effect of access to public infrastructure and services on consumption growth, pov- erty status, and agricultural income growth, controlling for initial household endowments and economic shocks using the following specification:
T ln(Yt) – ln(Yt–1) = δln(Yt–1) + αKt–1 + βXt–1 + γSt–1 + ε, (5.1) where ln is an abbreviation for log normal; t stands for 2002/03 and t–1 refers
to 1995/96; Yt is defined as per capita consumption or agricultural income in 2003/04; Kt–1 includes a set of public infrastructure variables in 1995/96— travel time in logarithm, access to irrigation, and having received visits from
extension agents; Xt–1 reflects lagged fixed characteristics of the household and the district, such as education level of the household head, the number
of working men, the number of working women, landholdings, and whether the household is headed by a female, as well as district characteristics such as the cumulative conflict variable that represents the number of killings in each district, the population size of the district in 1996, and the percentage of persons of Brahman ethnicity in the district, which potentially controls for
political influence; St–1 represents transitory shocks, such as rainfall and price changes; and ε is an error term. Estimating Equation 5.1 using ordinary linear square regressions will gen- erate biased results because the lagged per capita consumption variable is correlated with the error term. We address this issue by using a generalized method of moments (GMM) estimator. Following Dercon et al. (2009), we instru-
25
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