19
are two rates of compensation set for the entire sample: a higher level for samples in Sichuan Province,
which is located in the Yangtze River basin, and a lower one for households in Gansu and Shaanxi
Provinces, which are located in the Yellow River basin. We therefore include an interaction term
between a dummy variable for the Yangtze River basin rate and a year dummy variable for 2004.
Given the preceding considerations, we estimate the empirical model as
(-ι)i,t) = μ+θhime + tiD(i,l) + αθCu) +pLe(tθ) + ≡Cu) (1)
where t indicates time, which equals zero for the preprogram period and one for the postprogram period.
The coefficient α (from the DID estimator) is the parameter of interest. Because we have both
household and individual data, we estimate equation (1) at both the household and the individual level.
Since errors in the equation that uses individual data may be correlated within households, we report
model results that account for clustered errors at the household level. We also extend the DID
framework to test whether the intensity of participation in the program influences the program effect by
replacing the treatment variable D(i, t) with measures of intensity.
Strategy to Estimate How Liquid Assets Affect the Program’s Impact on Off-Farm Labor
Two of our variables that can be used as measures of liquidity (K, B) also depict different
trends between the participating and nonparticipating groups. Since we are specifically interested in
whether the program’s effect on labor allocation differs for households with different levels of liquidity,
we turn now to the strategy for testing this.14 Ideally, if we could directly classify households into those
that are liquidity-constrained and those that are not (e.g., Carter and Olinto, 2003), we could estimate the
program’s impact for each group and test whether there are statistically detectable differences between
More intriguing information
1. The name is absent2. Optimal Taxation of Capital Income in Models with Endogenous Fertility
3. The name is absent
4. SAEA EDITOR'S REPORT, FEBRUARY 1988
5. The name is absent
6. The mental map of Dutch entrepreneurs. Changes in the subjective rating of locations in the Netherlands, 1983-1993-2003
7. A Study of Adult 'Non-Singers' In Newfoundland
8. Publication of Foreign Exchange Statistics by the Central Bank of Chile
9. Cross-Country Evidence on the Link between the Level of Infrastructure and Capital Inflows
10. Does South Africa Have the Potential and Capacity to Grow at 7 Per Cent?: A Labour Market Perspective