Table 5.
Alcohol | Tobacco | Sweets | Soft drinks | |
---|---|---|---|---|
Panel A. Expenditures
| ||||
Cross-section | 1.337 | 0.044 | 2.100 | −15.433*** |
(1.570) | (0.706) | (3.206) | (4.142) | |
[30,246] | [30,246] | [30,246] | [30,246] | |
Panel | 8.568* | 0.021 | 23.916*** | −17.962* |
(4.458) | (1.451) | (8.390) | (9.970) | |
[9,046] | [9,046] | [9,046] | [9,046] | |
| ||||
Panel B. Conditional expenditures
| ||||
Cross-section | 69.771** | 1.404 | 10.191** | −26.752*** |
(33.008) | (8.031) | (4.579) | (7.737) | |
[1,413] | [2,418] | [15,823] | [11,909] | |
Panel | 124.537 | −25.291 | 32.898** | −48.194** |
(225.483) | (27.263) | (14.405) | (20.996) | |
[420] | [711] | [4,691] | [3,572] | |
| ||||
Panel C. Budget shares (%), trimming propensity score at 5% and 95%
| ||||
Cross-section | 0.07* | 0.02 | 0.09 | −0.14 |
(0.04) | (0.02) | (0.08) | (0.10) | |
[17,797] | [17,797] | [17,797] | [17,797] | |
Panel | 0.17* | 0.14* | 0.56*** | −0.08 |
(0.09) | (0.08) | (0.19) | (0.26) | |
[5,477] | [5,477] | [5,477] | [5,477] | |
| ||||
Panel D. Budget shares (%), immediate effects in 2010
| ||||
Cross-section | 0.15*** | 0.03 | 0.16* | −0.05 |
(0.06) | (0.02) | (0.10) | (0.12) | |
[18,785] | [18,785] | [18,785] | [18,785] | |
Panel | 0.16 | 0.14** | 0.46*** | 0.10 |
(0.10) | (0.06) | (0.17) | (0.20) | |
[7,686] | [7,686] | [7,686] | [7,686] | |
| ||||
Panel E. Frequency of purchases per week
| ||||
Cross-section | 0.2302 | - | 0.0764** | −0.0056 |
(0.1458) | - | (0.0350) | (0.0233) | |
[1,413] | [15,823] | [11,909] | ||
Panel | 1.9361 | - | 0.1829** | −0.0263 |
(1.6272) | - | (0.0896) | (0.0604) | |
[420] | [4,691] | [3,572] | ||
| ||||
Panel F. Expenditures
| ||||
Transfers per capita | 0.0650*** | 0.0296*** | 0.0715** | 0.1227*** |
(0.0234) | (0.0108) | (0.0311) | (0.0391) | |
[30,246] | [30,246] | [30,246] | [30,246] |
Note: The cross-sectional models include the covariates listed in Table A6 and fixed effects for year, region, and community type. The panel models include a quadratic for log(expenditures) and fixed effects for year and household. Robust standard errors, clustered at the district level, are shown in parentheses. Sample sizes are shown in brackets. Significance:
p < 0.01
p < 0.05
p < 0.10.