Table 5.
Placebo test
| (1) | (2) | (3) | (4) | (5) | (6) | |
|---|---|---|---|---|---|---|
| Discontinuity | 56 | 57 | 58 | 62 | 63 | 64 |
| Panel A (Dependent variable: consciously controlling sugar intake) | ||||||
| Pension | −4.962 | −0.978 | −0.416 | 0.439 | 0.056 | 6.246 |
| (20.453) | (1.672) | (1.465) | (0.357) | (0.584) | (27.588) | |
| Bandwidth | 2.898 | 3.818 | 4.783 | 3.460 | 4.577 | 3.898 |
| N | 529;797 | 773; 863 | 1047;887 | 441;665 | 608;822 | 470;650 |
| Panel B (Dependent variable: consciously controlling salt intake) | ||||||
| Pension | 0.675 | −2.320 | −1.412 | 0.198 | 0.270 | 8.830 |
| (6.096) | (2.342) | (1.214) | (0.327) | (0.612) | (36.632) | |
| Bandwidth | 2.992 | 4.110 | 3.168 | 3.336 | 4.566 | 3.967 |
| N | 529;797 | 995;994 | 779;720 | 441;665 | 608;822 | 470;650 |
The robust standard error of clustering to the township level is reported in parentheses, with * p < 0.1, * * p < 0.05, * * * p < 0.01. The default minimum mean square error bandwidth is used in the table