Table S6.
Outcome variable | Model Component | Metric | NM | NC | NR | I | D | C |
Deforestation | Certification | Coef | −0.64*** | −0.27 | −0.57*** | −0.22 | −0.30 | −0.64** |
P | <0.001 | 0.11 | 0.0016 | 0.29 | 0.18 | 0.024 | ||
Precertification | Coef | 0.41*** | 0.25 | 0.39** | 0.31** | 0.37*** | 0.62*** | |
P | 0.0041 | 0.12 | 0.015 | 0.020 | 0.0018 | 0.0041 | ||
Temperature | Coef | 0.0096*** | 0.0073*** | 0.0093*** | 0.010*** | 0.0098*** | 0.0085*** | |
P | <0.001 | 0.0010 | <0.001 | <0.001 | <0.001 | <0.001 | ||
Precipitation | Coef | 1.1** | 4.2*** | 2.3*** | 3.0*** | 3.0*** | 2.9* | |
P | 0.021 | <0.001 | 0.0068 | <0.001 | 0.0036 | 0.069 | ||
Constant | Coef | −147*** | −112*** | −142*** | −161*** | −150*** | −133*** | |
P | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | ||
N, control | 1473 | 0 | 173 | 395 | 295 | 55 | ||
N, treatment | 89 | 89 | 89 | 73 | 64 | 44 | ||
N, districts (clusters) | 133 | 34 | 71 | 91 | 29 | 40 | ||
Primary Forest Clearing | Certification | Coef | −0.81*** | −0.37* | −0.76*** | −0.67*** | −0.64** | −1.1*** |
P | <0.001 | 0.069 | <0.001 | 0.0054 | 0.011 | <0.001 | ||
Precertification | Coef | 0.30 | 0.12 | 0.29 | 0.24 | 0.25 | 0.50 | |
P | 0.11 | 0.54 | 0.15 | 0.22 | 0.29 | 0.13 | ||
Temperature | Coef | 0.0099*** | 0.0092*** | 0.0097*** | 0.012*** | 0.011*** | 0.0099*** | |
P | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | 0.0024 | ||
Precipitation | Coef | 1.12 | 6.4*** | 2.7** | 2.8** | 2.9** | 2.7 | |
P | 0.11 | <0.001 | 0.038 | 0.013 | 0.041 | 0.21 | ||
Constant | Coef | −154*** | −141*** | −148*** | −184*** | −163*** | −153*** | |
P | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | 0.0019 | ||
N, control | 1236 | 0 | 148 | 295 | 225 | 48 | ||
N, treatment | 71 | 71 | 71 | 56 | 46 | 30 | ||
N, districts (clusters) | 125 | 30 | 63 | 78 | 25 | 34 | ||
Peatland Forest Clearing | Certification | Coef | −0.68*** | −0.43 | −0.67** | −0.97*** | −0.69** | −1.5*** |
P | 0.0054 | 0.15 | 0.012 | <0.001 | 0.026 | <0.001 | ||
Precertification | Coef | 0.22 | 0.17 | 0.26 | 0.47 | 0.48* | 0.98*** | |
P | 0.25 | 0.50 | 0.19 | 0.15 | 0.094 | 0.0023 | ||
Temperature | Coef | 0.0094*** | 0.0041 | 0.0070*** | 0.0086*** | 0.0075*** | 0.0090*** | |
P | <0.001 | 0.24 | <0.001 | <0.001 | <0.001 | 0.0027 | ||
Precipitation | Coef | 1.7** | 5.9*** | 2.8* | 2.5* | 2.6 | 2.8 | |
P | 0.035 | 0.010 | 0.057 | 0.052 | 0.11 | 0.41 | ||
Constant | Coef | −144*** | −66 | −110*** | −134*** | −117*** | −140*** | |
P | <0.001 | 0.20 | <0.001 | <0.001 | <0.001 | 0.0021 | ||
N, control | 597 | 0 | 70 | 132 | 112 | 14 | ||
N, treatment | 33 | 33 | 33 | 20 | 23 | 9 | ||
N, districts (clusters) | 79 | 19 | 42 | 44 | 16 | 13 | ||
Fire Rate | Certification | Coef | −0.80*** | 0.10 | −0.61*** | −0.39** | −0.38*** | −0.51* |
P | <0.001 | 0.71 | <0.001 | 0.017 | 0.010 | 0.099 | ||
Precertification | Coef | −0.21 | −0.27 | −0.21 | −0.26 | −0.22 | −0.49 | |
P | 0.12 | 0.13 | 0.16 | 0.11 | 0.26 | 0.11 | ||
Temperature | Coef | 0.0095*** | 0.011*** | 0.0091*** | 0.011*** | 0.011*** | 0.0097*** | |
P | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | ||
Precipitation | Coef | −2.5*** | −3.3** | −1.6 | −3.4*** | −3.3*** | 0.097 | |
P | <0.001 | 0.023 | 0.30 | <0.001 | <0.001 | 0.96 | ||
Constant | Coef | −144*** | −170*** | −138*** | −169*** | −164*** | −147*** | |
P | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | ||
N, control | 1473 | 0 | 173 | 395 | 295 | 55 | ||
N, treatment | 89 | 89 | 89 | 73 | 64 | 44 | ||
N, districts (clusters) | 133 | 34 | 71 | 91 | 29 | 40 |
The anticipatory model is described in Temporal Effects and Parallel Trends. A full description of matching and subsampling methods is provided in SI Text, Matching and Subsetting. For each outcome variable, columns compare models estimated on differently matched subsets of the dataset. We present coefficients of certification (Coef), the P value associated with the certification coefficient, and the number (N) of controls, treatments, and administrative districts. Three models use unmatched samples: NM, no matching; NC, no matching, restricted to certified plantations; and NR, no matching, restricted to RSPO member plantations. Models using matched samples through 2003 include: I, simple matching within island; D, matching within district; and C, matching within RSPO member company. Asterisks indicate degree of coefficient significance (*P < 0.10, **P < 0.05, ***P < 0.01).