Table 2 Pooled OLS regressions showing that lobbying activities lead to less severe FDA recall classification between 2012 and 2019. Huber-White robust standard errors are clustered by firm; 95% confidence intervals are shown in square brackets; p-values are shown in parentheses; ∗ ∗ ∗ and ** indicates statistical significance at the 1% and 5% levels, respectively
From: The effects of lobbying on the FDA’s recall classification
Dependent variable: | Class | ||
|---|---|---|---|
(1) | (2) | (3) | |
Panel A: All Recalls | |||
Lobby | 0.145*** | ||
[0.052, 0.239] | |||
(0.002) | |||
Reports | 0.045*** | ||
[0.016, 0.073] | |||
(0.002) | |||
Amount ($M) | 0.309** | ||
[0.041, 0.577] | |||
(0.024) | |||
N | 46,519 | 46,519 | 46,519 |
Adjusted R-squared | 0.024 | 0.024 | 0.024 |
Panel B: Food Recalls | |||
Lobby | 0.567*** | ||
[0.426, 0.708] | |||
(< 0.001) | |||
Reports | 0.567*** | ||
[0.426, 0.708] | |||
(< 0.001) | |||
Amount ($M) | 1.012*** | ||
[0.760, 1.264] | |||
(< 0.001) | |||
N | 17,652 | 17,652 | 17,652 |
Adjusted R-squared | 0.043 | 0.043 | 0.043 |
Panel C: Drug Recalls | |||
Lobby | 0.022 | ||
[-0.128, 0.172] | |||
(0.771) | |||
Reports | 0.021 | ||
[-0.026, 0.068] | |||
(0.385) | |||
Amount ($M) | 0.112 | ||
[-0.342, 0.566] | |||
(0.629) | |||
N | 9852 | 9852 | 9852 |
Adjusted R-squared | 0.135 | 0.136 | 0.136 |
Panel D: Device Recalls | |||
Lobby | -0.004 | ||
[-0.089, 0.082] | |||
(0.935) | |||
Reports | -0.002 | ||
[-0.024, 0.020] | |||
(0.867) | |||
Amount ($M) | -0.048 | ||
[-0.245, 0.150] | |||
(0.637) | |||
N | 19,015 | 19,015 | 19,015 |
Adjusted R-squared | 0.035 | 0.035 | 0.035 |