Where the number comes from
−0.262
The wage against its province's own poverty line, once the national year effect is held constant
Log minimum wage on log poverty line, 238 province-years, 34 provinces. The first specification asks whether a province's wage follows its own costs; the second removes whatever was happening to the whole country at once.
| specification | β | se | se, clustered | t | 95% CI |
|---|---|---|---|---|---|
| province FE | +1.146 | 0.034 | 0.042 | 27.12 | +1.063 – +1.229 |
| + year FE | −0.262 | 0.206 | 0.218 | −1.20 | −0.689 – +0.164 |
The first interval is nowhere near zero; the second comfortably contains it. The apparent tracking was the national inflation trend the whole time.
cross-section, one year at a time
| year | n | r | ρ | elasticity | r² |
|---|---|---|---|---|---|
| 2015 | 34 | 0.557 | 0.510 | 0.646 | 23.4% |
| 2016 | 34 | 0.471 | 0.354 | 0.568 | 16.6% |
| 2017 | 34 | 0.522 | 0.426 | 0.528 | 22.7% |
| 2018 | 34 | 0.531 | 0.452 | 0.519 | 22.9% |
| 2019 | 34 | 0.542 | 0.465 | 0.515 | 24.2% |
| 2020 | 34 | 0.517 | 0.416 | 0.497 | 21.9% |
| 2022 | 34 | 0.496 | 0.412 | 0.493 | 21.1% |
Even ignoring time, the poverty line accounts for a median 22.7% of the variation in what provinces set. Most of what fixes a minimum wage is not what that province costs.
