Table 6.
Triple difference regressions results
| Reservation days | Number of listings | Reservation days per listing | Number of new entrants | Number of withdrawals | ||
|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | |
| After Enforcement x Bordeaux area x Inside the city | − 47.57*** (14.77) | − 47.57*** (13.88) | − 7.79*** (1.53) | 0.56 (0.44) | − 0.39*** (0.18) | 0.01 (0.06) |
| Effect as a % of mean† | − 35.33% | − 35.33% | − 53.9% | 5.95% | − 60.18% | 1.99% |
| Area FE | x | x | x | x | x | x |
| Month FE | x | x | x | x | x | x |
| Controls | x | x | x | x | x | |
| N | 8064 | 8064 | 8064 | 7621 | 8064 | 8064 |
| Districts | 168 | 168 | 168 | 168† † | 168 | 168 |
| Adjusted R2 | 0.22 | 0.31 | 0.27 | 0.26 | 0.23 | 0.09 |
| F -Statistic | 12.50*** (df = 194; 7869) | 19.19*** (df = 197; 7866) | 16.51*** (df = 197; 7866) | 14.94*** (df = 197; 7423) | 13.35*** (df = 197; 7866) | 4.91*** (df= 197; 7866) |
*p < 0.05; **p < 0.01; ***p < 0.001
†Mean in districts on both sides of Bordeaux’s border after regulation
† †Reservation days per district data are unbalanced, as we are unable to assign values for a given month in districts that do not have any listings for that month
We are interested in the triple difference interaction for each dependent variable, noted by “After Enforcement x Bordeaux area x Inside the city” in the table. The general specification can be found in the “Methodology” section. Standard errors are clustered at the area level and controls include population in the district as well as the number of restaurants and the number of hotels
Reading (2): Bordeaux’s districts had 47.57 fewer STR reservation days per month, relative to outside districts (in the 500 -m radius of interest), on average than they would have had without regulation. This equates to 35.33% percent of the mean STR activity in those districts, within that radius