The impact of JUUL market entry on cigarette sales: evidence from a major chain retailer in Canada

Table 4 Robustness checks

Monthly cigarettes volume sales	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Monthly cigarettes volume sales	No other control	Cluster at the city level	w/cig price	w/province-specific linear trend	w/city-specific season control	w/year and month FE separately	w/city-specific linear time trends
Monthly average (09/2017–08/2018)				737.1
Linear Trend after JUUL Entry	− 14.25***	− 12.15**	− 14.80***	− 10.44**	− 12.01***	− 13.09***	− 10.99***
Linear Trend after JUUL Entry	(4.073)	(4.660)	(4.264)	(4.125)	(4.236)	(0.860)	(0.596)
Cigarette Price			− 0.00226**
			(0.000876)
Observations	19,296	19,296	19,296	19,296	19,296	19,296	19,296
Adjusted R²	0.950	0.950	0.950	0.952	0.956	0.948	0.957
Store FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year-month FE	Yes	Yes	Yes	Yes	Yes
Province-specific linear time trend				Yes
City-specific seasonal control					Yes
Year FE and month FE						Yes	Yes
City-specific linear time trend							Yes
Robust errors clustered at the city level		Yes
Other controls		Yes	Yes	Yes	Yes	Yes	Yes

***p < 0.01, **p < 0.05, *p < 0.1
Year-month fixed effects (FE) are unique for each year and month. A total of 32 year-month fixed effects were included in the regressions. Control variables for local temperature, gasoline price, CPI, and unemployment rate were included in the specifications (2) to (5) but omitted in the table. Robust errors for coefficient estimates were clustered at the store level, except column (2). The dataset included 603 stores in 154 cities and towns from five provinces including Ontario. The analysis used observations from January 2017 to August 2019

ISSN: 1477-7517