WEBVTT 1 00:00:04.600 --> 00:00:09.779 Brad Davis: Welcome to the Tobacco Online Policy Seminar, TOPS. Thank you for joining us today. 2 00:00:09.970 --> 00:00:15.350 Brad Davis: I'm Brad Davis, a postdoctoral research fellow at the University of Missouri. 3 00:00:16.040 --> 00:00:22.309 Brad Davis: TOPS is organized by Mike Pesco at the University of Missouri, C. Shang at The Ohio State University. 4 00:00:22.430 --> 00:00:25.319 Brad Davis: Michael Darden at John Hopkins University. 5 00:00:25.540 --> 00:00:32.529 Brad Davis: Jamie Hartman-Boyce at University of Massachusetts Amherst, and Justin White at Boston University. 6 00:00:33.070 --> 00:00:37.359 Brad Davis: The seminar will be one hour, with questions from the moderator and discussant. 7 00:00:37.990 --> 00:00:52.019 Brad Davis: The audience may pose questions and comments in the Q&A panel, and the moderator will draw from these questions and comments in conversation with the presenter. Please review the guidelines on tobaccopolicy.org for acceptable questions. 8 00:00:52.230 --> 00:00:56.530 Brad Davis: Please keep the questions professional and related to the research being discussed. 9 00:00:56.820 --> 00:01:05.980 Brad Davis: Questions that meet the seminar series guidelines will be shared with the presenter afterwards, even if they are not read aloud. Your questions are very much appreciated. 10 00:01:06.190 --> 00:01:15.999 Brad Davis: This presentation is being video recorded and will be made available along with presentation slides on the TOPS website, tobaccopolicy.org. 11 00:01:16.350 --> 00:01:23.810 Brad Davis: I will turn the presentation over to today's moderator, Michael Darden from Johns Hopkins University, to introduce our speaker. 12 00:01:25.170 --> 00:01:44.779 Michael Darden: Thanks, Brad. Today, we continue our Winter 2026 season with a single paper presentation by Nita Mukhan entitled, Effect of Tobacco Sales Bans on Retail Sales in Beverly Hills and Manhattan Beach, California. This presentation was selected via a competitive review process by submission through the TOPS website. 13 00:01:45.050 --> 00:02:00.339 Michael Darden: Nita Mukond is a PhD candidate in epidemiology and translational sciences at UC San Francisco. Her current research focuses on tobacco control policies and tobacco use in persistently impoverished areas in California. 14 00:02:00.340 --> 00:02:05.779 Michael Darden: She completed a PharmD slash MBA at the University of Chicago… Illinois at Chicago. 15 00:02:06.000 --> 00:02:29.419 Michael Darden: Her clinical rotations in pulmonary and oncology settings spurred an interest in the harms of tobacco and its roles in perpetuating health inequities. Her previous research used cancer registry data to identify racial and ethnic disparities in cancer care and outcomes. Justin White, an associate professor at Boston University, is a co-author on the study and will answer select questions in the Q&A. 16 00:02:29.570 --> 00:02:32.259 Michael Darden: Nito, thanks for presenting for us today. 17 00:02:39.990 --> 00:02:42.339 Nita Mukand: Hi, thank you so much for having me. 18 00:02:47.600 --> 00:02:49.370 Nita Mukand: Is everyone able to see? 19 00:02:52.250 --> 00:02:53.950 Nita Mukand: Thumbs up would be appreciated. 20 00:02:54.750 --> 00:03:01.050 Michael Darden: Yes, if you could just speak up as loud as you can, or turn up the volume a little bit. Thanks. 21 00:03:01.050 --> 00:03:02.679 Nita Mukand: Yeah, thank you. 22 00:03:03.000 --> 00:03:23.259 Nita Mukand: So my name is Nita Mokand. I am a PhD candidate at the University of California at San Francisco, and I'd like to thank the Tobacco Online Policy Seminar Leadership Team for the opportunity to present the first in the nation tobacco retail sales bands in Beverly Hills and Manhattan Beach. 23 00:03:26.750 --> 00:03:43.520 Nita Mukand: This research was funded by the California Department of Public Health California Tobacco Prevention Program. The findings and conclusions of this work are those of the authors, and don't necessarily represent the views or opinions of CDPH or the California Health and Human Services Agency. 24 00:03:43.560 --> 00:03:55.519 Nita Mukand: My training was supported by a T32 from the National Institute on Aging, and I have no other competing interest to disclose and have never received funding from the tobacco-nicotine industry. 25 00:03:55.520 --> 00:04:05.489 Nita Mukand: The data used in this analysis requires the following disclaimer, which is that the results and conclusions are those of the researchers, and do not reflect the views of Nielsen IQ. 26 00:04:07.080 --> 00:04:25.740 Nita Mukand: I'd like to thank my co-authors, Elizabeth Anderson-Rogers, Rafael Colonna at the California Department of Public Health, Wendy Max at UCSF, and Justin White at BU. I'd also like to thank my dissertation committee, and especially Justin, for their support in conducting this research. 27 00:04:27.730 --> 00:04:32.510 Nita Mukand: This analysis was published this past November in Tobacco Control. 28 00:04:35.100 --> 00:04:46.620 Nita Mukand: On January 1st, 2021, two cities in LA County, Beverly Hills and Manhattan Beach, became the first in the nation to ban the sale of all tobacco products, including e-cigarettes. 29 00:04:46.920 --> 00:05:01.370 Nita Mukand: Both cities offered retailers a one-year hardship exemption, and conducted outreach to individual retailers. Beverly Hills also included a permanent carve-out for three existing cigar lounges, and offered small businesses consulting services. 30 00:05:01.520 --> 00:05:07.999 Nita Mukand: Additionally, Beverly Hills planned a City Council review of the impact on tourism after 3 years. 31 00:05:08.220 --> 00:05:19.280 Nita Mukand: These laws are enforced by relying on public comment and sending inspectors to look for tobacco products, and noncompliance would result in civil penalties from city prosecutors. 32 00:05:22.350 --> 00:05:35.610 Nita Mukand: But this policy is not without precedent. Bhutan banned the sale of commercial tobacco from 2004 to 2020, and during the COVID pandemic, South Africa, Botswana, and India implemented temporary bans. 33 00:05:35.800 --> 00:05:37.040 Nita Mukand: last year. 34 00:05:37.230 --> 00:05:54.430 Nita Mukand: Two cities in Marin banned the sale of tobacco products, and though there were no retailers in those cities, the law will prevent future tobacco sales there. Also last year, to protect the coastal environment, Santa Cruz banned the sale of tobacco products with plastic tips or filters. 35 00:05:55.690 --> 00:06:10.190 Nita Mukand: In addition to sales bans, tobacco-free generation laws, or nicotine-free generation laws, annually increase the minimum legal sales age such that anyone born after the cutoff will be unable to purchase commercial tobacco products. 36 00:06:10.340 --> 00:06:33.249 Nita Mukand: These have been implemented by 21 cities in Massachusetts, starting with Brookline. In the Maldives, those born after January 1st, 2007 will no longer be able to purchase commercial tobacco, and a nicotine-free generation law just passed in the House of Lords this Monday, and includes a clause that adults who purchase nicotine products on behalf of children 37 00:06:33.340 --> 00:06:34.760 Nita Mukand: We'll face a fine. 38 00:06:35.300 --> 00:06:54.049 Nita Mukand: These are just a few examples of tobacco and game strategies, which also include retail licensing restrictions, which limit the number or location of tobacco retailers, which stores can sell tobacco products, price increases, product restrictions, and advertising restrictions. 39 00:06:58.230 --> 00:07:15.419 Nita Mukand: This groundbreaking policy has already been the subject of public research… published research. A secret shopper study by Henriksen and colleagues found that 81% of Beverly Hills and 94% of Manhattan Beach retailers were compliant with the ban. 40 00:07:15.420 --> 00:07:19.559 Nita Mukand: And in a 2023 study by McDaniel and colleagues. 41 00:07:19.560 --> 00:07:28.299 Nita Mukand: Minimal difficulty in sales losses were reported by large chain stores, whereas smaller stores reported losing revenue and customers. 42 00:07:30.830 --> 00:07:36.320 Nita Mukand: This is the first quantitative study of the economic impacts of the ban. 43 00:07:36.860 --> 00:07:46.690 Nita Mukand: Our objective was to estimate the effects of the tobacco sales bans on retail sales in Beverly Hills and Manhattan Beach, which will be referred to as the treated cities. 44 00:07:46.990 --> 00:08:00.349 Nita Mukand: The primary outcomes were overall tobacco sales, and tobacco sales by category, which included cigarettes, cigars, smokeless tobacco, and electronic nicotine delivery systems, which includes e-cigarettes. 45 00:08:01.150 --> 00:08:06.079 Nita Mukand: We also included aims that address possible consequences of the ban. 46 00:08:06.210 --> 00:08:19.769 Nita Mukand: These included estimating tobacco sales in the areas surrounding Beverly Hills and Manhattan Beach to determine whether the tobacco sales that would have occurred in the treated cities were displaced to the surrounding areas. 47 00:08:20.750 --> 00:08:37.969 Nita Mukand: We also estimated sales of non-tobacco products in the treated cities and surrounding areas to ascertain the broader economic impacts of the ban, and especially whether shifts in tobacco purchasing were accompanied by other spending changes that might impact local retailers. 48 00:08:37.970 --> 00:08:43.890 Nita Mukand: For example, if people are buying tobacco less often, are they purchasing other products less often, too? 49 00:08:44.010 --> 00:08:49.479 Nita Mukand: Or are they spending more money on non-tobacco products, because now they aren't spending their money on tobacco? 50 00:08:54.220 --> 00:09:01.579 Nita Mukand: This analysis leveraged Nielsen's retail scanner dataset of store-level information on individual product sales. 51 00:09:01.810 --> 00:09:13.419 Nita Mukand: These data are generated from universal product codes and point-of-sale systems. UPCs are commonly known as barcodes and uniquely identify retail goods. 52 00:09:13.890 --> 00:09:27.240 Nita Mukand: The data was collected from a subset of large retail, grocery, drug, and convenience store chains that pre-approved sharing their store-level data, which included store addresses. 53 00:09:28.300 --> 00:09:33.720 Nita Mukand: This dataset covers approximately 16% of California's cigarette sales. 54 00:09:33.830 --> 00:09:44.930 Nita Mukand: And we use Nielsen's categorization of tobacco products, which includes cigarettes, cigars, smokeless tobacco, electronic nicotine delivery systems, which includes e-cigarettes. 55 00:09:46.130 --> 00:10:03.440 Nita Mukand: Our analysis included sales data between April 1st, 2018 and December 31st, 2022, which covers 32 months prior to the ban and 24 months after the ban, and it also precedes California's statewide restriction on the sale of flavored tobacco. 56 00:10:04.360 --> 00:10:15.199 Nita Mukand: The data were aggregated to quarterly sales, and we used the sum of the sales of all tobacco products to create a combined all-tobacco category. 57 00:10:19.920 --> 00:10:35.229 Nita Mukand: To estimate the effect of the bans, we compared store-level unit sales of tobacco products in Beverly Hills and Manhattan Beach with counterfactuals comprised of tobacco sales in a weighted subset of California stores not in the bordering areas. 58 00:10:35.560 --> 00:10:45.370 Nita Mukand: Additionally, stores that contributed to the counterfactual were matched based on treated store type, which included drug, grocery, and convenience stores. 59 00:10:45.880 --> 00:10:53.669 Nita Mukand: We used Synthetic Difference and Differences, which is a state-of-the-art causal inference technique that was first introduced in 2019. 60 00:10:59.170 --> 00:11:08.539 Nita Mukand: Synthetic Difference and Differences is a new, quasi-experimental approach that's a hybrid of two different methods, synthetic control and difference in differences. 61 00:11:08.690 --> 00:11:13.279 Nita Mukand: It's appropriate for panel data, also known as cross-sectional data. 62 00:11:13.800 --> 00:11:16.740 Nita Mukand: Like standard difference in differences. 63 00:11:16.850 --> 00:11:22.890 Nita Mukand: Synthetic difference in differences includes two-way fixed effects for unit and time. 64 00:11:22.960 --> 00:11:37.849 Nita Mukand: The unit fix effect, in our case stores, allows for each unit to have a different intercept, and accounts for stable unit characteristics, such as a city's tobacco use norms, or tobacco retail environment. 65 00:11:37.990 --> 00:11:50.810 Nita Mukand: The time-fixed effects account for period-specific shocks that affect all units the same. Examples would be statewide tobacco policies like the flavor restriction, or the onset of COVID-19. 66 00:11:51.010 --> 00:12:05.920 Nita Mukand: The synthetic control method takes a weighted average of the outcomes in cities that could potentially serve as comparators, driving weights that produce the best match in pre-policy outcomes and other prognostic factors that we adjusted for. 67 00:12:06.350 --> 00:12:21.690 Nita Mukand: And synthetic difference in differences makes use of the time and unit-specific fixed effects and difference in differences, incorporates unit-specific weights from the synthetic control method, and adds time-specific weights. 68 00:12:21.770 --> 00:12:32.279 Nita Mukand: And these time-specific weights balance out the control group, making the parallel trends assumption upon which the validity of difference in differences rests, more plausible. 69 00:12:34.910 --> 00:12:42.520 Nita Mukand: Furthermore, Synthetic difference in differences yields more precise and less biased estimates than either approach alone. 70 00:12:46.840 --> 00:12:59.930 Nita Mukand: The equations presented here will hopefully clarify what was explained in the previous slide. So on the left, difference and differences, two-way fixed effects regression with a unit fixed effect shown as alpha in the blue circle. 71 00:12:59.980 --> 00:13:09.059 Nita Mukand: On the right, synthetic control creates a weighted average using unit-specific weights, which is estimated omega in the yellow circle. 72 00:13:09.060 --> 00:13:20.710 Nita Mukand: This allows the synthetic control model to up-weight controls with the most similar pre-treatment trends in the treatment… to the treated group, which is necessary for the difference-in-differences approach. 73 00:13:20.750 --> 00:13:26.799 Nita Mukand: And then at the bottom, you can see how Synthetic Difference in Differences incorporates these elements. 74 00:13:27.100 --> 00:13:33.040 Nita Mukand: With the addition of the time-specific weight, estimated lambda, in the green circle. 75 00:13:37.740 --> 00:13:48.610 Nita Mukand: Our model included store and year-quarter fixed effects, and store type, and whether there was a local flavor tobacco sales restriction as covariates. 76 00:13:48.840 --> 00:13:58.470 Nita Mukand: We also performed a synthetic difference in differences event study to test pre-policy balance and see how the effects of the policy changed over time. 77 00:14:03.670 --> 00:14:21.029 Nita Mukand: To assess the effect of the ban on retail sales in the surrounding area, we analyzed data from stores of the same type within a 30-minute drive of Beverly Hills and Manhattan Beach, which is a similar amount of time to the average commute for both cities per the 2023 American Community Survey. 78 00:14:21.350 --> 00:14:37.300 Nita Mukand: The GeoAppify's Isoline application program interface was used to identify stores in the commuting area. Similar to the primary analysis, we used synthetic difference and differences with matching by and adjusting for store type. 79 00:14:37.700 --> 00:14:49.840 Nita Mukand: Beverly Hills is shown on the left, and Manhattan Beach is shown on the right. As you can see, both borders include the other treated cities, which was omitted as a possible synthetic control in these analyses. 80 00:14:52.940 --> 00:15:00.729 Nita Mukand: To determine whether stores in the treated cities lost business to those in the surrounding areas that did not have tobacco retail sales bans. 81 00:15:00.730 --> 00:15:16.960 Nita Mukand: We estimated the total dollar sales of non-tobacco products in the treated border areas using synthetic difference and differences. The non-tobacco sales included sales from all store departments except for alcohol and pet supplies, which are licensed separately. 82 00:15:18.800 --> 00:15:33.480 Nita Mukand: We used a permutation-based approach to estimate the confidence intervals for the synthetic difference-in-differences model. This involves dropping the treated units from the data and randomly assigning controls to serve as treated units. 83 00:15:33.480 --> 00:15:39.189 Nita Mukand: This new data structure is then used to generate 100 placebo estimates and calculate a variance. 84 00:15:42.300 --> 00:15:49.219 Nita Mukand: In addition to the main analysis, we conducted the following sensitivity analyses, which are included in the manuscript supplement. 85 00:15:49.310 --> 00:16:08.909 Nita Mukand: A pre-ban test of trend fit using the root-mean-square prediction error. A leave-one-out synthetic difference in differences where we sequentially re-estimate the model while excluding each donor city, as this approach helps to identify whether the results were driven by poor treatment fit in any particular donor city. 86 00:16:09.290 --> 00:16:28.869 Nita Mukand: We also performed analyses, incorporating an alternative, smaller definition of the border area that was limited to neighboring zip codes. An analysis to account for possible purchasing changes in package size by estimating dollar sales was performed, and non-tobacco sales were also analyzed 87 00:16:28.870 --> 00:16:31.640 Nita Mukand: By store type and store department. 88 00:16:34.490 --> 00:16:40.520 Nita Mukand: And before moving on to the results, take a moment for questions. Oh, a few moments. 89 00:16:40.850 --> 00:16:50.860 Michael Darden: Thanks so much, Nita. We're going to pivot to our discussant today, who is Adam Leventhal from the University of Southern California, for some questions and comments. 90 00:16:55.070 --> 00:17:03.880 Adam Leventhal: Hello, everyone. Yeah, thanks for having me, and this is a really interesting, important, and highly rigorous study. 91 00:17:04.420 --> 00:17:21.300 Adam Leventhal: I'll spend more time, maybe towards, you know, after we see the results and discussion, you know, talking a little bit more about the conceptual things. I'll call out just a couple of methodological questions that, Dr. Mukan, you can, like, just expand upon a little bit. 92 00:17:21.560 --> 00:17:38.579 Adam Leventhal: just for clarity purposes, I think you mentioned sometime, that your data set represents 16% of California's total tackle, tobacco sales. Can you unpack that and explain the context of what that means a little bit more? 93 00:17:38.750 --> 00:17:48.840 Nita Mukand: Absolutely, and I will go back to the slide that focuses on the Nielsen data to better illustrate that point. 94 00:17:49.820 --> 00:17:55.320 Nita Mukand: So, the way this data works is that Nielsen… 95 00:17:55.320 --> 00:18:15.400 Nita Mukand: licenses their data that they obtain from stores to be used for research purposes, and it would be infeasible for them to do so with every mom and pop, you know, corner or bodega. And so, unfortunately, it's limited to big chains. And in California, there's been a study, I believe it was 96 00:18:15.400 --> 00:18:18.810 Nita Mukand: about… it was tax-based? Justin, maybe you can… 97 00:18:18.840 --> 00:18:27.550 Nita Mukand: way in later, that estimated how much of sales were captured in this data. And so it's… 98 00:18:28.640 --> 00:18:35.869 Nita Mukand: It's the best data available to do this analysis, but it is limited to these big stores, and, you know, we… 99 00:18:35.870 --> 00:18:49.629 Nita Mukand: acknowledge that the conclusions aren't necessarily generalizable to independent stores, online sales, liquor stores, that aren't included in this dataset. That's a really important point. Thank you, Adam. 100 00:18:50.080 --> 00:19:07.209 Adam Leventhal: Oh, sure, okay. Well, maybe we could talk later, but, whether you have an estimate of, like, a little bit more narrow about Beverly Hills and, and Manhattan Beach, and, like, yeah, they may, they may be representative or non-representative of the statewide. 101 00:19:07.210 --> 00:19:10.130 Adam Leventhal: The distribution of tobacco retailer types. 102 00:19:10.350 --> 00:19:13.100 Nita Mukand: Yeah, yeah, that'll get addressed in the results. 103 00:19:13.100 --> 00:19:13.880 Adam Leventhal: Sounds great. 104 00:19:13.880 --> 00:19:14.790 Nita Mukand: Definitely important. 105 00:19:16.140 --> 00:19:38.920 Michael Darden: There are a few questions in the chat that I, want to just raise. Justin is actually doing a remarkable job of answering questions on the fly here. So, you know, one of the concerns with a total ban is always the kind of black market opportunities, and, one question was the extent to which you've thought about that, and, like, perhaps you could look at law enforcement data to see to what extent there was kind of… 106 00:19:39.150 --> 00:19:42.969 Michael Darden: that stuff going on, or that stuff increased during this period? 107 00:19:43.310 --> 00:19:44.839 Nita Mukand: Yeah, so… 108 00:19:45.020 --> 00:20:02.329 Nita Mukand: We're not able to measure that, though, as it relates to illegal sales, if they were to be conducted by stores in these municipalities, there is clearly on the Beverly Hills and Manhattan Beach municipal websites, numbers 109 00:20:02.330 --> 00:20:21.449 Nita Mukand: to call to report them, and there would be civil action against them if they were identified. And there was recently civil action against a large chain of tobacco retailers in LA that had been breaking the flavor ban. So it's not that there isn't any interest in… 110 00:20:21.510 --> 00:20:31.949 Nita Mukand: going after… businesses that flout the rules. So should there have been, any… 111 00:20:32.610 --> 00:20:35.159 Nita Mukand: Illegal sales through these stores. 112 00:20:35.280 --> 00:20:36.410 Nita Mukand: I think they… 113 00:20:36.410 --> 00:20:47.300 Michael Darden: I think the spirit of the question is more on the black market for cigarettes themselves. So, the sale of cigarettes outside of any formal establishment, so informal markets. 114 00:20:48.960 --> 00:20:57.550 Nita Mukand: Yeah, I mean, it'd be great if there were data to measure that, but things that happen outside of, you know. 115 00:20:59.480 --> 00:21:00.310 Nita Mukand: like. 116 00:21:00.610 --> 00:21:01.539 Michael Darden: Well, the suggest… 117 00:21:01.540 --> 00:21:02.160 Nita Mukand: Yes. 118 00:21:02.160 --> 00:21:06.080 Michael Darden: The suggestion was perhaps to use law enforcement data. 119 00:21:06.720 --> 00:21:11.700 Nita Mukand: Okay, I mean, we could. We could look into that. I don't know… 120 00:21:12.500 --> 00:21:22.430 Justin White: If I could maybe just jump in on that, I would say that I think that there are data on product seizures, for example, that might be used, but I'm not sure that they… 121 00:21:23.750 --> 00:21:31.940 Justin White: are released on the local level, but at any rate, our data and what we're reporting on are only taxpay sales, so it is a good point. 122 00:21:33.880 --> 00:21:51.119 Michael Darden: And then one other question, which Justin commented on in the chat already, but I'll ask you as well, Nita, just to think about, like, a lot of people buy their cigarettes at gas stations, and so what do you think about the absence of gas stations in your data, in terms of your results? 123 00:21:51.750 --> 00:21:59.120 Nita Mukand: Convenience stores are those that are or are not attached to gas stations, so I believe it captures gas stations. 124 00:21:59.120 --> 00:22:07.039 Michael Darden: Oh, so it does capture, okay, okay, great. Okay, great. Alright, well, let's, let's move on then and see the results. So, great, great work so far. 125 00:22:14.300 --> 00:22:18.870 Nita Mukand: So the, the first slide addresses your question, 126 00:22:19.390 --> 00:22:22.249 Nita Mukand: Adam, it was a good one. So… 127 00:22:22.300 --> 00:22:32.119 Nita Mukand: See, I'll start with the descriptive statistics. So the Nielsen IQ data included 2 tobacco retailers in Beverly Hills and 5 in Manhattan Beach. 128 00:22:32.120 --> 00:22:42.480 Nita Mukand: But their border areas include over 250 stores in 387 cities, and 300 stores in 350 cities, respectively. 129 00:22:42.790 --> 00:22:59.560 Nita Mukand: While not directly related to this analysis, per McDaniel and colleagues, three stores were reported to have closed in Beverly Hills and one in Manhattan Beach in the 18 months following the ban, though the authors couldn't determine whether the closures were due to the ban or other factors. 130 00:23:03.060 --> 00:23:12.629 Nita Mukand: These graphs depict trends in tobacco sales as a percentage of mean sales prior to the ban. The implementation date is indicated by the dashed black line. 131 00:23:12.890 --> 00:23:31.929 Nita Mukand: On the left, prior to the ban, there was a spike in tobacco sales in grocery stores, which may reflect stocking up in anticipation of the ban taking effect. Within our sample, tobacco sales ceased within 3 months in Beverly Hills and within a year in Manhattan Beach, indicating full compliance in our sample stores. 132 00:23:34.360 --> 00:23:44.739 Nita Mukand: This graph shows the results of the primary synthetic difference-in-differences analysis of unit sales of all tobacco by store before and after the ban. 133 00:23:45.130 --> 00:23:54.530 Nita Mukand: The treated cities, Beverly Hills and Manhattan Beach, are shown in red, and there's synthetic controls in blue. As you can see on the left-hand side of each graph. 134 00:23:54.580 --> 00:24:06.990 Nita Mukand: The treated and controlled stores exhibit parallel trends, meaning the unit sales per store in the matched comparator group provide a reasonably good fit for the results in the treated cities prior to the ban, which is 135 00:24:06.990 --> 00:24:19.600 Nita Mukand: necessary for these results to be valid. And on the right-hand side of each graph, sales in the treated city, again, shown in red, dropped precipitously, reflecting the end of tobacco sales in both areas. 136 00:24:22.460 --> 00:24:37.050 Nita Mukand: This graph shows all tobacco sales by store before and after the bands in the border areas. Here, border areas are shown in red, there are synthetic controls in blue. Both graphs also show very good pretend matches. 137 00:24:37.060 --> 00:24:49.679 Nita Mukand: And if tobacco sales had been displaced to the border, we would expect to see higher post-policy sales trends, but the observed lower trend is a strong indicator of no increases in cross-border shopping. 138 00:24:51.730 --> 00:25:09.189 Nita Mukand: This table shows the estimated effects of the bans on tobacco purchases in the border areas for all tobacco products and by-product category. The sale of all tobacco decreased in both border areas, though the difference was not statistically significant for the Beverly Hills border area. 139 00:25:10.840 --> 00:25:16.470 Nita Mukand: And cigars were the only product category with increased sales in the border areas. 140 00:25:20.340 --> 00:25:31.800 Nita Mukand: These four figures depict the results of the event studies of the all-tobacco results. In the treated cities, shown in the upper two panels, the sale of tobacco decreased precipitously. 141 00:25:33.380 --> 00:25:40.300 Nita Mukand: And there was no detectable increase in the sale of all tobacco in the border area, as shown in the lower two panels. 142 00:25:40.720 --> 00:25:45.860 Nita Mukand: These findings in the border area indicate a lack of evidence of cross-border shopping. 143 00:25:48.690 --> 00:25:53.760 Nita Mukand: Here, we have the event studies for the non-tobacco sales analyses. 144 00:25:53.970 --> 00:26:13.820 Nita Mukand: As you can see in the top panels, dollar sales of non-tobacco products did not decline in Beverly Hills, but decreased non-significantly in Manhattan Beach. Furthermore, there were no statistically significant changes in dollar sales of non-tobacco products in the treated or border areas when compared to the time before the sales ban. 145 00:26:14.170 --> 00:26:28.220 Nita Mukand: This suggests that stores didn't experience detectable changes in the sale of non-tobacco products after the ban took place. So any changes in foot traffic or substitution away from tobacco didn't affect their bottom line. 146 00:26:31.640 --> 00:26:40.739 Nita Mukand: The sensitivity analyses indicated that there was a close pre-band fit, such that the root-mean-squared prediction error was within 7.6% 147 00:26:40.740 --> 00:26:53.509 Nita Mukand: of the pre-band mean, or 0.1 standard deviations. In our next sensitivity analysis, we tested whether one comparator unit was responsible for our estimates by conducting a leave-one-out analysis. 148 00:26:53.510 --> 00:26:57.580 Nita Mukand: And found that our results didn't change when we dropped one unit at a time. 149 00:26:58.580 --> 00:27:06.149 Nita Mukand: The effect of the bands in the adjacent zip codes was similar to those observed in the larger border areas used in the primary analysis. 150 00:27:06.400 --> 00:27:17.580 Nita Mukand: And an analysis using dollar sales as the outcome yielded similar results as the analysis using units, suggesting that package size switching was unlikely to be a major concern. 151 00:27:17.860 --> 00:27:21.660 Nita Mukand: And there were also no differential effects by store type. 152 00:27:24.940 --> 00:27:37.420 Nita Mukand: Let me acknowledge some limitations of our analysis. It was restricted to major chains, as Nielsen does not have contracts with small businesses to collect their point of sales data. We lack… 153 00:27:37.560 --> 00:27:43.160 Nita Mukand: Information on independent stores, tobacco specialty stores, liquor stores, and online sales. 154 00:27:43.710 --> 00:27:48.279 Nita Mukand: Consequently, our results may not be generalizable to these stores. 155 00:27:48.740 --> 00:28:08.299 Nita Mukand: Additionally, the number of treated stores in Beverly Hills and Manhattan Beach is small, but the estimated effects come not only from those stores, but also the very large donor pool. So more than 1,200 stores for Beverly Hills, and more than 2,100 for Manhattan Beach contributed positive weight to the synthetic controls. 156 00:28:09.420 --> 00:28:25.960 Nita Mukand: We also don't have a means in this data set of measuring illicit tobacco sales, and any inference we would seek to make on tobacco use is limited by the fact that purchases are an imperfect proxy for tobacco consumption. 157 00:28:30.230 --> 00:28:45.649 Nita Mukand: Key takeaways from the study include that stores complied promptly with the tobacco sales bans, confirming Henrikin's findings of high compliance. As shown in the pre- and post-policy store-level figures, tobacco sales quickly dropped to zero. 158 00:28:46.290 --> 00:28:56.189 Nita Mukand: When analyzed as a group, tobacco sales did not increase significantly in the border areas, with the exception of cigars, demonstrating a lack of cross-border shopping. 159 00:28:56.790 --> 00:29:10.889 Nita Mukand: And Beverly Hills and Manhattan Beach retailers' fears of lost sales due to the ban were not borne out in statistically significant differences in the dollar sales of non-tobacco products evinced in the event study results. 160 00:29:14.200 --> 00:29:23.140 Nita Mukand: This study has important implications for policymakers by demonstrating a proof of concept and feasibility of tobacco sales bans. 161 00:29:23.960 --> 00:29:43.760 Nita Mukand: Tobacco sales bans, though, are just one tool in a larger toolbox of tobacco retail endgame strategies, which include the nicotine-free generation, tobacco retail licensing restrictions, which limit the number or location of tobacco retailers, which types of stores can sell tobacco products, price increases. 162 00:29:43.900 --> 00:29:47.500 Nita Mukand: Product restrictions and advertising restrictions. 163 00:29:48.190 --> 00:29:59.689 Nita Mukand: Additional experimentation will help researchers evaluate the impact of these policies, as Beverly Hills and Manhattan Beach may not be representative of other cities in California. 164 00:30:00.080 --> 00:30:01.620 Nita Mukand: or the U.S. as a whole. 165 00:30:03.670 --> 00:30:05.250 Nita Mukand: Thank you for listening. 166 00:30:05.420 --> 00:30:07.380 Nita Mukand: Happy to answer questions. 167 00:30:07.870 --> 00:30:12.700 Nita Mukand: And if any should come up later, you can contact me or Justin. 168 00:30:13.560 --> 00:30:31.369 Michael Darden: Thanks so much, that's a really terrific presentation, very clear, too. It was great. There's been a really active chat in the Q&A, so thanks to everybody, and thanks to Justin for his hard work there. We'll switch to Adam, Adam Leventhal from USC to provide some comments and questions. 169 00:30:33.550 --> 00:30:51.539 Adam Leventhal: Sure, yeah, great presentation, Nita. Obviously, like, a really rigorous and important study of a policy, where there's not a lot of opportunities to do these types of evaluation studies, so kudos to you and your team for capitalizing on this opportunity. 170 00:30:51.540 --> 00:31:00.760 Adam Leventhal: And, and addressing it. I think some bigger picture issues about sales bans. 171 00:31:01.900 --> 00:31:12.419 Adam Leventhal: I think, like, in some… in some senses, we kind of start thinking about prohibition, and kind of as an extreme… this is like an extreme policy tool. 172 00:31:12.420 --> 00:31:29.009 Adam Leventhal: Right? And it brings up some of those concepts that people might have, and your study really does a great job at showing and rigorously evaluating the potential collateral unintended consequences, which didn't seem 173 00:31:29.010 --> 00:31:31.190 Adam Leventhal: to, arise. 174 00:31:31.190 --> 00:31:49.659 Adam Leventhal: At least according to your data, with the exception of the cigar increase in the neighboring, locations, but that seemed to be, on the… on the whole, from a public health impact perspective, probably a modest. That may not have outweighed the, the positive impact. 175 00:31:49.660 --> 00:31:57.579 Adam Leventhal: Although there are, of course, other, potential issues that couldn't be addressed in your study, as you mentioned, right? About the, 176 00:31:57.580 --> 00:32:10.369 Adam Leventhal: the non-legal market. I think one thing that's interesting about this type of a policy is a lot of policies in tobacco control can be, ambiguous sometimes. 177 00:32:10.570 --> 00:32:13.129 Adam Leventhal: So, you know, flavors. 178 00:32:13.130 --> 00:32:31.020 Adam Leventhal: Right? What is a flavor? What is not a flavor? You know, California's had quite a time trying to clarify what that is. I think with a sales band, it's pretty clear, right? I think retailers have a good idea of what, they need to do. 179 00:32:31.150 --> 00:32:39.530 Adam Leventhal: So, I think that's another interesting aspect of this policy that sets it apart from others. 180 00:32:39.730 --> 00:32:46.179 Adam Leventhal: I think some issues to consider in the future, 181 00:32:46.370 --> 00:33:05.400 Adam Leventhal: like any types of study like this, when we're using sales data, it's always interesting to think about the correlation between sales and actual use in the population. And so, you know, if you do show a reduction in sales. 182 00:33:05.400 --> 00:33:08.670 Adam Leventhal: To what extent does that impact the… 183 00:33:08.770 --> 00:33:19.339 Adam Leventhal: person-level topography of tobacco use? Do we see, like, some people maybe using more and accounting for, or let… or maybe there's… 184 00:33:19.480 --> 00:33:39.410 Adam Leventhal: you know, not… not as big of an impact on the total number of the population who use tobacco products, but maybe the frequency or amount that they use goes down. Those are the types of questions, that… that kind of come up, but that's why we have a big field of tobacco control to triangulate. 185 00:33:39.410 --> 00:33:43.700 Adam Leventhal: And I think, across different types of research methods, so… 186 00:33:43.760 --> 00:34:00.829 Adam Leventhal: you know, some interesting future directions, and I'm not sure whether this is planned, is, are there… is there a possibility for doing a survey, of residents in these areas, and taking a look at their tobacco product use? 187 00:34:01.050 --> 00:34:05.950 Adam Leventhal: Or is there any data that's previously existing? 188 00:34:06.060 --> 00:34:12.939 Adam Leventhal: And, like, the kind of final few bigger picture comments I'll make, 189 00:34:13.170 --> 00:34:19.840 Adam Leventhal: is, I think one of the issues you brought up was maybe your dataset didn't have alcohol? 190 00:34:20.449 --> 00:34:44.359 Adam Leventhal: If I understood correctly, so it's always, like, an intriguing question with these types of studies, right, is whether, you know, you see a substitution effect, and people using more of another psychiatric agent, or a complementary effect, you know, which would mean that, you know, in this case, they would use less alcohol. Same goes for cannabis. 191 00:34:44.449 --> 00:34:57.459 Adam Leventhal: Right? Especially in this environment, in Southern California, where, you know, cannabis policies seem to be becoming less restrictive, and there's more products available. 192 00:34:57.620 --> 00:35:02.629 Adam Leventhal: And then kind of the final question and kind of point to make is. 193 00:35:03.690 --> 00:35:22.069 Adam Leventhal: there's… it's such an interesting policy. I don't know, and maybe our, our health economist experts may have some… some good ideas here, but do we have a comparator, right? Like, for another type of a product that… where a policy happened? 194 00:35:22.940 --> 00:35:33.209 Adam Leventhal: And we could kind of have… we could use those data and kind of that experience to think about how it might apply to tobacco control. 195 00:35:33.470 --> 00:35:38.500 Adam Leventhal: and this policy. So I don't know whether there were… there have been, 196 00:35:38.820 --> 00:35:49.619 Adam Leventhal: you know, sales bands of other products, and if there's lessons learned from that, also. And then, relatedly. 197 00:35:50.580 --> 00:35:55.320 Adam Leventhal: How much will this simulate when you have 198 00:35:55.600 --> 00:35:59.870 Adam Leventhal: A sales band for a few, like, two communities? 199 00:35:59.920 --> 00:36:19.509 Adam Leventhal: in a very, like, densely packed population, versus if a state or a county, like a large county like Los Angeles County, which has 10 million people, were to have a policy like this, would we see a totally different set of results or not? I don't know. 200 00:36:19.610 --> 00:36:34.829 Adam Leventhal: But your study did a great job at looking under every potential rock possible and giving us the best clues available as to what we can anticipate. So thank you, Nita, for your excellent study and great presentation. 201 00:36:35.940 --> 00:36:49.949 Nita Mukand: Thanks, Adam. It seemed like there were several questions in there. I'll do my best to address all of them, but should I, omit any, please just remind me to address them. One that is of particular interest is… 202 00:36:50.360 --> 00:37:06.120 Nita Mukand: whether there's survey data to indicate if there's any changes in smoking in response to these policies. And, this was part of my dissertation, so I, you know, my involvement is somewhat finite, but I wondered if… 203 00:37:06.390 --> 00:37:14.899 Nita Mukand: CHIS, the California Health Interview Survey, because they collect a lot of data on tobacco use, and they tend to collect it, 204 00:37:14.990 --> 00:37:34.260 Nita Mukand: like, a high density of sampling, like, much more so than Burfice, might be a good source of data to look at the consumption of tobacco products in these cities and around these cities before the bans and subsequent to the bans. 205 00:37:35.540 --> 00:37:49.290 Nita Mukand: Regarding a product that has also undergone a ban, That could be compared to… Tobacco sales bans. 206 00:37:50.660 --> 00:37:55.559 Nita Mukand: I can't think of anything that comes to mind, but I'm not an economist. 207 00:37:55.880 --> 00:38:08.919 Nita Mukand: And so, if anyone else, does think of a good comparator, I'd be really interested to hear that perspective. 208 00:38:09.160 --> 00:38:11.590 Nita Mukand: Let's see… your additional… 209 00:38:12.050 --> 00:38:28.110 Nita Mukand: questions. Oh yes, the substitution with liquor and, tobacco and other psychoactive substances. Yeah, it's a… it's a function of the way the data is collected, and it's unfortunate, but it's the best data we have. So while it's… 210 00:38:28.240 --> 00:38:33.820 Nita Mukand: disappointing to be able to give an incomplete picture. I think this is… 211 00:38:34.060 --> 00:38:43.170 Nita Mukand: a great start to the research on these policies, because, at least in our sample. 212 00:38:43.210 --> 00:38:50.580 Nita Mukand: There was no evidence of considerable economic changes for 213 00:38:50.610 --> 00:39:07.079 Nita Mukand: stores in the cities that implemented the ban, such that, policymakers in future might be concerned about economic consequences in whatever future municipalities, counties, states are looking at policies. 214 00:39:07.270 --> 00:39:08.660 Nita Mukand: like these. 215 00:39:09.210 --> 00:39:12.650 Nita Mukand: And Adam? 216 00:39:13.300 --> 00:39:14.930 Nita Mukand: What, what am I missing? 217 00:39:14.930 --> 00:39:20.539 Adam Leventhal: No, I think you hit the high points, so thank you. I really appreciate it. Great job. 218 00:39:20.830 --> 00:39:24.379 Nita Mukand: All great comments and questions. Thank you. 219 00:39:25.330 --> 00:39:36.239 Michael Darden: Several, several questions in the chat. One of them, you know, is kind of impossible to answer, but is always asked, and so I'll have to ask it as well. You have… 220 00:39:36.490 --> 00:39:50.740 Michael Darden: two extremely outlier places in the United States here, in Beverly Hills and Manhattan Beach. What do you think about the external validity of your results? 221 00:39:50.820 --> 00:39:56.520 Michael Darden: If you were to do this in a less densely populated area, or a lower SES area. 222 00:39:56.670 --> 00:39:58.770 Michael Darden: Can you conclude anything here? 223 00:40:00.830 --> 00:40:02.399 Nita Mukand: the way I see it. 224 00:40:03.010 --> 00:40:17.500 Nita Mukand: is that someone has to go first with these policies, and it might as well be a place where it's most feasible. And so, yes, the prevalence of smoking, the Henriksen paper looked at 225 00:40:17.750 --> 00:40:37.329 Nita Mukand: It is lower in Beverly Hills and Manhattan Beach than it is in California overall, and in California, it's the third lowest in the country, as far as states go. So we have already two levels of an uncommon willingness to control tobacco, and yes, a lot of financial resources. 226 00:40:37.560 --> 00:40:51.550 Nita Mukand: I think the generalizability is… limited, but… I… I hope that other… Areas won't necessarily take… 227 00:40:52.410 --> 00:41:00.660 Nita Mukand: differences that exist between them and Beverly Hills and Manhattan Beach as reasons not to try. 228 00:41:01.840 --> 00:41:14.420 Justin White: So maybe if I could add just one more point about that. So this doesn't get to the generalizability specifically, but I think that there is internal validity, even though these places are different. We fully recognize that 229 00:41:14.420 --> 00:41:23.989 Justin White: Beverly Hills and Manhattan Beach are not typical, in any way, but we, you know, I think the SDID procedure is designed to match 230 00:41:23.990 --> 00:41:39.899 Justin White: those cities to other, you know, a comparator group that looks more similar to them, and so that's what we tried to do in terms of estimating the changes. So it's not comparing Beverly Hills to all California cities, it's matched 231 00:41:40.030 --> 00:41:43.980 Justin White: A weighted comparator that sort of more resembles them. 232 00:41:44.240 --> 00:41:45.000 Nita Mukand: That's a great point. 233 00:41:45.000 --> 00:41:51.330 Michael Darden: Was it true in your results that there was a ramp-up pre-policy, in terms of sales? 234 00:41:51.330 --> 00:42:09.540 Nita Mukand: only in grocery stores, and to Justin's point, I think it was 250 and 300, respectively, something around there, cities contributed stores with positive weights for the synthetic controls. So, 235 00:42:09.700 --> 00:42:16.920 Nita Mukand: The sales trends that are being observed in these cities before the bans. 236 00:42:17.030 --> 00:42:23.909 Nita Mukand: were similar to sales trends observed elsewhere in California at that time period. 237 00:42:24.700 --> 00:42:38.440 Adam Leventhal: And just one other, like, point to add about generalizability, particularly Manhattan Beach, but Beverly Hills to some extent, too, is while the residents of those areas are definitely higher income. 238 00:42:38.440 --> 00:42:59.140 Adam Leventhal: A lot of the, people who work in those areas are from the service industry and are a little bit more representative of lower and middle incomes, so that's something to consider about the impact on those populations who typically buy their tobacco products, you know, on their, right after they get off their shift, before they go home, where they work. 239 00:43:00.060 --> 00:43:04.679 Adam Leventhal: Even if they're commuting a distance to get back to their residential community. 240 00:43:06.340 --> 00:43:16.049 Michael Darden: Can you speak a little bit to the public health implications of something like this, where they choose to ban both combustible tobacco and other tobacco products? 241 00:43:16.200 --> 00:43:28.659 Michael Darden: So, you know, if you're in the kind of harm reduction camp, you might think that this is not obviously good, because some people would have difficulty substituting because e-cigarettes are no longer available. 242 00:43:28.840 --> 00:43:34.479 Michael Darden: Just from a policy design point, can you comment on that? 243 00:43:35.670 --> 00:43:41.369 Nita Mukand: Yeah, so I think… It is seen as a… 244 00:43:42.340 --> 00:43:52.440 Nita Mukand: The fact that the policy doesn't cover a broader geographic area, I think, allows for individuals who take… 245 00:43:52.660 --> 00:43:59.850 Nita Mukand: A harm reduction approach to their tobacco consumption, more feasible. 246 00:44:00.100 --> 00:44:18.119 Nita Mukand: you know, they… these are relatively small cities. You saw what large geographic areas were covered by the half hour, and that's the average commute for both those cities, so that's not, like, an unusual amount of driving for people who live there. 247 00:44:18.750 --> 00:44:22.700 Nita Mukand: And so, it may be marginally harder. 248 00:44:22.840 --> 00:44:29.009 Nita Mukand: to… Obtain products that might be less harmful. 249 00:44:29.460 --> 00:44:38.050 Nita Mukand: But, it's… the way this is structured, it's, I think, a relatively low burden. 250 00:44:39.860 --> 00:44:42.999 Michael Darden: But that speaks to the scalability of the policy, right? 251 00:44:43.000 --> 00:44:52.420 Nita Mukand: It does, but I think future policies will have to address that in a way that this one, because of how it was designed, does not. 252 00:44:53.590 --> 00:44:54.160 Michael Darden: Right. 253 00:44:54.510 --> 00:44:55.220 Michael Darden: Right. 254 00:44:55.220 --> 00:44:58.379 Nita Mukand: They sort of kicked the can down the road on that one. 255 00:44:58.380 --> 00:44:59.340 Michael Darden: Because… Yeah. 256 00:44:59.410 --> 00:45:00.940 Nita Mukand: It was immaterial. 257 00:45:01.420 --> 00:45:13.120 Michael Darden: I don't know… I don't know that much about Los Angeles. Maybe Adam can comment on this, but, one… one question would be, it would be kind of nice to know where these service workers are commuting from, to kind of think about, 258 00:45:13.270 --> 00:45:17.469 Michael Darden: Characteristics of those people, to think about the representativeness and the external validity. 259 00:45:19.520 --> 00:45:35.900 Adam Leventhal: I don't have any specifics, but just my guesstimate is just abroad. This is the most heterogeneous population, in terms of county-wide, I think, that exists in the country, or one of them at least, and of course, most populous. 260 00:45:35.900 --> 00:45:40.019 Adam Leventhal: So, it's unique in some senses, but, 261 00:45:40.170 --> 00:45:47.859 Adam Leventhal: Yeah, so I'm not really sure, but some of us do commute, you know, an hour each way from work, or even more. 262 00:45:48.020 --> 00:45:49.810 Adam Leventhal: Me not included, luckily. 263 00:45:51.060 --> 00:45:56.600 Nita Mukand: And we do have demographic characteristics for the treated cities. 264 00:45:56.820 --> 00:46:14.170 Nita Mukand: their synthetic controls, and also the entirety of California. So you can compare how much Beverly Hills and Manhattan Beach are similar in that regard to places from which we drew sales data and the state overall. 265 00:46:15.680 --> 00:46:22.960 Michael Darden: One of the things that I'm interested in these kind of contexts is the supply response, the industry response. 266 00:46:23.000 --> 00:46:42.550 Michael Darden: So, you know, there's a tobacco control policy that's put in place, and industry responds in some way that potentially undercuts that response. In this context, it's such a small area, as you've mentioned a number of times, maybe there's no industry response. But if you tried to scale this, like, there's a producer of cigarettes, right? 267 00:46:42.550 --> 00:46:47.580 Michael Darden: And they might behave in fundamentally different ways. Do you have any ideas there? 268 00:46:49.060 --> 00:46:56.139 Nita Mukand: Well, I'm sure they will, pull from their… established playbook. 269 00:46:56.410 --> 00:47:05.370 Nita Mukand: To address the economic concerns that would come with the restriction of the sale of a highly lucrative product. 270 00:47:06.020 --> 00:47:09.119 Nita Mukand: But as a public health researcher. 271 00:47:09.320 --> 00:47:20.050 Nita Mukand: and specifically someone more on the epi side than the policy side, I'm really more interested in, what this could do for people's health. 272 00:47:20.250 --> 00:47:31.829 Nita Mukand: maybe, Justin, you might have… because he's… I'm relatively new to tobacco control. As was mentioned in my introduction, I came from the cancer control space. So, 273 00:47:32.010 --> 00:47:34.360 Nita Mukand: Justin might be a bit better equipped. 274 00:47:34.360 --> 00:47:43.590 Justin White: So, one thing I would mention is, one of the earlier studies that Nita had mentioned by Lisa Henriksen and colleagues. 275 00:47:43.630 --> 00:47:59.209 Justin White: They did, I think, also look at pricing of products in the surrounding areas to try to see if there were shifts there in terms of discounts, for example, and my recollection is that at least, 276 00:47:59.610 --> 00:48:18.280 Justin White: maybe within the immediate surrounding area, there weren't big, shifts in price after the bans went into effect. But that would be sort of, like, one study to think about. But more generally, I think you're right that, you know, there clearly could be supply-side responses to 277 00:48:18.490 --> 00:48:27.439 Justin White: these policies if they were scaled, and that is something I would expect. And I think it depends a little bit, sort of, on the… 278 00:48:27.640 --> 00:48:40.320 Justin White: you know, if other places don't adopt the policy exactly how Beverly Hills and Manhattan Beach is, that could also make a difference in terms of the types of products that get substituted towards in those sort of substitution patterns. 279 00:48:40.950 --> 00:48:46.720 Michael Darden: Yeah. One question that came in, just thinking about concurrent policies, so… 280 00:48:46.720 --> 00:49:02.389 Michael Darden: regarding Beverly Hills, there might be… there are bans of smoking outdoors unless you're actively walking, bans in parks of smoking, things like that. Are you worried about any kind of contamination of concurrent policy changes during this period? 281 00:49:04.360 --> 00:49:11.070 Nita Mukand: We incorporated the flavor bans for, but not other policies, yes. 282 00:49:11.230 --> 00:49:30.339 Nita Mukand: Into the analysis. The way I see those, and maybe, Dustin, you have a… something that you can add, is that places have their own cultures surrounding tobacco consumption. And so Beverly Hills and Manhattan Beach have already positioned themselves 283 00:49:30.460 --> 00:49:36.250 Nita Mukand: As cities that are not… Favorable to tobacco use. 284 00:49:36.530 --> 00:49:41.539 Nita Mukand: And… Maybe places that don't have such 285 00:49:41.680 --> 00:49:58.339 Nita Mukand: stringent regulations will find this more difficult to implement, but I'm really excited to see, hopefully other municipalities will follow suit, and we'll have more data to study how this operates differently in different places. 286 00:49:59.040 --> 00:50:08.839 Michael Darden: One thing to piggyback off of that, there was a question about whether or not you could get data on cigarette tax revenues. 287 00:50:09.320 --> 00:50:12.400 Michael Darden: I don't know if you thought about it, maybe you can comment on that. 288 00:50:13.170 --> 00:50:29.339 Justin White: So I can maybe jump in on that. I didn't actually… I haven't seen exactly where that is, but I believe at one point, we'd been trying to look at revenue data, and I think it's really hard to get at the local level, within California. So, yeah, that's a… 289 00:50:29.700 --> 00:50:35.150 Justin White: something that would be interesting, but I think is not available based on the way that it gets, collected. 290 00:50:37.200 --> 00:50:38.679 Nita Mukand: So it's a state-level tax. 291 00:50:38.920 --> 00:50:43.420 Nita Mukand: Like, there aren't necessarily local… Ways of splitting it out. 292 00:50:43.620 --> 00:50:45.830 Nita Mukand: Is that… do you know Justin? 293 00:50:46.140 --> 00:50:57.130 Nita Mukand: If that's part of the problem, that you can't… because we had store addresses, which was essential for the feasibility of this analysis, but a lot of data exists without geolocation information. 294 00:50:57.810 --> 00:51:07.719 Justin White: Yeah, I mean, I think the way that it gets reported to the California Department of, whatever it is, Tax and Finance Administration, it's… they don't have local data that they can provide. 295 00:51:09.170 --> 00:51:19.430 Michael Darden: Another thing that's come up, you know, it's not just service workers, these are highly touristy areas as well, so you have huge numbers of people coming to these areas. 296 00:51:19.670 --> 00:51:21.570 Michael Darden: They don't live there, and they're not working there. 297 00:51:22.310 --> 00:51:44.130 Nita Mukand: Beverly Hills did plan on an evaluation 3 years after the policy, though I have yet to see the results of it, on the effects on tourism. So it's something that is on their radar. But again, as mentioned, these are small places, it is not overly burdensome for people to go to a neighboring area and purchase tobacco, should that be their choice. 298 00:51:46.190 --> 00:51:53.299 Adam Leventhal: I think that, Michael, you brought up this point, though, about, like, the industry response, and I think that's something really… 299 00:51:53.370 --> 00:52:10.939 Adam Leventhal: important to think about, right? And kind of, like, do the thought experiment. And, you know, I guess there… I haven't been following, like, most recently, but I've heard that, you know, some of the bigger tobacco companies have been investing in cannabis, right? 300 00:52:11.060 --> 00:52:26.310 Adam Leventhal: And so, I mean, that's one thing to think about if there is some sort of, like, a substitution effect, and there's more marketing. You know, how do we think about that as a potential consequence? Certainly relative to combusted tobacco. 301 00:52:26.490 --> 00:52:33.390 Adam Leventhal: I think that some of us might feel that that's, not the worst trade-off, right? 302 00:52:33.420 --> 00:52:51.440 Adam Leventhal: But it'd be interesting to think about and study, and to what extent is… are the kind of the bigger tobacco industries, if we're, like, a large county or states were to start doing this, right? You know, are they going to think about alcohol? I'm not sure. Or, of course, potentially new psychoactives, 303 00:52:51.440 --> 00:52:54.040 Adam Leventhal: That would be outside of, 304 00:52:54.040 --> 00:52:57.699 Adam Leventhal: You know, the jurisdiction of tobacco enforcement. 305 00:53:02.680 --> 00:53:05.879 Michael Darden: Okay, so… 306 00:53:06.200 --> 00:53:19.899 Michael Darden: Any other questions? So, I think there is broad interest from universities and agencies in studying these things. I think my… my… I guess my overarching question for the… for you, for both of you, is… is this… 307 00:53:19.900 --> 00:53:30.730 Michael Darden: you know, this is kind of the logical end of the kind of tobacco control policy spectrum that we've seen, right? We have flavor bans, we have much higher taxes, we have indoor smoking laws. 308 00:53:30.920 --> 00:53:35.860 Michael Darden: But, you know, this is just outright banning it. Do… so… 309 00:53:35.880 --> 00:53:54.150 Michael Darden: have there been… it seems like litigation is going to come in at some point about this, right? And maybe, can either of you comment on, kind of, what that looks like going forward, as we scale these things to larger areas? Or we try these things in bigger areas? I mean, is this just going to get struck down? 310 00:53:55.920 --> 00:54:10.609 Justin White: So I'm not an expert on this, but I do think that that was initially a concern that Beverly Hills had. Sort of they, you know, were talking with their city lawyers about whether or not it would… 311 00:54:10.620 --> 00:54:18.149 Justin White: past monster or not, and I think the… neither of these cities, to my knowledge, have faced… 312 00:54:18.290 --> 00:54:20.779 Justin White: Legal challenges, 313 00:54:20.960 --> 00:54:31.560 Justin White: The thinking is that because it doesn't discriminate against certain types of products or certain retailers, it is sort of across the board that the bans that 314 00:54:31.560 --> 00:54:42.449 Justin White: It may actually be, legal to implement. But clearly, if it got scaled up, maybe, you know, there could be legal challenges elsewhere. 315 00:54:46.070 --> 00:54:47.520 Adam Leventhal: We can't hear you, Anita. 316 00:54:47.960 --> 00:54:56.640 Nita Mukand: The legal challenge that comes to my mind immediately is the ceremonial use of tobacco by American Indians and Alaska Natives. 317 00:54:56.890 --> 00:55:00.240 Nita Mukand: But, let's see… 318 00:55:02.090 --> 00:55:08.869 Michael Darden: Kathy makes a good point in the chat. It's not a ban on smoking, it's a ban on the selling in a retail environment. 319 00:55:09.640 --> 00:55:14.470 Michael Darden: Which brings back the question about, you know, black markets and illicit activity, but Adam? 320 00:55:14.790 --> 00:55:22.840 Adam Leventhal: Oh, I was gonna say, do we know anything about those dry counties for alcohol, and how the alcohol industry has responded to those? 321 00:55:24.250 --> 00:55:34.929 Michael Darden: Well, I think there's a large literature on that in the economics… in the economics literature. I mean, the obvious comparison here is prohibition of alcohol, right, which didn't go particularly well. 322 00:55:35.030 --> 00:55:36.890 Michael Darden: But, you know. 323 00:55:36.890 --> 00:55:39.170 Nita Mukand: 100 years ago, when things were harder to regulate. 324 00:55:39.170 --> 00:55:42.440 Michael Darden: Yeah, it's very difficult to understand how that would relate to now. 325 00:55:44.810 --> 00:55:51.559 Nita Mukand: It's also a product that doesn't directly harm people who are not consuming it like secondhand smoke does, but… 326 00:55:51.880 --> 00:55:56.260 Michael Darden: Well, I mean, there's… Well, I think people are responsible sex with alcohol, right? 327 00:55:56.260 --> 00:55:57.720 Nita Mukand: alcohol, but anyways… 328 00:56:02.910 --> 00:56:12.839 Michael Darden: Was there an uptick in sales in the cigar lounges that were able to stay open? I guess that must be yes, because you found that, and there were some that were allowed to… 329 00:56:12.840 --> 00:56:13.370 Nita Mukand: We did, right? 330 00:56:13.370 --> 00:56:13.750 Michael Darden: Are we… 331 00:56:13.750 --> 00:56:16.890 Nita Mukand: Have data from the cigar lounges. 332 00:56:17.250 --> 00:56:23.269 Nita Mukand: And so, because those are small, independent stores, those do not feed into Nielsen. 333 00:56:23.430 --> 00:56:26.029 Michael Darden: But you found something on cigars, didn't you? 334 00:56:26.030 --> 00:56:28.509 Nita Mukand: yes, was in the border area, and I suspect. 335 00:56:29.040 --> 00:56:32.400 Nita Mukand: That's a relatively inelastic demand for that product. 336 00:56:32.920 --> 00:56:33.620 Michael Darden: Right. 337 00:56:33.930 --> 00:56:36.610 Nita Mukand: I think people who smoke cigars are gonna smoke cigars. 338 00:56:39.030 --> 00:56:53.370 Adam Leventhal: Those, those are, you know, like, Nita, those are, like, cigar… like, large cigar, premium cigar retailers, or would… would this also include, like, the small cigars, little filtered cigars and things like that? Cigarios? 339 00:56:53.370 --> 00:56:57.790 Nita Mukand: I'm not sure what products, if you're asking… oh. 340 00:56:57.790 --> 00:57:01.559 Justin White: It does include little cigars and cigaritos as well, yeah. 341 00:57:02.280 --> 00:57:10.690 Nita Mukand: But do you know anything, Justin, about if there's any restrictions on the products that can be sold by the three exempted locations? 342 00:57:12.750 --> 00:57:15.869 Justin White: I mean, they can only sell cigars, I believe. 343 00:57:16.000 --> 00:57:17.569 Justin White: Yeah. And, you know… 344 00:57:17.570 --> 00:57:20.289 Nita Mukand: I don't know if they specify whether cigaretas are… 345 00:57:20.800 --> 00:57:22.710 Justin White: Oh, I see. Yeah, I'm not sure. 346 00:57:23.230 --> 00:57:29.839 Justin White: I imagine not, I mean, they're… yeah, they tend to be, I think premium cigars are what gets sold there. 347 00:57:32.060 --> 00:57:44.349 Michael Darden: Alright, well, thank you so much for a wonderful presentation. It's very interesting work, and I applaud the publication. So, we are just about out of time. We're going to kick it back to Brad to take us out. 348 00:57:51.330 --> 00:57:55.700 Brad Davis: We are out of time. Thank you to our presenter, moderator, and discussant. 349 00:57:55.840 --> 00:58:01.579 Brad Davis: Finally, thank you to our audience of 135 people for your participation. Have a tops-notch weekend.