08:59:37	 From  Lita-NBER Staff : Welcome to the Summer Institute 2020 HE Workshop
09:56:30	 From  Matthew Lindquist : test
10:26:19	 From  Kitt Carpenter/Nashville/he, him : Q&A starts in 5 minutes. If you'd like to get in the queue, please raise your hand or indicate so in the chat (privately to me or publicly is fine).
10:30:40	 From  David Slusky : Is it plausible to extrapolate these results to the U.S. setting?  Or are the prison systems and cultures so substantially different that it would be ill advised?
10:31:54	 From  David Slusky : Could you match this up with Peter Nilson's paper on lead data?
10:52:53	 From  Karen Eggleston : Great paper. And of course good economics need not be justified only by relevance to the US context.
11:00:09	 From  William Dow : Randi: Fascinating paper. In reading through it in more detail I see these are for relatively short sentences.  Are you able to say anything further about sentences on the longer side?
11:03:23	 From  Randi Hjalmarsson : Thanks William. This is for sentences of 4-48 months. There are few sentences longer. We are still working on whether we can say something about whether there are non-linear effects (in the 4-48 month interval), but it is tricky given that the marginal offender's characteristics differ so much across sentence length.
11:15:00	 From  karen norberg : Many thanks for this paper about under or over diagnosis in mental health - it is a serious issue, tough to measure, and this paper uses an interesting and plausible natural experiment.   
11:15:24	 From  Nicolas Ziebarth : Is the referral to a specific predetermined specialist? do you have provider networks or can patients freely choose their docs and get several opinions? 
11:19:10	 From  Nick Wilson : Is one mechanism why diagnosis in this setting affects life outcomes is that it is done by the employer?  For example, a mental health diagnosis by the employer adjusts the employer’s view of the employees productivity and raises, promotions, etc.
11:19:41	 From  karen norberg : topics crossing during the airwaves - my comments are going to be about the interpretation of the results. For those who don't know me - I'm retired child psychiatrist, but have also been a member of NBER community for many decades. 
11:19:46	 From  Janet Currie : It would be interesting to unpack the mechanisms.  e.g. could you compare people who were deemed unfit to serve for mental illness reasons to those unfit to serve for other reasons?  Can you see what drugs people take subsequent to diagnosis (since some of them have stronger side effects than others)?  One could also potentially look at longterm health consequences of some types of drugs such as diabetes or kidney problems.
11:20:28	 From  Kitt Carpenter/Nashville/he, him : As both authors are presenting, they will answer these great questions during the Q&A. keep them coming, and i'll call on you at that time. Thx!
11:20:29	 From  Peter Savelyev : Is it likely that some young men faked mental issues to avoid military service? Implications for your paper?
11:26:13	 From  Inas Kelly : Just curious: why is the number of observations in column (2) for labor market outcomes greater than that in column (1)?
11:26:13	 From  Peter Savelyev : As mental health diagnosis affects military service, can effects of these two treatments be separated?
11:27:20	 From  David Slusky : What about fertility? Are these men more or less likely to have children?
11:36:57	 From  Daniel Grossman : Does not serving in the military act as a potential source of stigma? So by not serving, employers are less likely to hire non-veterans?
11:38:29	 From  Randi Hjalmarsson : I have a question about heterogeneity….(can't fnd the raise hand function)...
11:41:51	 From  Sebastian Tello Trillo : Fine if doesn’t get on the queue: In order to determine leniency, was the severity (1-9) used? or only share of diagnosis per doctor?
11:42:01	 From  Rosalie Pacula : You are identifying on the "low severity cases" as the marginal cases, but it presumes there are clear diagnostic tools for these 3 specific diagnoses that are most likely to be "low severity cases" can be extremely difficult to diagnose -- any other health information on these people at the time of diagnosis.
11:44:01	 From  Sebastian Tello Trillo : Another vote for Rosalie question too ^
11:49:41	 From  karen norberg :  Hi, my computer does not seem to offer audio ... I'm following along, will give some comments to the authors off-line. Apologies!    
12:10:38	 From  Kitt Carpenter/Nashville/he, him : Do states/localities also regulate e-cig use on school grounds, and is that picked up by anrf? same question for e-cig advertising restrictions - are these part of the contemporaneous policy controls?
12:11:54	 From  Erik Nesson : Hi Kitt, I do not believe we include these as controls in our models.  Our ANR data may have these variables, but I am not sure about that.
12:12:06	 From  Erik Nesson : That is a good suggestion for us to look into.
12:15:05	 From  Rosalie Pacula : For Cotti et al paper: A 2019 study by David Levy, Frank Chaloupka and others published in Tob Regulatory Science looked at the market structure of the END industry and find that retail outlets represent only 40% of the purchase location for e-cigs (with internet and vape shops - not captured in the Nielson data) being more common... given selection issues regarding which individuals will buy e-cigs in traditional retail stores - how confident are you that these findings based on Nielsen data are generalizable?
12:15:29	 From  Mike Pesko : Hi Kitt, in terms of e-cig advertising restrictions, I'm not aware of much action by the end of 2017 at the state or sub-state level, but also worth looking into.
12:20:01	 From  Erik Nesson : Rosalie, this is a great question and one we've thought about quite a bit.  The NRSD provides a lot of advantages, but it is true that it doesn't capture these other avenues of purchases.
12:21:23	 From  David Cutler : Do the same taxes apply to internet sales?
12:22:26	 From  Mike Pesko : Hi David, shipping address is used to assess state e-cig taxes, if any.
12:22:57	 From  Ce Shang : Would the outcome make a difference? E.g. sales per capita or sales per tobacco user instead of overall sales?
12:23:07	 From  Benjamin Hansen : Related to internet sales, a lot of the advertising for e-cigs is on Instagram and social media.  Do individual states restrict this advertising???
12:23:18	 From  David Slusky : Building off David's question, is there an issue like there used to be with Amazon where internet retailers with no physical presence in the delivery state refuse to collect taxes?
12:23:28	 From  David Cutler : Thanks Mike. So that means there wouldn't be obvious substitution into internet sales as state taxes go up? If that occurred, your Nielsen elasticity of purchase would significantly overstate the usage elasticity. 
12:24:14	 From  Ce Shang : Another important policy to be controlled for may be state and local level tobacco flavor regulation.
12:24:43	 From  Mike Grossman : Saffer et al did not estimate own price elasticity for e-cigs
12:24:44	 From  John Cawley : There could be cross-border shopping to evade the tax (which would also lead to an overstatement of the true price elasticity of demand). You could test this - is there an impact on e-cig sales at stores just over the border from treated cities/counties.  
12:24:46	 From  Mike Pesko : Hi David, as long as the internet retailer is collecting the e-cigarette tax I don't see any reason for substitution. I have checked some of the major e-cigarette sellers to see if they add the tax and it appears they do so.
12:24:48	 From  Erik Nesson : We Ce, that's a nice suggestion.  My guess is that, given the fixed effects, the per-capita outcome will not be that different.
12:24:58	 From  Erik Nesson : *Hi Ce.
12:25:22	 From  Rosalie Pacula : I agree with David -- and finding of substitution for this particular population can be misconstrued to suggest that raising taxes on e-cigs are a bad idea as they encourage people to smoke cigarettes which are more dangerous... which may not be the case for the majority of the market … if there is substitution to other vaping products for the sample buying on-line and in vape shops rather than cigarettes -- seems like an important limitation for interpreting yoru results.
12:26:05	 From  Erik Nesson : John, thanks.  This is a good suggestion.  We find that the tax pass through is higher in states imposing taxes as opposed to localities.  We think this is consistent with cross-border shopping, but we have not examined this explicitly.
12:26:53	 From  Mike Grossman : Saffer et al did not estimate an own price elasticity for e0cigs
12:27:04	 From  Erik Nesson : Ben, this is a good question. To my knowledge, states cannot restrict advertising on social media.
12:27:31	 From  Marianne Bitler (she/her) : Do you know what share of total sales you capture in your data? Do you think the areas not in the data are more or less concentrated than the ones you have?
12:27:32	 From  Erik Nesson : Thanks, Kitt!
12:27:37	 From  Mike Pesko : Thanks Ce, I think most of the flavor regulations occurred after 2017, when our study period ends. But worth looking into.
12:27:44	 From  Ce Shang : Survey data often suggest the majority of sales happen at vape shops or online in the US.
12:28:27	 From  David Cutler : Is there a cite to the fact that most sales are online / in vape shops?
12:28:29	 From  Ce Shang : Sorry not “majority” I should have said “significant share”
12:30:36	 From  Ce Shang : https://www.mdpi.com/1660-4601/16/3/338
12:30:52	 From  Rosalie Pacula : Levy et al (2019) "An Economic Analysis of the PRe-Deeming US Market for Nicotine Vaping PRoducts Tobacco Regulatory Science 5(2): 169-181
12:43:11	 From  Erik Nesson : Ben, to your earlier question about substitution: we did examine the liquid per product and whether new products were introduced after a tax and did not find an effect.
12:43:12	 From  Charles Courtemanche : for johns question: we have results separately estimating impacts of state vs county taxes. elasticities are larger for county taxes, which is consistent w border crossing story (albeit indirect)
12:43:38	 From  Benjamin Hansen : Cool cool.  Thanks Erik!
12:44:18	 From  Erik Nesson : Note, Ben: This does not directly address your question.  But, we do have nicotine contents as well, so we could examine a similar outcome to explicitly examine nicotine contents.
12:44:28	 From  Erik Nesson : This is an excellent suggestion.
12:45:44	 From  Benjamin Hansen : The key issue is whether prices differ across product types, and if the taxes are per unit/ i.e. effectively are expensive products taxed less.  
12:45:48	 From  kaest : I suspect there is a substantial amount of variation within-area, particularly if price at all sales outlets were measured. The pass-thru of  a tax is likely to differ. If consumers with different consumption patterns shop at different outlets, it could bias estates significantly.
12:47:38	 From  Benjamin Hansen : Kitt, are you going to follow up on randomization to estimate the peer effects of social groups on research producitivty? 
12:49:03	 From  Nicholas Sanders : Love to participate, but have to “run off” to EEE session – hope everyone is safe and well
12:52:29	 From  Marianne Bitler (she/her) : 4-6
12:53:24	 From  Rosalie Pacula : Congratulations Kitt on a very nice job managing your first HE program as director very successfully!!   Super job!
12:56:00	 From  Nicolas Ziebarth : I agree. Excellent job!
12:57:06	 From  Mike Grossman : Much better than I ever did.  All I ever did  was ring the bell.
13:01:39	 From  Peter Savelyev : Great job, Kitt! Amazing multitasking.
13:03:28	 From  David Slusky : Daniel's you're coming through garbled
13:14:29	 From  Daniel Rees : I've had no problem with Zoom until now.  Oh well.  My apologies for audio. 
I wonder if we could invite people to discuss their own papers in a graduate class?

13:30:02	 From  Heather Royer : Great time! Thanks Kitt and Mike!
13:30:06	 From  Alison Cuellar : Thank you, Kitt and Mike!
13:30:17	 From  John Cawley : Thank you Mike and Kitt!
13:30:21	 From  Laura Wherry : Thank you!
13:35:41	 From  Johanna Maclean : Thank you Kitt and Mike!  And everyone!
13:39:39	 From  Lita-NBER Staff  to  Kitt Carpenter/Nashville/he, him(Privately) : Please return me as the host so that I can end the meeting when it is time
13:40:01	 From  Kitt Carpenter/Nashville/he, him  to  Lita-NBER Staff(Privately) : done. thank you so much.
13:41:02	 From  Lita-NBER Staff  to  Kitt Carpenter/Nashville/he, him(Privately) : Thank you and you are welcome
13:45:50	 From  David Slusky : I think we're at 5, which is on the small size (we normally have 8) but 2-3 may not be able to get visas, and if they show up late they can't get their fellowships.
13:45:56	 From  Marianne Bitler (she/her) : we got a little less than 2/3 of our usual yield.
13:46:34	 From  Marianne Bitler (she/her) : if higher ed survives
13:46:43	 From  Kitt Carpenter/Nashville/he, him : agree.
13:47:03	 From  Marianne Bitler (she/her) : UC is allowing another year for grad students although I think that it hasn’t been approved.