10:15:19 From Lita-NBER Staff : Welcome to the Summer Institute 2020 EFMM Workshop 11:44:02 From Marco Del Negro : question for the Duarte paper: there are huge compositional effects on wages over the past few months. Unless you control for them hard to use wage data to draw conclusions, it seems to me. 11:45:22 From John Grigsby to All panelists : ^^ Agreed. You could try using wage data from the employment cost index to get a simple composition-adjustment for the wage. It's not perfect but at least controls for shifts in the occupation x industry mix through time. 11:46:17 From Miguel Faria-e-Castro : Marco: agree that is the main challenge to our exercise. We have a section in the paper devoted to trying to address that. We believe that doing the analysis at the NAICS-3 partly helps, as well as splitting into production vs supervisory employees. John: thanks a lot for that suggestion, we will definitely try that. 11:49:39 From Marco Del Negro : Thanks Miguel :-) Another issue: you are showing a point estimate. I would think that the bands are as large as I-95. if so, what is the point of showing point estimates? 11:53:07 From Miguel Faria-e-Castro : Thanks Marco. Good point, we couldn't figure a "readable" way of showing the bands without overwhelming the reader of the graph :) we have tables with the intervals in the paper, as well as tests for the difference between the supply and demand shocks. Tables 3-5 here: https://s3.amazonaws.com/real.stlouisfed.org/wp/2020/2020-011.pdf (apologies for throwing tables at you) 11:56:19 From Miguel Faria-e-Castro : For the results we highlight in both the paper and the slides (i.e. positive demand shock for Information in March), the shocks are statistically different from zero. 11:56:33 From Joao B. Duarte : Thank you so much for the comments Marco and John. Marco, there is indeed a lot of uncertainty. However, we get that for 6/13 of the sectors at NAICS-2 the difference in the relative size of the supply shock vs demand is statistically different than 0 with 68% credible intervals 12:01:34 From Marco Del Negro : last point: may want to see to what extent the results are robust to the Kitagawa - Giacomini approach. 12:01:47 From Marco Del Negro : Thank you for your answers btw :-) 12:02:24 From Miguel Faria-e-Castro : Thank you, we will look into that! 12:03:17 From Joao B. Duarte : Thanks. Will check that! 12:13:09 From Ivan Werning : One interesting extension we did not have time to present of our model is limited mobility, which this Roy model is another very interesting way to get at this. 12:17:25 From Christian Moser : @Mark & Iván: I was wondering about the Roy-style worker relocation mechanism… do we have evidence of the magnitude of sector switchers in the data? Seems like industry-specific human capital and frictions would prevent such relocation, especially if people expect the COVID-19 shock / lockdowns to be temporary. 12:24:17 From Ivan Werning : Following up on mobility: one point we make in the paper is that this mobility improves the outcome (lower drop) but can make the gap between outcome and efficient output greater 12:24:23 From Ivan Werning : Sorry it’s in the paper not presentation. 12:24:36 From James D. Hamilton : question for Ludwig et al. As I understand it, Keynesian supply shock comes in large part from people saying I can't buy A today so I'll spend less on B today so I can get more A tomorrow. Two questions. (1) Where does extra saving go in general equilibrium? (2) If I have the general belief that intertemporal substitution is low, would I be skeptical of this channel? 12:26:00 From Ludwig Straub to All panelists : Thanks for the Q. (1) In GE, income in sector B would decline (spending = income) and so there would actually not be any extra saving in equilibrium. 12:27:44 From Ivan Werning : To add to Ludwig’s respond: there can be an increase in saving though if at the same time government policy issues a lot of debt (as in reality) to pay for greater stimulus 12:27:45 From Ludwig Straub to All panelists : (2) One thing to keep in mind is that here the EIS should be inclusive of durables. Altogether, we think it was likely above 1 (as I mentioned: effective cost of consumption increased and at the same time spending decreased). But best evaluation would probably use the sufficient statistic exercise I flashed, that measures at the individual level how much people raised their consumption of unconstrained sector 12:31:49 From James D. Hamilton : Are the first two papers available somewhere? 12:34:45 From David Baqaee to All panelists : Hi Jim, the paper can be found at: https://www.dropbox.com/s/pmc2jjpttiolwrg/Keynesian_draft_v2.pdf?raw=1 12:35:51 From Guido Lorenzoni : https://www.nber.org/papers/w26918 12:36:03 From Ivan Werning : I just emailed you Jim 12:37:00 From James D. Hamilton : thanks! 13:19:25 From Gianluca Violante : Gianluca Violante is inviting you to a scheduled Zoom meeting. Topic: Break Out Rooms Time: Jul 15, 2020 01:00 PM Eastern Time (US and Canada) Join Zoom Meeting https://princeton.zoom.us/j/95572510080 Meeting ID: 955 7251 0080