1 00:00:00,030 --> 00:00:01,830 ♪ (music) ♪ 2 00:00:03,800 --> 00:00:05,800 (narrator) Welcome to Nobel Conversations. 3 00:00:07,270 --> 00:00:10,300 In this episode, Josh Angrist and Guido Imbens, 4 00:00:10,300 --> 00:00:13,800 sit down with Isaiah Andrews to discuss the key ingredients 5 00:00:13,800 --> 00:00:15,700 in their nobel-winning collaboration. 6 00:00:16,700 --> 00:00:19,800 Josh and Guido, first congratulations on the Nobel Prize! 7 00:00:19,800 --> 00:00:20,790 Thank you. 8 00:00:20,790 --> 00:00:23,190 (Isaiah) The work you did together, particularly the work 9 00:00:23,190 --> 00:00:27,300 on the local average treatment effect, or late framework 10 00:00:27,300 --> 00:00:28,900 was cited as one of the big reasons you won the prize. 11 00:00:29,200 --> 00:00:32,600 At the same time, you only overlapped at Harvard for a year-- 12 00:00:32,900 --> 00:00:34,300 if I'm remembering correctly-- 13 00:00:34,700 --> 00:00:36,700 it would be great to hear a bit more about how you started this collaboration 14 00:00:38,500 --> 00:00:41,500 and sort of what made your working relationship productive. 15 00:00:41,500 --> 00:00:43,600 Are there ways in which you felt like you complimented each other, 16 00:00:43,600 --> 00:00:46,790 sort of what got things started on such a productive, trajectory. 17 00:00:46,790 --> 00:00:48,800 Your job talk, as I recall Guido, it wasn't very interesting 18 00:00:49,600 --> 00:00:52,660 but I think it was a choice-based sampling-- 19 00:00:52,660 --> 00:00:53,330 It was. It was. 20 00:00:53,330 --> 00:00:54,000 (laughter) 21 00:00:54,600 --> 00:00:57,700 I was a very marginal hire there because they didn't actually interview me 22 00:00:58,400 --> 00:01:02,400 on the regular job market, but I think they were very desperate to get 23 00:01:02,500 --> 00:01:04,900 someone else to actually teach that course. 24 00:01:04,900 --> 00:01:08,470 It was after they had a couple of seminars already 25 00:01:08,470 --> 00:01:12,950 and it was still looking in econometrics, so Gary called me and kind of-- 26 00:01:12,950 --> 00:01:14,100 Gary Chamberlain? 27 00:01:14,100 --> 00:01:16,700 Gary Chamberlain called me and interviewed me over the telephone. 28 00:01:17,400 --> 00:01:21,292 He said, "Okay, well, my don't you come out and give a talk?" 29 00:01:21,292 --> 00:01:27,100 I remember this talk a little bit. 30 00:01:27,100 --> 00:01:29,000 I remember the dinner that you and Gary and I had. 31 00:01:29,000 --> 00:01:32,900 I remember not being very excited about your job market paper, 32 00:01:33,600 --> 00:01:38,220 but I saw that Gary was and luckily, Gary's view prevailed... 33 00:01:38,580 --> 00:01:39,600 Yes. 34 00:01:39,600 --> 00:01:41,900 ...and Harvard made you an offer 35 00:01:42,400 --> 00:01:46,300 and I think we started talking to each other pretty pretty soon after 36 00:01:46,300 --> 00:01:49,810 you arrived in the fall of 1990, right? 37 00:01:49,810 --> 00:01:51,600 Now as I said, I came and I didn't have a very clear agenda. 38 00:01:51,600 --> 00:01:55,700 I was a little intimidated getting there. 39 00:01:56,000 --> 00:01:58,700 But Gary kind of said, "No, you should talk to Josh." 40 00:01:58,800 --> 00:02:01,500 You should go to the labor seminar, kind of see what these people do. 41 00:02:01,500 --> 00:02:06,600 They're doing very interesting things there." 42 00:02:06,800 --> 00:02:08,700 I listened to Gary. 43 00:02:10,000 --> 00:02:12,900 As we did. 44 00:02:13,800 --> 00:02:15,700 As we did in the those days and ever since. 45 00:02:15,700 --> 00:02:16,700 I think it helped it, we were neighbors. 46 00:02:16,700 --> 00:02:18,400 So we both lived in Harvard's junior faculty housing, 47 00:02:21,500 --> 00:02:25,600 partly because housing costs were very high in Cambridge 48 00:02:25,600 --> 00:02:27,400 relative to our salary, which was very low. 49 00:02:27,800 --> 00:02:31,100 I think it also kind of made a difference, neither of us came from Cambridge, 50 00:02:31,200 --> 00:02:36,300 so there were a lot of MIT people who kind of already had their whole networks, 51 00:02:36,800 --> 00:02:38,000 kind of our collaborators. 52 00:02:38,300 --> 00:02:39,300 ♪ (music) ♪ 53 00:02:39,600 --> 00:02:43,700 (Josh) Well, I think we had figured out a mode of working together also. 54 00:02:43,800 --> 00:02:46,600 We had kind of a regular date, so we were neighbors 55 00:02:47,000 --> 00:02:48,900 and we often did our laundry together. 56 00:02:49,300 --> 00:02:52,000 We didn't have laundry machines at our apartments. 57 00:02:52,500 --> 00:02:56,800 But we used to do our laundry and we were talking 58 00:02:56,800 --> 00:02:59,300 and you had a way of very systematically, 59 00:03:00,100 --> 00:03:03,300 addressing questions that would come up in our discussions 60 00:03:03,300 --> 00:03:05,800 and the one thing that I was very impressed by, 61 00:03:06,400 --> 00:03:09,000 our early interaction, is you would follow up. 62 00:03:10,000 --> 00:03:10,500 Yeah, 63 00:03:10,500 --> 00:03:11,770 You would write some things down. 64 00:03:11,770 --> 00:03:13,400 Looking back at those days, sort of clearly, 65 00:03:13,400 --> 00:03:16,460 just had a lot more time to actually think 66 00:03:16,460 --> 00:03:19,100 -- I mean, I look at my junior college now-- -- You don't have time to think now. 67 00:03:19,600 --> 00:03:23,800 (Guido) No, but for me that is kind of one thing, 68 00:03:24,100 --> 00:03:26,200 but I feel now a lot of my junior colleagues 69 00:03:26,200 --> 00:03:27,200 don't actually have a lot of time to think. 70 00:03:27,200 --> 00:03:31,500 People are just doing so many projects, and it's actually so hard 71 00:03:31,800 --> 00:03:34,560 and there's so much pressure on people to publish that. 72 00:03:34,560 --> 00:03:39,200 I remember spending a lot of time sitting in my office and thinking, 73 00:03:39,700 --> 00:03:41,600 "Wow, what shall I do now?" 74 00:03:42,050 --> 00:03:43,050 (laughter) 75 00:03:43,500 --> 00:03:45,300 But it would give me a lot of time to actually think about these problems 76 00:03:45,300 --> 00:03:49,100 and trying to figure it them out 77 00:03:49,100 --> 00:03:50,900 and I could actually go to seminars 78 00:03:52,000 --> 00:03:57,100 and then the next day have coffee or lunch with Josh or Gary 79 00:03:57,300 --> 00:03:58,500 and actually talk about those things. 80 00:03:58,700 --> 00:04:01,300 (Isaiah) You guys weren't actually at Harvard together all that long, 81 00:04:01,300 --> 00:04:03,300 so you started working together pretty quickly. 82 00:04:03,300 --> 00:04:06,600 Were you both in the mindset that you were looking for co-authors, 83 00:04:06,600 --> 00:04:09,200 or looking for a particular type of types of co-authors at the time 84 00:04:09,400 --> 00:04:11,600 or was it more sort of fortuitous than that? 85 00:04:11,700 --> 00:04:12,700 (Josh) I think we were lucky. 86 00:04:13,500 --> 00:04:17,700 I don't remember I was that I was looking 87 00:04:17,700 --> 00:04:18,700 Now that I think, it was more fortuitous. 88 00:04:18,700 --> 00:04:19,900 I said I came in, I'd done my job market paper, 89 00:04:21,600 --> 00:04:24,500 and another paper for my thesis 90 00:04:24,500 --> 00:04:25,500 and I was just very happy to come to Harvard 91 00:04:25,500 --> 00:04:29,000 and suddenly there were all these seminars to go to, 92 00:04:29,200 --> 00:04:30,700 and lots of interesting people to talk to, 93 00:04:31,200 --> 00:04:36,000 but it wasn't a very conscious thing on my part. 94 00:04:36,300 --> 00:04:39,200 Looking back, I think there was a moment for me, 95 00:04:39,300 --> 00:04:41,800 where I was discussing instrumental variables, 96 00:04:42,200 --> 00:04:45,900 potential outcomes, treatment effects with Guido 97 00:04:47,000 --> 00:04:50,600 and we had a pretty good discussion, 98 00:04:51,300 --> 00:04:54,600 but then he also sent me some notes 99 00:04:56,000 --> 00:05:01,400 and the notes were very methodical write-up of our discussion 100 00:05:01,800 --> 00:05:03,300 and what you thought, 101 00:05:03,600 --> 00:05:07,900 we had been concluding in a fairly formal way 102 00:05:08,200 --> 00:05:10,630 and I thought, "Well, that's great." 103 00:05:10,630 --> 00:05:13,100 Talk is cheap, right, but with somebody.. 104 00:05:13,100 --> 00:05:15,544 - (Guido) Yeah, but-- - ...really writes out their story. 105 00:05:15,900 --> 00:05:18,500 (Guido) For me, it really helps writing things down 106 00:05:18,500 --> 00:05:22,960 and I do remember working with Josh 107 00:05:23,330 --> 00:05:25,200 and sitting in my office and writing things out 108 00:05:25,400 --> 00:05:28,130 and you guys have all had the discussions with Gary 109 00:05:29,500 --> 00:05:32,551 where afterwards we need to then sit down 110 00:05:32,551 --> 00:05:34,850 and actually write things up 111 00:05:34,850 --> 00:05:36,700 to figure out exactly what was going on. 112 00:05:37,400 --> 00:05:39,600 I think the other thing we had, Guido, 113 00:05:39,600 --> 00:05:41,815 is we had some very concrete questions 114 00:05:41,815 --> 00:05:43,750 that came from applications. 115 00:05:43,750 --> 00:05:45,000 (Guido) Yeah. 116 00:05:45,600 --> 00:05:50,400 A lot of econometrics, in my view, 117 00:05:50,400 --> 00:05:51,400 that we were schooled in was about models, 118 00:05:51,400 --> 00:05:55,000 Here's a model and what can you say about this model? 119 00:05:55,300 --> 00:06:00,200 I think we were thinking about, here's a particular scenario, 120 00:06:00,500 --> 00:06:03,800 draft eligibility is an instrument for whether you serve in the Army. 121 00:06:04,400 --> 00:06:06,300 What do we learn from that? 122 00:06:06,300 --> 00:06:07,300 (Guido) That's right. 123 00:06:07,300 --> 00:06:09,200 That's right, and that's sort of where your influence 124 00:06:10,800 --> 00:06:15,200 on the way I do research now is still very clear-- 125 00:06:15,700 --> 00:06:16,700 ♪ (music) ♪ 126 00:06:17,200 --> 00:06:19,400 (Isaiah) I guess zooming out a little bit, just thinking about 127 00:06:19,400 --> 00:06:22,200 when you guys started working on this, 128 00:06:22,200 --> 00:06:23,000 when you started working together, 129 00:06:23,100 --> 00:06:24,500 any thoughts for folks 130 00:06:24,500 --> 00:06:27,200 who are just interested in finding productive 131 00:06:27,200 --> 00:06:28,400 co-authors being productive? 132 00:06:28,400 --> 00:06:30,900 I mean, Guido already mentioned the importance of having time, 133 00:06:31,200 --> 00:06:34,000 right, which it is. 134 00:06:34,000 --> 00:06:35,000 It is very easily not to have a lot of time to think-- 135 00:06:35,000 --> 00:06:38,000 You definitely have to make time. 136 00:06:38,000 --> 00:06:40,400 That's a great question, though, Isaiah, 137 00:06:40,400 --> 00:06:45,700 and I tell my students that you should pick your co-authors 138 00:06:45,800 --> 00:06:46,800 as carefully maybe more carefully than you pick your spouse. 139 00:06:52,200 --> 00:06:52,900 You want to find co-authors who, 140 00:06:53,800 --> 00:07:01,200 you have some complementarity 141 00:07:01,700 --> 00:07:02,700 and that's what makes a strong relationship. 142 00:07:03,200 --> 00:07:06,900 You don't want to work with somebody who sees the world exactly like you 143 00:07:09,300 --> 00:07:13,800 and as much as Guido and I agree about things, 144 00:07:14,200 --> 00:07:16,800 we often disagree about things to this day 145 00:07:16,900 --> 00:07:19,250 and it's fruitful to have those discussions 146 00:07:19,250 --> 00:07:21,400 and we had complimentary skills. 147 00:07:21,400 --> 00:07:24,700 I was very empirical. I'm not really an abstract thinker. 148 00:07:25,500 --> 00:07:29,800 Guido was great at figuring out what the principles were. 149 00:07:30,100 --> 00:07:34,500 Yeah, that's right and I totally agree, kind of [a different spot.] 150 00:07:34,700 --> 00:07:37,829 These are incredibly important relationships 151 00:07:37,847 --> 00:07:42,400 and you see a lot of people working together 152 00:07:42,600 --> 00:07:46,600 and not necessarily working very well 153 00:07:47,000 --> 00:07:49,300 and then it's very hard often to get out of this relationship. 154 00:07:52,900 --> 00:07:56,000 A good partnering is a beautiful thing, like a marriage. 155 00:07:56,500 --> 00:07:58,500 It produces wonderful children, 156 00:07:59,500 --> 00:08:04,000 the fruits of the scholarship are potentially wonderful 157 00:08:04,200 --> 00:08:07,700 and they exceed the capacity of the partners to do it on their own 158 00:08:07,900 --> 00:08:13,200 but a bad co-authorship can be very destructive and time consuming and painful, 159 00:08:13,200 --> 00:08:16,800 just like a bad marriage. 160 00:08:16,900 --> 00:08:20,800 Arguments may start about who did what when 161 00:08:21,100 --> 00:08:23,700 and intellectual property type issues, 162 00:08:23,700 --> 00:08:24,700 especially when it when it goes a little sour 163 00:08:24,700 --> 00:08:28,250 and somebody thinks the other party is not pulling their weight. 164 00:08:30,100 --> 00:08:32,300 There's more co-authorship now in economics, 165 00:08:32,300 --> 00:08:33,700 I think that's been documented, much more. 166 00:08:33,700 --> 00:08:34,700 (Guido) Yes. 167 00:08:34,700 --> 00:08:37,700 There's more teams and there's larger teams 168 00:08:38,100 --> 00:08:41,400 and I think that's great, I love working on teams. 169 00:08:41,400 --> 00:08:46,800 We do work on schools with big teams. 170 00:08:47,000 --> 00:08:50,100 I work often with PI teammates like Parag Pathak and David Autor 171 00:08:50,100 --> 00:08:51,100 and then a team of graduate students, 172 00:08:51,100 --> 00:08:55,400 but I see that the students are not always, 173 00:08:55,400 --> 00:08:57,700 in some ways they're a little too promiscuous, 174 00:08:57,700 --> 00:08:58,700 in my view, in their partnering. 175 00:08:58,700 --> 00:09:02,600 They don't think it through. 176 00:09:02,600 --> 00:09:03,600 It's difficult to think it's through. 177 00:09:03,600 --> 00:09:08,500 I think, for me, working with people always has involved 178 00:09:08,800 --> 00:09:11,400 spending a lot of one-on-one time with people, 179 00:09:11,600 --> 00:09:16,400 you need to figure out how they think 180 00:09:16,900 --> 00:09:18,500 and what kind of problems are interested 181 00:09:18,600 --> 00:09:23,400 and how they think about these problems, how they like to write, to make that-- 182 00:09:23,600 --> 00:09:26,600 And it takes some maturity on everybody's part. 183 00:09:26,600 --> 00:09:27,600 Yes. Yes. 184 00:09:27,600 --> 00:09:29,600 In what sense? 185 00:09:29,600 --> 00:09:30,600 Just in the sense of knowing what's going to work for them, 186 00:09:30,600 --> 00:09:32,900 knowing when things are versus aren't working? 187 00:09:33,000 --> 00:09:36,500 (Josh) Maturity in the sense of having some judgment 188 00:09:36,600 --> 00:09:40,100 to be able to face it honestly, if it's not going well, 189 00:09:40,300 --> 00:09:45,100 sometimes you have to have some difficult discussions. 190 00:09:45,250 --> 00:09:46,250 Is it worth continuing? 191 00:09:46,400 --> 00:09:49,100 "I was hoping you would do this, and you didn't," 192 00:09:49,200 --> 00:09:51,400 maybe it turns out there's some 193 00:09:51,400 --> 00:09:54,155 feeling in the other direction, the same way. 194 00:09:54,155 --> 00:09:56,800 And Josh is very good (chuckles) 195 00:09:56,800 --> 00:09:59,600 in the being honest, part from the beginning, 196 00:10:00,000 --> 00:10:03,600 (Josh) For better or worse. 197 00:10:03,600 --> 00:10:04,600 (Guido) I would write this stuff and then I remember the 198 00:10:04,600 --> 00:10:09,343 first version of the paper with Rubin, 199 00:10:09,343 --> 00:10:11,710 Josh was in Israel at the time, 200 00:10:12,900 --> 00:10:15,500 Don and I were in Cambridge and so I would talk with Don regularly, 201 00:10:16,300 --> 00:10:18,600 but Don wasn't really doing much writing in those days, 202 00:10:18,600 --> 00:10:20,400 I would write things and then I would fax them to Josh 203 00:10:20,500 --> 00:10:25,200 and they would come back, first page just one big cross, No, 204 00:10:25,300 --> 00:10:29,400 second page, one big line, No 205 00:10:30,800 --> 00:10:31,700 and that would go for awhile 206 00:10:31,700 --> 00:10:32,600 but he still does that. 207 00:10:32,600 --> 00:10:36,800 I sent him the first draft of my Nobel lecture, 208 00:10:36,900 --> 00:10:38,100 and Josh goes, No, no! 209 00:10:38,800 --> 00:10:43,300 I've gotten some PDF comments like that from Josh, very helpful. 210 00:10:45,700 --> 00:10:46,600 Omit needless words. 211 00:10:47,800 --> 00:10:52,000 I have few co-authors who are willing to do that. 212 00:10:53,200 --> 00:10:58,400 Especially as you get older, it's harder to put up with that. 213 00:10:59,300 --> 00:11:03,100 I would find it harder now to start working with people who did that 214 00:11:03,800 --> 00:11:05,600 early on in a co-author relationship. 215 00:11:05,600 --> 00:11:08,900 It's also very hard because you need to have enough trust. 216 00:11:09,300 --> 00:11:15,400 Josh, for being willing to be very critical, 217 00:11:15,700 --> 00:11:20,800 he was also willing to admit being wrong. 218 00:11:21,150 --> 00:11:22,150 ♪ (music) ♪ 219 00:11:22,500 --> 00:11:25,600 (Josh) But you have to be on the lookout for good partners, 220 00:11:25,800 --> 00:11:29,800 somebody who can help you answer questions that you can't answer yourself. 221 00:11:30,200 --> 00:11:33,000 I think there's a natural tendency for people to gravitate 222 00:11:33,000 --> 00:11:34,400 to people who are similar in outlook and skills 223 00:11:35,400 --> 00:11:41,000 and that's not as useful 224 00:11:41,000 --> 00:11:42,000 Josh is right, nowadays it's very tempting 225 00:11:42,000 --> 00:11:46,500 to find people who think about the same problems 226 00:11:46,500 --> 00:11:50,500 you're already thinking about, who think along the same lines 227 00:11:53,000 --> 00:11:56,400 and that may not lead to very novel stuff. 228 00:11:58,500 --> 00:12:02,700 But at the same time finding people who actually have very different ideas, 229 00:12:02,800 --> 00:12:05,000 it's going to take a lot of time. 230 00:12:05,200 --> 00:12:08,390 Guido, you mentioned in passing how working with Josh has influenced 231 00:12:08,390 --> 00:12:10,290 how you do research, 232 00:12:10,290 --> 00:12:11,600 could you say a little more about that? 233 00:12:11,600 --> 00:12:15,100 I'd also be interested to hear from Josh, did working with Guido 234 00:12:15,100 --> 00:12:17,200 influence the way that you do research? 235 00:12:17,500 --> 00:12:20,900 (Guido) Nowadays, I'm much more conscious of the fact that, for me, 236 00:12:20,900 --> 00:12:25,300 good economic research comes out of talking to people doing empirical work, 237 00:12:25,600 --> 00:12:29,300 and it's really not reading econometrica 238 00:12:29,800 --> 00:12:31,500 or the reading the stats journals, 239 00:12:31,500 --> 00:12:35,000 but it's actually talking to people doing empirical work, 240 00:12:35,100 --> 00:12:37,200 going to the empirical seminars. 241 00:12:38,100 --> 00:12:40,300 When I was at Berkeley, 242 00:12:40,400 --> 00:12:45,500 David Carr and [inaudible] as colleagues there 243 00:12:45,500 --> 00:12:46,700 and I would talk to them and listen to them, 244 00:12:46,900 --> 00:12:48,200 trying to figure out 245 00:12:49,900 --> 00:12:54,500 how are they solving their problems and other things there 246 00:12:54,700 --> 00:12:57,424 where I'm not really quite happy with the way they're doing things 247 00:12:57,424 --> 00:13:04,200 and trying to look for methodological problems, 248 00:13:04,200 --> 00:13:07,900 where there's some more general solutions possible. 249 00:13:07,900 --> 00:13:11,700 I tried to tell it to my students that I encourage them to work 250 00:13:11,800 --> 00:13:14,500 as research assistants also, 251 00:13:14,500 --> 00:13:18,724 for the people doing empirical work at Stanford. 252 00:13:19,700 --> 00:13:22,100 There was no [subbing] but that I learned while I was in graduate school, 253 00:13:22,100 --> 00:13:25,000 but it really came out of working with Josh. 254 00:13:25,000 --> 00:13:26,000 as well as talking to Gary, 255 00:13:26,000 --> 00:13:31,000 Gary us was always encouraging of doing that 256 00:13:31,000 --> 00:13:33,600 and because he done that himself, 257 00:13:33,600 --> 00:13:36,900 he'd worked with on empirical problems with Zvi Griliches 258 00:13:36,900 --> 00:13:39,500 early in his career. 259 00:13:39,500 --> 00:13:40,500 Yeah. 260 00:13:40,500 --> 00:13:44,600 Well, I became more more interested in the econometric theory 261 00:13:45,400 --> 00:13:47,000 through our interaction, 262 00:13:47,100 --> 00:13:52,400 and I think empiricists are often impatient with econometric theory, 263 00:13:52,400 --> 00:13:55,500 partly because empirical work is very time-consuming, 264 00:13:56,000 --> 00:13:59,100 and you may have a sense that something is 265 00:13:59,300 --> 00:14:02,400 convincing and sensible 266 00:14:03,000 --> 00:14:04,100 and you haven't really fully made the case for that, 267 00:14:04,100 --> 00:14:05,100 but you're convinced 268 00:14:05,100 --> 00:14:09,700 and that motivates you to pursue it, like the draft lottery story. 269 00:14:10,700 --> 00:14:17,300 I was pretty sure that was worth doing 270 00:14:17,300 --> 00:14:21,100 and I came away from working with Guido 271 00:14:21,100 --> 00:14:24,800 seeing that there was the potential to say something 272 00:14:24,800 --> 00:14:25,800 more than just about that particular problem, 273 00:14:25,800 --> 00:14:29,500 and I think over the those early years in the 90s, 274 00:14:29,500 --> 00:14:35,000 our thinking evolved together that there's actually a framework, 275 00:14:35,100 --> 00:14:37,800 a way to solve a lot of problems 276 00:14:38,200 --> 00:14:41,700 and I think that that is the power of the late framework, 277 00:14:41,700 --> 00:14:42,800 is it answers a lot of questions in some sense. 278 00:14:43,150 --> 00:14:44,150 ♪ (music) ♪ 279 00:14:44,500 --> 00:14:46,300 In some sense, did you find that, 280 00:14:46,300 --> 00:14:50,700 email versus facts versus in -person, the medium mattered 281 00:14:50,700 --> 00:14:52,000 to how collaboration went 282 00:14:52,100 --> 00:14:55,200 or they're ways that you felt like it was the most useful to collaborate? 283 00:14:55,300 --> 00:14:59,700 To me, I think what matters most is, initially you have a period-- 284 00:15:00,000 --> 00:15:04,900 We needed that initial period, that was very intense with almost 285 00:15:05,100 --> 00:15:08,800 daily interaction and we also became friends. 286 00:15:08,900 --> 00:15:13,900 You don't develop the kind of friendship, electronically usually 287 00:15:15,000 --> 00:15:19,000 but once you have that foundation you can be pen pals 288 00:15:19,300 --> 00:15:25,300 and we did use e-mail, though it wasn't as useful then 289 00:15:25,500 --> 00:15:28,400 but it worked, but we definitely had a lot of faxes. 290 00:15:28,400 --> 00:15:34,000 I still have these faxes, long faxes 291 00:15:34,000 --> 00:15:35,000 and then in the summer, I would come to Cambridge, 292 00:15:35,000 --> 00:15:40,300 usually to the NBR meetings and hang around for a few weeks 293 00:15:40,300 --> 00:15:43,000 and you visited me in Israel. 294 00:15:43,000 --> 00:15:44,000 I visited in Israel. 295 00:15:44,000 --> 00:15:48,400 But yeah, there was good foundation from that that year 296 00:15:48,500 --> 00:15:51,000 and in some sense that was enough. 297 00:15:51,500 --> 00:15:53,000 and nowadays, 298 00:15:53,300 --> 00:15:56,600 I have the co-authors in lots of different places, 299 00:15:56,600 --> 00:15:59,100 but it's always been important 300 00:15:59,200 --> 00:16:01,400 to spend some time with people in the same place each year. 301 00:16:01,500 --> 00:16:04,900 You understand how they work, how they think, 302 00:16:05,000 --> 00:16:07,600 even to the point that, 303 00:16:07,600 --> 00:16:10,400 you know when they actually respond, whether they respond quickly or whether that means, 304 00:16:10,400 --> 00:16:14,100 they're not actually doing anything 305 00:16:14,100 --> 00:16:15,100 or that mean they're thinking hard about a problem 306 00:16:15,100 --> 00:16:17,300 and they just take take longer. 307 00:16:17,300 --> 00:16:20,200 but you do need to develop some understanding there. 308 00:16:20,200 --> 00:16:24,304 ♪ (music) ♪ 309 00:16:24,304 --> 00:16:25,900 We've talked about how your collaboration started, 310 00:16:26,900 --> 00:16:31,000 maybe just to step back slightly were they're sort of features about 311 00:16:31,000 --> 00:16:34,900 the environment at Harvard or in Cambridge, at the time, which you felt like contributed to it? 312 00:16:35,000 --> 00:16:37,400 Coming from Brown, 313 00:16:38,000 --> 00:16:42,100 I felt it was very intimidating place because it clearly was a very, very 314 00:16:43,500 --> 00:16:45,100 impressive set of people. 315 00:16:45,200 --> 00:16:48,200 Zvi Griliches was there, Dale Jorgensen-- 316 00:16:48,200 --> 00:16:49,200 Gary, Jerry Hausman, Whitney Newey, sometimes Jamie Robins. 317 00:16:52,600 --> 00:16:55,900 I mean, my view of that in retrospect, 318 00:16:55,900 --> 00:16:58,300 I can't say I loved every minute of every talk 319 00:16:58,300 --> 00:16:59,500 I ever gave in that Workshop, 320 00:16:59,500 --> 00:17:02,400 but that was the highest powered, that was the group you wanted to reach... 321 00:17:02,400 --> 00:17:03,400 (Guido) Yeah. 322 00:17:03,400 --> 00:17:04,900 and you would get 323 00:17:05,100 --> 00:17:10,600 extraordinarily insightful feedback, even if it wasn't always easy to swallow. 324 00:17:11,300 --> 00:17:12,940 Yeah, and I have for a while, 325 00:17:12,940 --> 00:17:16,200 I would basically give a talk every semester 326 00:17:16,200 --> 00:17:19,000 because we didn't have any money to be inviting people. 327 00:17:19,500 --> 00:17:22,000 Gary would say, "Well, why don't you give a talk?" 328 00:17:22,350 --> 00:17:23,350 (laughter) 329 00:17:26,800 --> 00:17:31,600 That was the arena for young people with our interest. 330 00:17:31,700 --> 00:17:34,700 (Guido) Yeah, it was really very impressive, 331 00:17:35,000 --> 00:17:36,600 but it was quite tough-- 332 00:17:36,700 --> 00:17:37,700 It was intimidating. 333 00:17:37,800 --> 00:17:41,000 People there had very strong views on what they thought was 334 00:17:41,200 --> 00:17:46,100 the way you should do econometrics, the way the direction things should go, 335 00:17:48,600 --> 00:17:53,300 now, I would think things were getting a little stale that in fact, 336 00:17:53,300 --> 00:17:56,000 we were bringing in a lot of the new ideas... 337 00:17:56,000 --> 00:17:57,000 (Josh) Yeah. 338 00:17:57,000 --> 00:18:01,900 ...and that wasn't necessary immediately appreciated. 339 00:18:02,800 --> 00:18:04,300 (Josh) But that's okay. - And that's fine. 340 00:18:04,300 --> 00:18:10,140 We were pushed and a lot of great discussions in that workshop about 341 00:18:11,250 --> 00:18:13,000 what should we make of late? 342 00:18:13,000 --> 00:18:15,800 But there were other questions that were just as interesting, 343 00:18:15,800 --> 00:18:18,000 like the role of the propensity score - 344 00:18:18,400 --> 00:18:19,600 that was a big deal in the 90s 345 00:18:19,700 --> 00:18:24,300 and econometrics was moving towards that 346 00:18:25,000 --> 00:18:27,800 and there were a lot of great questions. 347 00:18:27,900 --> 00:18:28,500 Yeah, 348 00:18:28,500 --> 00:18:33,300 I learned a huge amount there from the time I spent-- 349 00:18:33,300 --> 00:18:34,900 (Josh) I think the other thing that Guido and I 350 00:18:35,000 --> 00:18:36,900 both benefited from is we both, 351 00:18:37,400 --> 00:18:40,500 not at the same time, but in early in our careers, taught 352 00:18:40,800 --> 00:18:42,700 econometrics with Gary Chamberlain, 353 00:18:43,200 --> 00:18:46,500 and that was like an apprenticeship for us, I think. 354 00:18:46,800 --> 00:18:51,500 I taught a mixed graduate undergrad 1126, 355 00:18:51,500 --> 00:18:52,100 I don't know if they still have that number,... 356 00:18:52,500 --> 00:18:53,500 (Isaiah) Ahuh, they do. 357 00:18:53,900 --> 00:18:58,100 ...very interesting course that it had both graduate and undergraduate enrollment 358 00:18:58,800 --> 00:19:04,900 and it was relatively applied for an econometrics class, 359 00:19:05,000 --> 00:19:06,600 and I learned a lot by teaching that with Gary. 360 00:19:07,500 --> 00:19:10,100 But in that sense, Harvard was a great place, very flexible there. 361 00:19:13,600 --> 00:19:16,332 The other thing I remember about Harvard is, 362 00:19:16,710 --> 00:19:20,150 well I had very good students, 363 00:19:20,300 --> 00:19:25,100 I taught a lot of wonderful students who went on to have wonderful careers. 364 00:19:26,300 --> 00:19:31,750 Also, Harvard as an institution, you're probably are aware of this, Isaiah, 365 00:19:31,750 --> 00:19:34,800 as a junior faculty member, they didn't then ask much of us, 366 00:19:35,000 --> 00:19:37,300 other than teaching our classes. 367 00:19:37,800 --> 00:19:41,300 We didn't have administrative concerns, to speak of. 368 00:19:41,300 --> 00:19:45,300 I think I went to two faculty meetings in my two years at Harvard 369 00:19:46,600 --> 00:19:50,920 and so we're left-- 370 00:19:50,920 --> 00:19:53,400 You were given a lot of freedom and flexibility. 371 00:19:53,400 --> 00:19:58,100 I went to the chair said, "Can I teach this course with Rubin?" 372 00:19:59,000 --> 00:20:04,100 I think it was Friedman at the time. It was like, "Fine." 373 00:20:05,200 --> 00:20:10,100 It wasn't really any concern about what what it was about 374 00:20:10,700 --> 00:20:13,000 and again, that was a very intimidating experience, but it was a great experience. 375 00:20:13,350 --> 00:20:14,350 ♪ (music) ♪ 376 00:20:14,700 --> 00:20:17,600 (narrator) If you'd like to watch more Nobel Conversations, click here.