1 00:00:00,040 --> 00:00:02,460 The following content is provided under a Creative 2 00:00:02,460 --> 00:00:03,970 Commons license. 3 00:00:03,970 --> 00:00:06,910 Your support will help MIT OpenCourseWare continue to 4 00:00:06,910 --> 00:00:08,700 offer high quality, educational 5 00:00:08,700 --> 00:00:10,660 resources for free. 6 00:00:10,660 --> 00:00:13,460 To make a donation or view additional materials from 7 00:00:13,460 --> 00:00:17,390 hundreds of MIT courses, visit MIT OpenCourseWare at 8 00:00:17,390 --> 00:00:18,640 ocw.mit.edu. 9 00:00:25,630 --> 00:00:32,850 PROFESSOR: So where we left it last time is the access to 10 00:00:32,850 --> 00:00:35,860 family planning, the access to contraception, et cetera, did 11 00:00:35,860 --> 00:00:40,200 not play such an important role in keeping people's 12 00:00:40,200 --> 00:00:41,110 families small. 13 00:00:41,110 --> 00:00:43,130 It doesn't mean that it's not a good thing to do. 14 00:00:43,130 --> 00:00:45,295 It certainly helps people time their fertility. 15 00:00:45,295 --> 00:00:49,270 It makes them happier and all other good things, but it was 16 00:00:49,270 --> 00:00:52,880 not the main determinant of family size. 17 00:00:52,880 --> 00:00:56,280 And so then the next question, obviously, is then what is the 18 00:00:56,280 --> 00:00:58,530 determinant of family size? 19 00:00:58,530 --> 00:01:02,790 If people are not constrained by their lack of access to 20 00:01:02,790 --> 00:01:05,010 contraception to choose the family size that they want, 21 00:01:05,010 --> 00:01:08,670 how do they decide to choose their family size? 22 00:01:08,670 --> 00:01:14,610 And there are, I think, only two big classes of factors 23 00:01:14,610 --> 00:01:16,140 that might matter. 24 00:01:16,140 --> 00:01:22,280 One is social factors, so both what the community thinks is 25 00:01:22,280 --> 00:01:25,010 right, what the country thinks is right, what your husband 26 00:01:25,010 --> 00:01:26,810 thinks is right. 27 00:01:26,810 --> 00:01:33,550 And the other is the cost and benefits of having sex, which 28 00:01:33,550 --> 00:01:37,300 is pleasant, and having babies, which you might or 29 00:01:37,300 --> 00:01:40,620 might not want to have at a specific time of your life. 30 00:01:40,620 --> 00:01:44,420 So both of these things are traded off in some ways. 31 00:01:44,420 --> 00:01:47,310 And the issue is whether people do that effectively, 32 00:01:47,310 --> 00:01:53,040 whether they do that voluntarily, whether there is 33 00:01:53,040 --> 00:01:54,990 some correction coming from society or 34 00:01:54,990 --> 00:01:56,600 something like that. 35 00:01:56,600 --> 00:02:00,270 So social norms clearly play a role. 36 00:02:00,270 --> 00:02:03,190 So one example is this example of Matlab, in Bangladesh, 37 00:02:03,190 --> 00:02:06,500 where this program was introduced, and a community 38 00:02:06,500 --> 00:02:09,740 worker came and tried to persuade woman. 39 00:02:09,740 --> 00:02:14,170 One thing that has happened is that, at the beginning of the 40 00:02:14,170 --> 00:02:17,350 program, the introduction was very, very slow. 41 00:02:17,350 --> 00:02:20,080 The take-up of contraception was very, very slow. 42 00:02:20,080 --> 00:02:24,180 And then it started to explode, the take-up started 43 00:02:24,180 --> 00:02:25,340 to explode. 44 00:02:25,340 --> 00:02:28,090 And then it comes down again. 45 00:02:28,090 --> 00:02:33,320 So you get a little bit of an s-shape in adoption of the 46 00:02:33,320 --> 00:02:34,940 contraception. 47 00:02:34,940 --> 00:02:38,070 And what kind of mechanism would explain, what kind of 48 00:02:38,070 --> 00:02:41,500 social mechanism explains this kind of diffusion curve? 49 00:02:41,500 --> 00:02:43,960 You must have seen that, either in other classes or you 50 00:02:43,960 --> 00:02:45,210 can think about it. 51 00:02:47,574 --> 00:02:48,470 AUDIENCE: Viruses. 52 00:02:48,470 --> 00:02:49,140 PROFESSOR: Exactly. 53 00:02:49,140 --> 00:02:52,030 This is how viruses spread. 54 00:02:52,030 --> 00:02:57,940 This is also how iPads spread or any new innovation. 55 00:02:57,940 --> 00:03:01,600 And why does it spread in this way? 56 00:03:01,600 --> 00:03:04,301 Why does it have this shape? 57 00:03:04,301 --> 00:03:07,780 AUDIENCE: If it's contagious, initially, then a smaller 58 00:03:07,780 --> 00:03:10,265 group of people haven't. 59 00:03:10,265 --> 00:03:12,583 I guess that people that interact with each other at 60 00:03:12,583 --> 00:03:13,744 some point. 61 00:03:13,744 --> 00:03:16,974 I mean it kind of depends on something, I 62 00:03:16,974 --> 00:03:18,217 don't exactly know. 63 00:03:18,217 --> 00:03:20,720 I don't know the exact mechanism. 64 00:03:20,720 --> 00:03:21,215 PROFESSOR: Right. 65 00:03:21,215 --> 00:03:25,040 But you get the drift. 66 00:03:25,040 --> 00:03:30,130 It spreads mostly like viruses or new products, which spread 67 00:03:30,130 --> 00:03:32,555 mostly by infection. 68 00:03:35,090 --> 00:03:38,910 At the beginning, no one is infected, so the innovation or 69 00:03:38,910 --> 00:03:40,800 the virus doesn't spread. 70 00:03:40,800 --> 00:03:42,580 And then more and more people get infected. 71 00:03:42,580 --> 00:03:44,810 So they get chances to infect more people. 72 00:03:44,810 --> 00:03:47,090 So there is a sudden take off. 73 00:03:47,090 --> 00:03:49,950 And then after a while, everybody is infected. 74 00:03:49,950 --> 00:03:52,910 So the infection has nowhere to go anymore. 75 00:03:52,910 --> 00:03:56,530 So that's why we find this kind of mechanism, both for 76 00:03:56,530 --> 00:03:58,440 viruses and for new things. 77 00:04:02,665 --> 00:04:07,200 So the fact that in Matlab the contraception had the same 78 00:04:07,200 --> 00:04:11,430 spread, at first very slow and then a sudden take off and 79 00:04:11,430 --> 00:04:14,610 then flattening out again, when the take-up was larger, 80 00:04:14,610 --> 00:04:21,680 suggests there may have been a contagion phenomenon going on. 81 00:04:21,680 --> 00:04:26,490 So to look at it more directly, there's a paper by 82 00:04:26,490 --> 00:04:33,360 Kaivan Munshi, who looked at what happened, whether people 83 00:04:33,360 --> 00:04:37,050 do the same thing that their neighbors do. 84 00:04:37,050 --> 00:04:41,590 And so the first thing you can do is to say, well, if there 85 00:04:41,590 --> 00:04:45,350 are 20 women in my neighborhood and 19 have 86 00:04:45,350 --> 00:04:48,980 adopted the contraception, am I more likely to adopt then if 87 00:04:48,980 --> 00:04:51,270 there is 0 who have adopted? 88 00:04:51,270 --> 00:04:53,730 And when you do that, you find that indeed people, who had 89 00:04:53,730 --> 00:04:56,090 more people who adopted around them, were also 90 00:04:56,090 --> 00:04:57,690 more likely to adopt. 91 00:04:57,690 --> 00:05:01,670 So is that necessarily a sign that they're learning from 92 00:05:01,670 --> 00:05:02,340 each other? 93 00:05:02,340 --> 00:05:05,775 What else could be going on? 94 00:05:05,775 --> 00:05:07,659 AUDIENCE: It could just be a copycat effect. 95 00:05:07,659 --> 00:05:09,543 So it's not necessarily that you're learning, but you're 96 00:05:09,543 --> 00:05:11,130 just doing as you see someone else do. 97 00:05:11,130 --> 00:05:11,420 PROFESSOR: Exactly. 98 00:05:11,420 --> 00:05:13,600 So the learning could in a broad sense. 99 00:05:13,600 --> 00:05:15,370 First, it doesn't have to be learning something. 100 00:05:15,370 --> 00:05:17,310 It could just be pure imitation, where you're trying 101 00:05:17,310 --> 00:05:21,500 to figure out what's the fashionable thing to do. 102 00:05:21,500 --> 00:05:22,610 So there is no learning going on. 103 00:05:22,610 --> 00:05:25,190 That's right. 104 00:05:25,190 --> 00:05:27,410 So this is going to be difficult to separate, but 105 00:05:27,410 --> 00:05:28,250 even beyond that? 106 00:05:28,250 --> 00:05:28,735 Yeah? 107 00:05:28,735 --> 00:05:30,917 AUDIENCE: I guess if more people around you are doing 108 00:05:30,917 --> 00:05:33,585 it, there is greater demand in the area. 109 00:05:33,585 --> 00:05:35,040 [INAUDIBLE] 110 00:05:35,040 --> 00:05:37,480 supplies, but it's just inaccessible. 111 00:05:37,480 --> 00:05:37,560 PROFESSOR: Right. 112 00:05:37,560 --> 00:05:40,330 So there might be this effect, which is it might be that, in 113 00:05:40,330 --> 00:05:42,690 response to the first demand from the people, there is an 114 00:05:42,690 --> 00:05:43,930 excess in supply. 115 00:05:43,930 --> 00:05:46,920 Or it could even be that it turns out that the community 116 00:05:46,920 --> 00:05:54,800 health worker is better in that particular village. 117 00:05:54,800 --> 00:05:58,060 And so, because she's better and more energetic, she got 118 00:05:58,060 --> 00:06:00,860 your neighbors to adopt and, also, yourself to adopt. 119 00:06:00,860 --> 00:06:04,940 And the fact that you adopt when you neighbor adopts is 120 00:06:04,940 --> 00:06:08,450 not a sign that there is a causal effect of neighbor's 121 00:06:08,450 --> 00:06:11,340 adoption on your, which could be coming from copycat or 122 00:06:11,340 --> 00:06:14,370 coming from learning, but just that there is one common 123 00:06:14,370 --> 00:06:17,310 factor that leads them to adopt and leads you to adopt. 124 00:06:17,310 --> 00:06:21,150 So what he does is that he does something very ingenious. 125 00:06:21,150 --> 00:06:25,700 He exploits the fact that most of these villages have Hindus 126 00:06:25,700 --> 00:06:27,140 and Muslims. 127 00:06:27,140 --> 00:06:30,350 And there is only one community health worker for 128 00:06:30,350 --> 00:06:31,890 both communities. 129 00:06:31,890 --> 00:06:33,530 But Hindus don't speak to Muslims-- 130 00:06:33,530 --> 00:06:35,490 certainly not about contraception. 131 00:06:35,490 --> 00:06:37,970 And Muslims don't speak to Hindus. 132 00:06:37,970 --> 00:06:44,050 So now he's looking at, if it was the certain hypothesis 133 00:06:44,050 --> 00:06:47,420 that the reason why our actions are correlated is 134 00:06:47,420 --> 00:06:49,720 because of the community health worker. 135 00:06:49,720 --> 00:06:56,450 Should I see that, if I'm a Muslim woman, there is also a 136 00:06:56,450 --> 00:06:59,120 correlation between my adoption and that of the Hindu 137 00:06:59,120 --> 00:07:02,290 woman if it's due to the community health worker. 138 00:07:07,100 --> 00:07:08,140 AUDIENCE: Can you restate the question? 139 00:07:08,140 --> 00:07:10,640 PROFESSOR: So if it is due to the community health worker, 140 00:07:10,640 --> 00:07:13,160 should I see a relationship between the adoption of the 141 00:07:13,160 --> 00:07:16,392 Muslim woman and the adoption of Hindu woman? 142 00:07:16,392 --> 00:07:17,810 AUDIENCE: Yes. 143 00:07:17,810 --> 00:07:18,390 PROFESSOR: Yes. 144 00:07:18,390 --> 00:07:21,110 If it is due to the community health worker, because she's 145 00:07:21,110 --> 00:07:25,340 serving both communities, we should see this correlation. 146 00:07:25,340 --> 00:07:26,590 Sorry, one more came. 147 00:07:26,590 --> 00:07:28,112 Thank you. 148 00:07:28,112 --> 00:07:29,362 Good effort. 149 00:07:32,280 --> 00:07:36,100 But if it is due to the social learning, should we see a 150 00:07:36,100 --> 00:07:38,670 relationship between the adoption of the Hindu woman 151 00:07:38,670 --> 00:07:41,680 and adoption of the Muslim woman? 152 00:07:41,680 --> 00:07:42,470 No. 153 00:07:42,470 --> 00:07:45,360 So if it's due to the people learning from each other or 154 00:07:45,360 --> 00:07:48,060 even copying each other, they can't copy the Muslims, 155 00:07:48,060 --> 00:07:50,530 because they never will ask them if they are Hindu and 156 00:07:50,530 --> 00:07:51,460 vice versa. 157 00:07:51,460 --> 00:07:52,460 So he does that. 158 00:07:52,460 --> 00:07:55,860 And what he finds is that the correlation is only within 159 00:07:55,860 --> 00:07:58,920 community, which is an indication that, probably, 160 00:07:58,920 --> 00:08:01,385 people are trying to figure out what's the social norm 161 00:08:01,385 --> 00:08:05,370 around the social group or are copying each other. 162 00:08:05,370 --> 00:08:07,580 And it's just not a joint effect of the 163 00:08:07,580 --> 00:08:10,050 community health worker. 164 00:08:10,050 --> 00:08:13,220 Another way in which you would peak at the social norm is by 165 00:08:13,220 --> 00:08:15,490 watching what's on TV. 166 00:08:15,490 --> 00:08:21,120 And there is a very nice paper on the telenovelas in Brazil. 167 00:08:21,120 --> 00:08:28,640 So Brazil is a very Catholic country and, until relatively 168 00:08:28,640 --> 00:08:33,820 recently, was not a very free society, until about 169 00:08:33,820 --> 00:08:35,590 15, 20 years ago. 170 00:08:35,590 --> 00:08:38,390 And it was certainly not particularly encouraging 171 00:08:38,390 --> 00:08:40,030 family planning. 172 00:08:40,030 --> 00:08:46,130 But the telenovelas, which are like soap operas, became very, 173 00:08:46,130 --> 00:08:48,040 very popular. 174 00:08:48,040 --> 00:08:51,170 And they became very popular, in part, because they were 175 00:08:51,170 --> 00:08:52,700 quite good. 176 00:08:52,700 --> 00:08:54,950 And the reason why they were quite good-- 177 00:08:54,950 --> 00:08:59,060 I'm told, because I'm yet to watch one, my Portuguese being 178 00:08:59,060 --> 00:09:02,630 not quite up to crack-- 179 00:09:02,630 --> 00:09:06,890 is that a lot of people who would otherwise have been 180 00:09:06,890 --> 00:09:10,270 artists in other fields could not really do it very easily 181 00:09:10,270 --> 00:09:13,230 in Brazil, so they all went and did telenovelas. 182 00:09:13,230 --> 00:09:19,800 And they were aired on one channel, called Rede Globo, 183 00:09:19,800 --> 00:09:24,070 which was not available everywhere at the beginning, 184 00:09:24,070 --> 00:09:26,330 which progressively spread. 185 00:09:26,330 --> 00:09:29,840 So what this article did is that they looked at the 186 00:09:29,840 --> 00:09:35,160 diffusion of the Rede Globo all over Brazil and compared 187 00:09:35,160 --> 00:09:38,510 the number of children that a women had before and after the 188 00:09:38,510 --> 00:09:42,920 introduction of Rede Globo in places where it got 189 00:09:42,920 --> 00:09:43,910 introduced. 190 00:09:43,910 --> 00:09:46,260 So exploiting, in that way, the progressive introduction. 191 00:09:46,260 --> 00:09:48,670 Of course, the cities got it first and then 192 00:09:48,670 --> 00:09:49,830 the more rural areas. 193 00:09:49,830 --> 00:09:51,930 So places that got it first, in 194 00:09:51,930 --> 00:09:53,230 general, had fewer children. 195 00:09:53,230 --> 00:09:56,770 But if we compare before and after you get it, in places 196 00:09:56,770 --> 00:09:59,450 that get it and places that don't, then we get a sense 197 00:09:59,450 --> 00:10:01,840 about what is the effect of Rede Globo. 198 00:10:01,840 --> 00:10:07,040 And what they found is that the women, who were of 199 00:10:07,040 --> 00:10:10,320 reproductive age after Rede Globo was introduced, had 200 00:10:10,320 --> 00:10:13,700 fewer children in their lifetime then the places that 201 00:10:13,700 --> 00:10:17,090 were in reproductive age before Rede Globo came. 202 00:10:17,090 --> 00:10:20,230 And another thing that was kind of cute, as a result of 203 00:10:20,230 --> 00:10:26,030 that, is that many of the kids that were born in this area 204 00:10:26,030 --> 00:10:30,160 were born after the characters of the telenovelas. 205 00:10:30,160 --> 00:10:30,907 Yeah? 206 00:10:30,907 --> 00:10:33,769 AUDIENCE: [INAUDIBLE] 207 00:10:33,769 --> 00:10:38,000 these telenovelas, the main characters have few children? 208 00:10:38,000 --> 00:10:38,790 PROFESSOR: Yes. 209 00:10:38,790 --> 00:10:39,215 Exactly. 210 00:10:39,215 --> 00:10:44,930 The reason why this happened is that the telenovelas were 211 00:10:44,930 --> 00:10:49,520 promoting, not necessarily for the sake of it, but the main 212 00:10:49,520 --> 00:10:53,610 characters in the telenovelas were people, some of them with 213 00:10:53,610 --> 00:10:57,720 0 children, if they had children, with a small family. 214 00:10:57,720 --> 00:11:00,770 So they were promoting a model of like a modern woman who had 215 00:11:00,770 --> 00:11:05,560 few children, which kind of helped or contributed to 216 00:11:05,560 --> 00:11:06,710 slowly shift the norm. 217 00:11:06,710 --> 00:11:09,460 Yes, Brazil expertise. 218 00:11:09,460 --> 00:11:13,300 AUDIENCE: So it was is like the main channel. 219 00:11:13,300 --> 00:11:15,200 Like if there is one that people actually don't have to 220 00:11:15,200 --> 00:11:17,852 pay, necessarily, for it, you just get it for free, if you 221 00:11:17,852 --> 00:11:23,350 just have a TV So part of the access to that channel comes 222 00:11:23,350 --> 00:11:28,440 with some areas that are growing into big cities, 223 00:11:28,440 --> 00:11:30,722 becoming a little bit more industrialized. 224 00:11:30,722 --> 00:11:33,870 So that could be another thing, stuff like that. 225 00:11:33,870 --> 00:11:40,350 People are getting access to the TV channel, that means 226 00:11:40,350 --> 00:11:43,388 that the country as a whole is becoming more and more 227 00:11:43,388 --> 00:11:44,384 industrialized. 228 00:11:44,384 --> 00:11:49,613 And they are sort of adapting how many kids they have to fit 229 00:11:49,613 --> 00:11:51,870 the model a little bit better. 230 00:11:51,870 --> 00:11:52,280 PROFESSOR: Right. 231 00:11:52,280 --> 00:11:54,110 AUDIENCE: So the actual cause could be different. 232 00:11:54,110 --> 00:11:59,200 Maybe it's not just the fact that they were watching TV, 233 00:11:59,200 --> 00:12:01,510 but the fact that they gained access to it 234 00:12:01,510 --> 00:12:03,060 and what that implies. 235 00:12:03,060 --> 00:12:05,450 PROFESSOR: So what you are saying is that TV came on the 236 00:12:05,450 --> 00:12:08,060 tail of other changes in the same areas? 237 00:12:08,060 --> 00:12:09,310 AUDIENCE: Right. 238 00:12:12,402 --> 00:12:16,114 Some places becoming a little bit more urban 239 00:12:16,114 --> 00:12:19,070 and how that's changed. 240 00:12:19,070 --> 00:12:19,370 PROFESSOR: Yeah. 241 00:12:19,370 --> 00:12:22,100 So that's a very good point, which is always with this 242 00:12:22,100 --> 00:12:24,590 different and difference kind of strategy. 243 00:12:24,590 --> 00:12:28,160 The always are worried about that. 244 00:12:28,160 --> 00:12:31,040 At least we are not comparing directly the city to the rural 245 00:12:31,040 --> 00:12:34,440 areas, but we are comparing the rural areas as they are 246 00:12:34,440 --> 00:12:37,940 becoming industrial enough or developed 247 00:12:37,940 --> 00:12:39,590 enough to get the channel. 248 00:12:39,590 --> 00:12:43,600 Because, presumably, the channel decided to come when 249 00:12:43,600 --> 00:12:45,770 there was enough people who were rich enough to buy a 250 00:12:45,770 --> 00:12:46,580 television. 251 00:12:46,580 --> 00:12:50,700 And therefore, they may have been looking, not only at the 252 00:12:50,700 --> 00:12:55,410 level of development of the place, but how it was moving. 253 00:12:55,410 --> 00:12:58,840 And if that is the case, then we could be picking part of 254 00:12:58,840 --> 00:12:59,830 those trends. 255 00:12:59,830 --> 00:13:03,380 So that is something where we could never be completely 256 00:13:03,380 --> 00:13:05,520 sure, because we haven't randomized. 257 00:13:05,520 --> 00:13:10,590 So those underlying trends, we don't observe them. 258 00:13:10,590 --> 00:13:13,350 And so it's quite possible that you are right. 259 00:13:13,350 --> 00:13:17,780 So these doubts remain. 260 00:13:17,780 --> 00:13:22,150 What the author does is try to look at whether there are 261 00:13:22,150 --> 00:13:30,450 trends in fertility before, that the fertility was already 262 00:13:30,450 --> 00:13:34,960 a little bit declining before the channel came in. 263 00:13:34,960 --> 00:13:38,000 Which we would probably observe, if the stories you're 264 00:13:38,000 --> 00:13:42,550 talking about is the one, which is, as I see that the 265 00:13:42,550 --> 00:13:44,590 place is becoming more industrialized, richer et 266 00:13:44,590 --> 00:13:48,710 cetera, I'm putting the TV there, as the president of 267 00:13:48,710 --> 00:13:49,920 Rede Globo. 268 00:13:49,920 --> 00:13:51,400 And they don't see that. 269 00:13:51,400 --> 00:13:54,950 They see kind of shift in trend around the introduction 270 00:13:54,950 --> 00:13:55,900 of the television. 271 00:13:55,900 --> 00:13:57,420 But you might be entirely right. 272 00:13:57,420 --> 00:14:01,150 That's also why I like the first name thing, that the 273 00:14:01,150 --> 00:14:04,920 kids were named after the people in the series. 274 00:14:04,920 --> 00:14:08,320 Because that is less likely to becoming from 275 00:14:08,320 --> 00:14:11,360 industrialization in general and more likely to be coming 276 00:14:11,360 --> 00:14:13,390 from the fact that you've seen them. 277 00:14:13,390 --> 00:14:16,056 But that's very well taken. 278 00:14:16,056 --> 00:14:19,014 AUDIENCE: Did they observe a lag? 279 00:14:19,014 --> 00:14:22,465 Like the introduction of the TV show and the shift in 280 00:14:22,465 --> 00:14:22,970 [INAUDIBLE]. 281 00:14:22,970 --> 00:14:23,640 PROFESSOR: Right. 282 00:14:23,640 --> 00:14:23,790 Exactly. 283 00:14:23,790 --> 00:14:26,000 So that's exactly what they are exploiting. 284 00:14:26,000 --> 00:14:29,270 They are trying to look at, particularly, it doesn't 285 00:14:29,270 --> 00:14:34,700 reduce the fertility of people who were already past their 286 00:14:34,700 --> 00:14:37,530 prime fertile age when the TV came. 287 00:14:37,530 --> 00:14:41,370 It reduces only the fertility of women who were in their 288 00:14:41,370 --> 00:14:45,750 late teens and then early 20s when the TV came, and whose 289 00:14:45,750 --> 00:14:48,860 views were shaped by this. 290 00:14:48,860 --> 00:14:52,550 And this is exactly this timing that helps, to some 291 00:14:52,550 --> 00:14:56,220 extent, discriminate between the idea that the TV caused 292 00:14:56,220 --> 00:14:59,520 the reduction of the fertility and your perfectly legitimate 293 00:14:59,520 --> 00:15:02,990 point that, in fact, in part it could also be that TV is 294 00:15:02,990 --> 00:15:06,112 just a proxy for the social progress happening. 295 00:15:10,540 --> 00:15:13,740 So these are all indications that something, which is 296 00:15:13,740 --> 00:15:16,760 probably not particularly surprising, is in something as 297 00:15:16,760 --> 00:15:19,160 fundamental and deep as how many 298 00:15:19,160 --> 00:15:22,460 children you want to have. 299 00:15:22,460 --> 00:15:24,460 The social norms around you-- 300 00:15:24,460 --> 00:15:27,500 these are very local, your village, your community, your 301 00:15:27,500 --> 00:15:32,760 family, your mother-in-law, or society as a whole, as 302 00:15:32,760 --> 00:15:36,470 reflected to you by TV-- 303 00:15:36,470 --> 00:15:37,910 matter. 304 00:15:37,910 --> 00:15:44,690 But there are also just plain, simple, cost-benefit 305 00:15:44,690 --> 00:15:49,490 calculations about do I want a kid at this time, do I want to 306 00:15:49,490 --> 00:15:52,230 take the risk of a kid at this time by having unprotected 307 00:15:52,230 --> 00:15:54,140 sex, et cetera, et cetera. 308 00:15:54,140 --> 00:15:58,100 And one thing that I want to argue is that people are very 309 00:15:58,100 --> 00:16:01,290 good at making those calculations. 310 00:16:01,290 --> 00:16:08,970 And in the sense the key group to look at for a test case of 311 00:16:08,970 --> 00:16:11,200 this are adolescent girls. 312 00:16:11,200 --> 00:16:14,500 Because in these countries and elsewhere, we kind of suspect 313 00:16:14,500 --> 00:16:17,770 adolescent girls to be not the best at making those 314 00:16:17,770 --> 00:16:20,560 calculations, necessarily, which is why we see some 315 00:16:20,560 --> 00:16:22,510 teenage pregnancies. 316 00:16:22,510 --> 00:16:26,520 And so if they can calculate, then it would mean that, 317 00:16:26,520 --> 00:16:28,330 presumably, people can calculate them. 318 00:16:28,330 --> 00:16:31,440 Another reason why it's very interesting to look at 319 00:16:31,440 --> 00:16:35,030 adolescents, in particular, in Africa, is not only teenage 320 00:16:35,030 --> 00:16:37,240 pregnancy, but teenage pregnancy as a sign that 321 00:16:37,240 --> 00:16:43,420 you've had unprotected sex, which puts you at risk for a 322 00:16:43,420 --> 00:16:47,920 STDs, sexually transmitted diseases, in particular, HIV. 323 00:16:47,920 --> 00:16:52,620 So adolescents have been called the window of hope in 324 00:16:52,620 --> 00:16:57,720 the fight against HIV, because A, they are uninfected. 325 00:16:57,720 --> 00:17:01,660 And once you're infected with HIV, you're infected. 326 00:17:01,660 --> 00:17:03,980 You can't disinfect the people, so it's better to 327 00:17:03,980 --> 00:17:05,520 catch them early. 328 00:17:05,520 --> 00:17:07,560 Second-- 329 00:17:07,560 --> 00:17:10,069 and that goes back to the point that you made earlier 330 00:17:10,069 --> 00:17:11,319 about was there a lag-- 331 00:17:13,980 --> 00:17:17,650 people's sexual preferences, their habits, et cetera, have 332 00:17:17,650 --> 00:17:18,660 not yet been formed. 333 00:17:18,660 --> 00:17:24,619 And maybe it's right time to catch them to influence their 334 00:17:24,619 --> 00:17:26,829 behavior early, and then they'll keep those patterns 335 00:17:26,829 --> 00:17:28,720 for the rest of their life. 336 00:17:28,720 --> 00:17:32,770 So there is a lot of effort, the world over, to develop 337 00:17:32,770 --> 00:17:45,410 programs to try to teach kids to have a safer sex life. 338 00:17:45,410 --> 00:17:48,250 And UNICEF worked on a lot of these programs. 339 00:17:48,250 --> 00:17:52,480 But unfortunately, in a lot of countries, there is some 340 00:17:52,480 --> 00:17:56,920 pressure to have this program being mainly concerned about 341 00:17:56,920 --> 00:17:59,280 abstinence and faithfulness. 342 00:17:59,280 --> 00:18:02,890 So the strategy that is employed in eastern South 343 00:18:02,890 --> 00:18:05,730 Africa is called ABCD. 344 00:18:05,730 --> 00:18:11,350 So the A is for abstain, B for be faithful, C is for use a 345 00:18:11,350 --> 00:18:15,630 condom, and D is or you die. 346 00:18:15,630 --> 00:18:23,710 So that kind of encapsulates the HIV prevention strategy. 347 00:18:23,710 --> 00:18:27,850 But in primary schools, and primary school in Kenya goes 348 00:18:27,850 --> 00:18:30,310 till grade eight, and people start school late 349 00:18:30,310 --> 00:18:31,520 and repeat a lot. 350 00:18:31,520 --> 00:18:35,103 So really, kids are like 16, 17, they're 351 00:18:35,103 --> 00:18:36,600 still in primary school. 352 00:18:36,600 --> 00:18:40,380 So think of primary and junior high. 353 00:18:40,380 --> 00:18:43,035 There was a policy decision that was made in Kenya, which 354 00:18:43,035 --> 00:18:46,140 is still a pretty religious country, that condoms could 355 00:18:46,140 --> 00:18:48,750 not be discussed in school. 356 00:18:48,750 --> 00:18:53,740 So the ABCD program is really an ABD program, 357 00:18:53,740 --> 00:18:55,060 abstain or be faithful. 358 00:18:55,060 --> 00:18:59,440 And, in fact, it's a program, which is all day long, to 359 00:18:59,440 --> 00:19:02,910 scare kids away and to say, look everybody's dangerous. 360 00:19:02,910 --> 00:19:04,750 HIV is everywhere. 361 00:19:04,750 --> 00:19:07,900 The most dangerous kid is the kid you meet in a disco. 362 00:19:07,900 --> 00:19:09,110 I'm not making any of this up. 363 00:19:09,110 --> 00:19:13,160 I've read the little UNICEF books, where you have this 364 00:19:13,160 --> 00:19:16,400 little cartoon of a girl going to a disco and meeting this 365 00:19:16,400 --> 00:19:19,410 boy and having sex and then getting HIV. 366 00:19:19,410 --> 00:19:25,060 So this is the type of thing that they are trying to put 367 00:19:25,060 --> 00:19:27,970 into people's heads. 368 00:19:27,970 --> 00:19:34,140 So basically, the message becomes, abstain until you are 369 00:19:34,140 --> 00:19:37,310 of age to get married and then get married. 370 00:19:37,310 --> 00:19:38,960 Now the big danger with that is that kids 371 00:19:38,960 --> 00:19:40,580 decide, well, fine. 372 00:19:40,580 --> 00:19:41,910 I don't want to die. 373 00:19:41,910 --> 00:19:43,450 But I don't want to abstain. 374 00:19:43,450 --> 00:19:45,490 So let me get married. 375 00:19:45,490 --> 00:19:49,670 And the reason why it's so dangerous is that means that 376 00:19:49,670 --> 00:19:51,720 kids could get married early. 377 00:19:51,720 --> 00:19:55,350 For girls, it means they would have children early. 378 00:19:55,350 --> 00:19:58,750 And so we would have very early pregnancies, which means 379 00:19:58,750 --> 00:20:00,550 that they would interrupt their schooling, which means 380 00:20:00,550 --> 00:20:02,870 that they would be, typically, quite a bit younger than their 381 00:20:02,870 --> 00:20:09,330 husbands, which changes their status in the family, as we 382 00:20:09,330 --> 00:20:11,940 are going to see in a minute, et cetera. 383 00:20:11,940 --> 00:20:18,330 So we decided to try and test the impact of the ABCD program 384 00:20:18,330 --> 00:20:26,090 against doing nothing and against another strategy, 385 00:20:26,090 --> 00:20:29,860 which is to basically make it a little bit easier for kids 386 00:20:29,860 --> 00:20:31,470 to stay in school. 387 00:20:31,470 --> 00:20:33,220 How do you make it a bit easier for 388 00:20:33,220 --> 00:20:33,630 kids to stay in school? 389 00:20:33,630 --> 00:20:36,870 You just pay the cost of staying in school, which now 390 00:20:36,870 --> 00:20:40,090 is just like the uniform. 391 00:20:40,090 --> 00:20:43,760 So we ran an experiment with Pascaline Dupas and Michael 392 00:20:43,760 --> 00:20:48,380 Kremer, a pretty big experiment, 376 schools, which 393 00:20:48,380 --> 00:20:50,810 were assigned to one of four groups. 394 00:20:50,810 --> 00:20:52,270 One was a control group. 395 00:20:52,270 --> 00:20:54,300 One was a uniform group. 396 00:20:54,300 --> 00:20:58,290 And the other thing we did is we wanted to test the ABCD 397 00:20:58,290 --> 00:21:02,400 strategy, so we accelerated the training of teachers. 398 00:21:02,400 --> 00:21:06,120 Because, in principle, every school is supposed to 399 00:21:06,120 --> 00:21:09,300 experience the ABCD strategy, but, in practice, the teachers 400 00:21:09,300 --> 00:21:11,630 are very shy about talking about HIV. 401 00:21:11,630 --> 00:21:14,620 They don't want to be branded as an HIV teacher. 402 00:21:14,620 --> 00:21:17,280 People are still pretty prudish about the whole thing. 403 00:21:17,280 --> 00:21:21,320 So teachers won't do anything about HIV AIDS prevention 404 00:21:21,320 --> 00:21:23,640 until they get training to do it. 405 00:21:23,640 --> 00:21:25,890 So the government has a team of trainers. 406 00:21:25,890 --> 00:21:27,450 And they went, and they trained the people. 407 00:21:27,450 --> 00:21:31,090 And that way the teacher started to teach. 408 00:21:31,090 --> 00:21:32,060 So we had four groups. 409 00:21:32,060 --> 00:21:35,430 One, the teachers got trained. 410 00:21:35,430 --> 00:21:39,220 One, the kids got a free uniform. 411 00:21:39,220 --> 00:21:42,090 And in one group, they got both. 412 00:21:42,090 --> 00:21:44,780 And these are some of our girls. 413 00:21:44,780 --> 00:21:48,330 And then we followed the kids, all the way, 414 00:21:48,330 --> 00:21:50,830 from 2002 to 2011. 415 00:21:50,830 --> 00:21:52,490 So what I'm going to show you is really the 416 00:21:52,490 --> 00:21:54,940 results that are coming. 417 00:21:54,940 --> 00:21:57,070 The results are as of yesterday. 418 00:21:57,070 --> 00:22:01,190 So you get like prime news. 419 00:22:01,190 --> 00:22:04,410 Not yesterday, because I had the slides on two days ago, 420 00:22:04,410 --> 00:22:07,250 but like less than a week ago. 421 00:22:07,250 --> 00:22:09,630 And what do we find? 422 00:22:09,630 --> 00:22:13,210 So this are simple means. 423 00:22:13,210 --> 00:22:16,050 And I tell you what is significant in the means and 424 00:22:16,050 --> 00:22:17,200 what's not. 425 00:22:17,200 --> 00:22:21,030 This is your probability to complete primary school in the 426 00:22:21,030 --> 00:22:22,280 four groups. 427 00:22:26,260 --> 00:22:29,120 So what is not surprising in this graph? 428 00:22:29,120 --> 00:22:29,820 Can you read? 429 00:22:29,820 --> 00:22:32,800 You have the control group, uniform, teacher training, and 430 00:22:32,800 --> 00:22:35,800 then both is the uniform and teacher training. 431 00:22:35,800 --> 00:22:38,630 So what is not particularly surprising in this graph? 432 00:22:38,630 --> 00:22:41,510 AUDIENCE: They all exceed the control group. 433 00:22:41,510 --> 00:22:43,330 PROFESSOR: That they all exceed the control group. 434 00:22:43,330 --> 00:22:45,842 And, in particular? 435 00:22:45,842 --> 00:22:47,092 AUDIENCE: [INAUDIBLE]? 436 00:22:49,188 --> 00:22:51,220 PROFESSOR: Both exceed just teacher training. 437 00:22:51,220 --> 00:22:52,470 And? 438 00:22:55,684 --> 00:22:58,660 AUDIENCE: That both is-- oh, sorry. 439 00:22:58,660 --> 00:23:00,160 I was going to say what's surprising. 440 00:23:00,160 --> 00:23:01,850 PROFESSOR: Yeah, what is surprising is fine, too. 441 00:23:01,850 --> 00:23:05,260 AUDIENCE: It's the fact that both is less than uniforms. 442 00:23:05,260 --> 00:23:05,710 PROFESSOR: Exactly. 443 00:23:05,710 --> 00:23:08,740 So what is not surprising is uniform is much more than 444 00:23:08,740 --> 00:23:09,450 everybody else. 445 00:23:09,450 --> 00:23:11,920 In fact, control and teacher training are 446 00:23:11,920 --> 00:23:15,170 pretty much the same. 447 00:23:15,170 --> 00:23:17,640 I mean, they are not statistically different. 448 00:23:17,640 --> 00:23:20,150 Uniform is much more. 449 00:23:20,150 --> 00:23:23,140 I mean, you have to take care of the scales, here. 450 00:23:23,140 --> 00:23:24,390 These are not huge effects. 451 00:23:24,390 --> 00:23:27,870 But getting a uniform, which costs like a few dollars, 452 00:23:27,870 --> 00:23:30,850 really nothing, increased completion rates of primary 453 00:23:30,850 --> 00:23:33,140 school by about 6%. 454 00:23:36,590 --> 00:23:41,390 What is surprising is that both is less than uniform. 455 00:23:41,390 --> 00:23:43,680 You would think that it would at least be just the same, 456 00:23:43,680 --> 00:23:44,900 because at least they have a uniform. 457 00:23:44,900 --> 00:23:47,740 So why is both less than uniform? 458 00:23:47,740 --> 00:23:51,360 Not to answer the question, we are going to continue and see 459 00:23:51,360 --> 00:23:54,120 what else we're seeing in this data and then come back to try 460 00:23:54,120 --> 00:23:57,100 and figure out what's happening. 461 00:23:57,100 --> 00:24:00,270 That's a second thing which is happening is whether or not 462 00:24:00,270 --> 00:24:02,700 you are ever pregnant by age 17. 463 00:24:02,700 --> 00:24:05,140 It's like, if you're ever pregnant or pregnant by age 464 00:24:05,140 --> 00:24:08,670 17, it will give you the same result. 465 00:24:08,670 --> 00:24:10,990 So what is not surprising here? 466 00:24:15,670 --> 00:24:16,865 We can just describe the results. 467 00:24:16,865 --> 00:24:18,052 Yeah? 468 00:24:18,052 --> 00:24:20,971 AUDIENCE: The surprising part is that teacher training is 469 00:24:20,971 --> 00:24:22,210 the highest. 470 00:24:22,210 --> 00:24:24,140 PROFESSOR: Yeah, so the quite surprising part is that 471 00:24:24,140 --> 00:24:25,610 teacher training is the highest. 472 00:24:25,610 --> 00:24:28,570 So remember, this is an abstinence program. 473 00:24:28,570 --> 00:24:31,610 The abstinence program is leading to a slight increase 474 00:24:31,610 --> 00:24:33,420 in the number of pregnancies. 475 00:24:33,420 --> 00:24:35,280 That is not the sign of a successful program. 476 00:24:38,800 --> 00:24:41,690 Which is the program that seems to be doing the best? 477 00:24:41,690 --> 00:24:42,640 AUDIENCE: Uniforms. 478 00:24:42,640 --> 00:24:44,290 PROFESSOR: The uniform program is great. 479 00:24:44,290 --> 00:24:46,540 But what do we have, again, that is surprising, here? 480 00:24:50,080 --> 00:24:51,920 That both is above. 481 00:24:51,920 --> 00:24:55,400 So not only the teacher training does nothing, but it 482 00:24:55,400 --> 00:24:59,960 undoes some positive effect of the uniform program, both in 483 00:24:59,960 --> 00:25:04,520 terms of school attendance and in terms of teenage pregnancy. 484 00:25:04,520 --> 00:25:11,800 So now let's look at the STD infection. 485 00:25:11,800 --> 00:25:14,260 So we tested for HIV. 486 00:25:14,260 --> 00:25:16,520 And we tested for HSV2. 487 00:25:16,520 --> 00:25:19,070 HSV2 is a form of herpes. 488 00:25:19,070 --> 00:25:20,250 It's less bad than HIV. 489 00:25:20,250 --> 00:25:21,860 It's not going to kill you. 490 00:25:21,860 --> 00:25:25,250 But it has the same characteristic, which is once 491 00:25:25,250 --> 00:25:28,150 you have a biomarker for it in your body, it never goes away. 492 00:25:28,150 --> 00:25:32,225 You have the antibody for it, so we know that you've had it 493 00:25:32,225 --> 00:25:33,740 at some point in your life. 494 00:25:33,740 --> 00:25:35,280 It's also good to diagnose it, because 495 00:25:35,280 --> 00:25:37,540 it needs to be treated. 496 00:25:37,540 --> 00:25:42,060 It's also bad to have HSV2, because, as other sexually 497 00:25:42,060 --> 00:25:46,190 transmitted diseases, it greatly increases the chance 498 00:25:46,190 --> 00:25:48,520 that you get HIV. 499 00:25:48,520 --> 00:25:51,585 The transmission rate of HIV, if you have sex with an 500 00:25:51,585 --> 00:25:55,880 infected partner, is higher if you have another STD. 501 00:25:55,880 --> 00:25:59,280 Because there are lesions in your body that the 502 00:25:59,280 --> 00:26:02,800 virus uses to get in. 503 00:26:02,800 --> 00:26:05,390 So what do we find with HSV2? 504 00:26:05,390 --> 00:26:08,450 So the reason we don't use HIV is that one of the great news 505 00:26:08,450 --> 00:26:10,740 of this project is no one has HIV. 506 00:26:10,740 --> 00:26:14,420 We have about 1% of HIV infection. 507 00:26:14,420 --> 00:26:16,600 So that's very, very low, across all groups. 508 00:26:16,600 --> 00:26:17,790 So that's good. 509 00:26:17,790 --> 00:26:20,750 But there is much more of HSV2. 510 00:26:20,750 --> 00:26:24,090 It's about 9% in the control group. 511 00:26:24,090 --> 00:26:29,180 And here, what do we see, with this graph? 512 00:26:29,180 --> 00:26:31,470 So that is the result that is brand, brand new. 513 00:26:31,470 --> 00:26:34,190 The other results, we already had. 514 00:26:34,190 --> 00:26:35,800 That's the reason that this is completely new. 515 00:26:39,237 --> 00:26:41,692 AUDIENCE: If they had a uniform, it makes you more 516 00:26:41,692 --> 00:26:43,170 likely to [INAUDIBLE]. 517 00:26:43,170 --> 00:26:45,180 PROFESSOR: Somewhat more likely, yes. 518 00:26:45,180 --> 00:26:48,840 So you can call it the same, that both uniform and teacher 519 00:26:48,840 --> 00:26:50,140 training don't help. 520 00:26:53,620 --> 00:26:56,680 Those are all pretty similar in terms of standard error. 521 00:26:56,680 --> 00:26:58,090 But it doesn't help. 522 00:26:58,090 --> 00:27:01,300 So having a uniform makes you stay in school longer. 523 00:27:01,300 --> 00:27:05,360 It makes you less likely to a teenage pregnancy. 524 00:27:05,360 --> 00:27:08,410 But you're not less likely to have 525 00:27:08,410 --> 00:27:10,860 sexually transmitted diseases. 526 00:27:10,860 --> 00:27:14,540 Teacher training, again, is not doing anything good. 527 00:27:14,540 --> 00:27:17,840 And then what now makes a difference is the both 528 00:27:17,840 --> 00:27:23,050 program, where, if you have the two together, you're 529 00:27:23,050 --> 00:27:27,170 significantly less likely to be infected with HSV2. 530 00:27:27,170 --> 00:27:30,710 So what do you think is happening? 531 00:27:35,040 --> 00:27:38,890 Any kind of story you can try? 532 00:27:38,890 --> 00:27:41,890 AUDIENCE: [INAUDIBLE]. 533 00:27:41,890 --> 00:27:44,310 PROFESSOR: Yeah, but when we wrote the book, 534 00:27:44,310 --> 00:27:45,960 I didn't know this. 535 00:27:45,960 --> 00:27:48,430 So you will have to make one more. 536 00:27:48,430 --> 00:27:52,160 We can start with that, then you can retell the 537 00:27:52,160 --> 00:27:53,030 story of the book. 538 00:27:53,030 --> 00:27:57,088 And then you can say why that explains that as well. 539 00:27:57,088 --> 00:27:59,528 AUDIENCE: So my understanding was that the uniforms kept the 540 00:27:59,528 --> 00:28:00,992 students in school. 541 00:28:00,992 --> 00:28:05,872 And as a result, they had less time to do other things, like 542 00:28:05,872 --> 00:28:07,010 have sex, which could 543 00:28:07,010 --> 00:28:08,800 potentially give them diseases. 544 00:28:08,800 --> 00:28:12,674 So it makes sense to me why both would be more effective 545 00:28:12,674 --> 00:28:17,514 in reducing how many people were HSV2 positive, because 546 00:28:17,514 --> 00:28:20,418 you would be more likely to be in school and then you'd be 547 00:28:20,418 --> 00:28:22,838 getting the training from the teacher in respect what they 548 00:28:22,838 --> 00:28:26,720 should not be doing in order to avoid that. 549 00:28:26,720 --> 00:28:29,010 PROFESSOR: So that is possible. 550 00:28:29,010 --> 00:28:31,890 So Ben's idea is that, oh, you spend more time school. 551 00:28:31,890 --> 00:28:33,865 And therefore, you have more time to learn what the teacher 552 00:28:33,865 --> 00:28:34,670 is telling you. 553 00:28:34,670 --> 00:28:36,520 And therefore, that should be the most effective. 554 00:28:36,520 --> 00:28:39,660 But if that was the story, we should see all of them going 555 00:28:39,660 --> 00:28:44,060 down a little, and both going down the most, right? 556 00:28:44,060 --> 00:28:45,360 So there is something else. 557 00:28:45,360 --> 00:28:46,343 Yeah? 558 00:28:46,343 --> 00:28:49,241 AUDIENCE: Because it has to do with getting married. 559 00:28:49,241 --> 00:28:52,138 So it still kind of goes along with school, where I think the 560 00:28:52,138 --> 00:28:55,332 people with the uniforms seem to be able to stay in school. 561 00:28:55,332 --> 00:28:56,769 And so they don't get pregnant. 562 00:28:56,769 --> 00:28:58,206 Because I assume the pregnancy would come from 563 00:28:58,206 --> 00:28:59,170 when they get married. 564 00:28:59,170 --> 00:29:01,294 But they're probably, maybe having safe sex. 565 00:29:01,294 --> 00:29:03,258 Or, I guess, unsafe sex. 566 00:29:03,258 --> 00:29:05,713 PROFESSOR: Unsafe sex, yeah. 567 00:29:05,713 --> 00:29:08,659 AUDIENCE: They're still having sex, so they're still getting 568 00:29:08,659 --> 00:29:10,458 diseases, but they're not getting pregnant and dropping 569 00:29:10,458 --> 00:29:11,114 out of school. 570 00:29:11,114 --> 00:29:16,515 Whereas with both, it may be a category where 571 00:29:16,515 --> 00:29:17,988 they're getting married. 572 00:29:17,988 --> 00:29:19,952 And so now they only have one partner. 573 00:29:19,952 --> 00:29:21,916 So they're less likely to get diseases 574 00:29:21,916 --> 00:29:23,430 but not getting pregnant. 575 00:29:23,430 --> 00:29:24,463 PROFESSOR: Exactly. 576 00:29:24,463 --> 00:29:28,930 AUDIENCE: For this particular result, are you tracking the 577 00:29:28,930 --> 00:29:31,010 students who actually dropped out and got married? 578 00:29:31,010 --> 00:29:31,660 PROFESSOR: Yeah. 579 00:29:31,660 --> 00:29:34,920 We tracked everyone. 580 00:29:34,920 --> 00:29:39,530 That's why it took us forever and something like $2 million. 581 00:29:39,530 --> 00:29:48,000 But we attempted to track all of the cohort that was in the 582 00:29:48,000 --> 00:29:50,150 initial program. 583 00:29:50,150 --> 00:29:52,140 Of course, a lot of them have moved. 584 00:29:52,140 --> 00:29:53,820 So if they went to Nairobi, we went to 585 00:29:53,820 --> 00:29:54,900 Nairobi to look for them. 586 00:29:54,900 --> 00:29:56,230 If they went to Uganda, we went to 587 00:29:56,230 --> 00:29:57,680 Uganda to look for them. 588 00:29:57,680 --> 00:30:02,990 So the first wave of them, we managed to find about 60%. 589 00:30:02,990 --> 00:30:08,080 And then in the remaining 40% of them, we really put all of 590 00:30:08,080 --> 00:30:11,840 the means to find them and found about 80% of them. 591 00:30:11,840 --> 00:30:15,750 And so the effective tracking rate was about 85%. 592 00:30:15,750 --> 00:30:18,010 And this has accounted for that. 593 00:30:18,010 --> 00:30:20,140 So we have all the married people, the ones that have 594 00:30:20,140 --> 00:30:21,690 dropped out, the ones that didn't drop out. 595 00:30:21,690 --> 00:30:22,940 They're all here. 596 00:30:28,090 --> 00:30:31,080 So anybody wants to build on the story? 597 00:30:35,390 --> 00:30:39,620 So the story that I think is happening is something very 598 00:30:39,620 --> 00:30:42,340 similar to what you are [INAUDIBLE]. 599 00:30:42,340 --> 00:30:45,690 Basically, you can have two kinds of sex. 600 00:30:45,690 --> 00:30:47,180 You can have casual sex. 601 00:30:47,180 --> 00:30:50,810 And you can have sex with a regular partner. 602 00:30:50,810 --> 00:30:53,100 And then you can choose whether or not you want to use 603 00:30:53,100 --> 00:30:55,140 a condom or you don't want to use a condom. 604 00:30:55,140 --> 00:30:56,870 So you can choose your level of protection. 605 00:30:56,870 --> 00:31:00,000 Let's say you don't like it so much. 606 00:31:00,000 --> 00:31:00,670 You should like it. 607 00:31:00,670 --> 00:31:04,320 But let's say, they don't like it so much. 608 00:31:04,320 --> 00:31:07,630 And so there is some cost of using more protection. 609 00:31:07,630 --> 00:31:11,230 But the advantage of condoms is that you're less likely to 610 00:31:11,230 --> 00:31:14,530 get an STD, and you're less likely to get pregnant. 611 00:31:14,530 --> 00:31:19,280 And then what people look at is the cost and benefits of 612 00:31:19,280 --> 00:31:23,720 pregnancy, of a pregnancy now, as a teenager. 613 00:31:23,720 --> 00:31:25,930 And the cost of an STD. 614 00:31:25,930 --> 00:31:28,930 There is no benefit of getting an STD. 615 00:31:28,930 --> 00:31:31,730 And the probability that they might get pregnant and the 616 00:31:31,730 --> 00:31:33,760 probability that they might get an STD. 617 00:31:33,760 --> 00:31:37,030 And the cost and benefit of pregnancy is the value of 618 00:31:37,030 --> 00:31:39,000 having the child. 619 00:31:39,000 --> 00:31:41,640 You might like to have a child more if you have partner to 620 00:31:41,640 --> 00:31:44,700 have a child with then if you have to raise them alone. 621 00:31:44,700 --> 00:31:48,140 So the value of a child in marriage might be higher than 622 00:31:48,140 --> 00:31:50,810 the value of a child with a casual partner. 623 00:31:50,810 --> 00:31:54,450 And then on the cost side is foregone earnings, because if 624 00:31:54,450 --> 00:31:57,500 you have child, you have to take of the child. 625 00:31:57,500 --> 00:31:59,040 They kick you out of school anyway. 626 00:31:59,040 --> 00:32:02,030 You can absolutely not go to school when you're pregnant. 627 00:32:02,030 --> 00:32:05,140 In fact, they just passed a law saying that schools cannot 628 00:32:05,140 --> 00:32:09,310 evict children that are pregnant. 629 00:32:09,310 --> 00:32:13,380 When this was happening, it was the end of your schooling 630 00:32:13,380 --> 00:32:16,120 if you had a child. 631 00:32:16,120 --> 00:32:18,860 So the cost of the pregnancy is that you're losing a chance 632 00:32:18,860 --> 00:32:22,660 to further your education if you wanted to. 633 00:32:22,660 --> 00:32:25,690 So that's kind of the parameter of what's happening. 634 00:32:25,690 --> 00:32:28,545 Now we have the girls, and they are thinking, well. 635 00:32:28,545 --> 00:32:31,770 And of course, people like to have sex, girls, too. 636 00:32:31,770 --> 00:32:35,050 So they are like weighing, like, what should I do? 637 00:32:35,050 --> 00:32:37,140 How much sex should I have? 638 00:32:37,140 --> 00:32:39,700 Should I use contraception? 639 00:32:39,700 --> 00:32:47,740 And should I have a regular partner that 640 00:32:47,740 --> 00:32:51,190 might lead to a marriage? 641 00:32:51,190 --> 00:32:53,750 Or should I have casual partners that are not going to 642 00:32:53,750 --> 00:32:55,530 lead to marriage? 643 00:32:55,530 --> 00:32:59,770 And what the teacher training program is telling them, if 644 00:32:59,770 --> 00:33:02,800 you have casual sex, you're very, very, very 645 00:33:02,800 --> 00:33:04,040 likely to get an STD. 646 00:33:04,040 --> 00:33:07,130 It does change their probability that they might 647 00:33:07,130 --> 00:33:09,000 get an STD. 648 00:33:09,000 --> 00:33:11,110 But if you have sex within marriage, you're safe. 649 00:33:11,110 --> 00:33:13,670 You're fine. 650 00:33:13,670 --> 00:33:16,140 In fact, it might do the two things at the same time. 651 00:33:16,140 --> 00:33:19,810 Maybe, girls have not realized the difference between casual 652 00:33:19,810 --> 00:33:21,090 sex and married sex. 653 00:33:21,090 --> 00:33:23,110 But after the teacher training program, they think married 654 00:33:23,110 --> 00:33:24,390 sex is very safe. 655 00:33:24,390 --> 00:33:27,330 And they think that casual sex is very risky. 656 00:33:27,330 --> 00:33:29,760 And the truth is that married sex is not necessarily that 657 00:33:29,760 --> 00:33:33,320 safe, because the candidate for marriage, for a young 658 00:33:33,320 --> 00:33:36,780 girl, is usually not a young boy, who would be unable to 659 00:33:36,780 --> 00:33:37,570 support a child. 660 00:33:37,570 --> 00:33:40,210 It's usually an older man, what they call a "sugar 661 00:33:40,210 --> 00:33:45,630 daddy", who is able to provide for the girl and then provide 662 00:33:45,630 --> 00:33:46,700 for the child. 663 00:33:46,700 --> 00:33:49,690 And the girls are convinced that the "sugar daddies" are 664 00:33:49,690 --> 00:33:52,540 faithful to themselves and their wives. 665 00:33:52,540 --> 00:33:54,600 But that might not be entirely true. 666 00:33:54,600 --> 00:33:59,695 And so the actual risk of infection in married sex and 667 00:33:59,695 --> 00:34:01,620 in casual sex might not be that different. 668 00:34:04,742 --> 00:34:07,840 But when the teacher training program arrived, they feel, 669 00:34:07,840 --> 00:34:10,350 oh, my god, I have to stop having casual sex. 670 00:34:10,350 --> 00:34:12,920 That's very dangerous. 671 00:34:12,920 --> 00:34:15,929 Let me reduce, a little bit, the sex I'm going to have. 672 00:34:15,929 --> 00:34:20,130 But let me also move towards married sex. 673 00:34:20,130 --> 00:34:23,120 And by doing that, they're much more likely to become 674 00:34:23,120 --> 00:34:26,650 pregnant, because, within marriage, they think it's fine 675 00:34:26,650 --> 00:34:28,170 to have a kid. 676 00:34:28,170 --> 00:34:30,460 And also, they think that, within marriage, they don't 677 00:34:30,460 --> 00:34:32,630 need to be protected as much. 678 00:34:32,630 --> 00:34:35,400 So for these two reasons, the pregnancy rate will tend to 679 00:34:35,400 --> 00:34:36,900 increase if anything. 680 00:34:36,900 --> 00:34:39,409 On the other hand, they also have less sex, so that 681 00:34:39,409 --> 00:34:42,830 balances out, which is why we have a slight increase in the 682 00:34:42,830 --> 00:34:46,679 number of pregnancies, in the teacher training group, 683 00:34:46,679 --> 00:34:48,429 instead of seeing a decrease. 684 00:34:48,429 --> 00:34:50,400 Because we see a reduction in sex, but that's 685 00:34:50,400 --> 00:34:54,300 counterbalanced by a shift towards this single 686 00:34:54,300 --> 00:34:55,330 relationship. 687 00:34:55,330 --> 00:34:57,560 And because the single relationships are actually not 688 00:34:57,560 --> 00:35:02,040 that safe, we also don't see much in terms of reduction in 689 00:35:02,040 --> 00:35:04,560 HSV2 infection. 690 00:35:04,560 --> 00:35:08,020 Now in the uniform group, suddenly the cost of staying 691 00:35:08,020 --> 00:35:10,610 in school has become much lower. 692 00:35:10,610 --> 00:35:13,560 So staying in school becomes an option, something you can 693 00:35:13,560 --> 00:35:15,640 actually do. 694 00:35:15,640 --> 00:35:18,080 So then you feel, OK, I'm going to 695 00:35:18,080 --> 00:35:19,270 try and stay in school. 696 00:35:19,270 --> 00:35:21,860 But then I really shouldn't get pregnant. 697 00:35:21,860 --> 00:35:27,180 So you will move from married sex to casual sex, use more 698 00:35:27,180 --> 00:35:29,450 contraception, both because you don't want to have a kid 699 00:35:29,450 --> 00:35:35,610 and because you don't want to be pregnant. 700 00:35:35,610 --> 00:35:37,370 And that's going to reduce. 701 00:35:37,370 --> 00:35:38,850 So you're going to dropout less. 702 00:35:38,850 --> 00:35:40,290 You're going to have fewer kids. 703 00:35:40,290 --> 00:35:41,038 Yep? 704 00:35:41,038 --> 00:35:43,478 AUDIENCE: Why does the teacher training still have like a 705 00:35:43,478 --> 00:35:48,590 large percentage of kids with HSV2 positives? 706 00:35:48,590 --> 00:35:50,720 PROFESSOR: Because the reality is that married sex is not 707 00:35:50,720 --> 00:35:55,990 that less dangerous, in terms of HSV 708 00:35:55,990 --> 00:36:00,392 infection, than casual sex. 709 00:36:00,392 --> 00:36:02,058 AUDIENCE: But you're saying there's not a statistical 710 00:36:02,058 --> 00:36:04,200 difference between training and both. 711 00:36:04,200 --> 00:36:05,300 PROFESSOR: Oh, there is. 712 00:36:05,300 --> 00:36:07,390 So teacher training does nothing. 713 00:36:07,390 --> 00:36:08,640 So, oops. 714 00:36:11,960 --> 00:36:14,260 There is no difference between teacher 715 00:36:14,260 --> 00:36:16,240 training and control group. 716 00:36:16,240 --> 00:36:19,145 That's because with teacher training people switch towards 717 00:36:19,145 --> 00:36:19,880 married sex. 718 00:36:19,880 --> 00:36:21,010 And they think that's safe. 719 00:36:21,010 --> 00:36:22,520 But that's not really. 720 00:36:22,520 --> 00:36:25,640 Because they're having sex with sugar daddies who are 721 00:36:25,640 --> 00:36:27,140 maybe themselves infected. 722 00:36:27,140 --> 00:36:29,270 So they do that, which is why they have more kids. 723 00:36:29,270 --> 00:36:30,490 But they think that it's safe. 724 00:36:30,490 --> 00:36:33,510 But it's not safe, which is why you get no. 725 00:36:33,510 --> 00:36:36,310 And then I'm going to talk about both in a moment. 726 00:36:36,310 --> 00:36:38,780 So I've done teacher training. 727 00:36:38,780 --> 00:36:41,205 I've done uniform and now both. 728 00:36:41,205 --> 00:36:43,890 Well, what is happening with both is that, on the one hand, 729 00:36:43,890 --> 00:36:45,600 you think casual sex is very dangerous. 730 00:36:45,600 --> 00:36:46,830 You want to avoid it. 731 00:36:46,830 --> 00:36:48,570 On the other hand, you don't want to get pregnant. 732 00:36:48,570 --> 00:36:50,430 So married sex is not great at all. 733 00:36:50,430 --> 00:36:53,040 So you can't substitute. 734 00:36:53,040 --> 00:36:56,180 The teacher training program only pushes you towards one 735 00:36:56,180 --> 00:36:57,430 kind of sex. 736 00:36:57,430 --> 00:36:59,240 And therefore, so that you don't have to 737 00:36:59,240 --> 00:37:01,450 reduce your sex so much. 738 00:37:01,450 --> 00:37:04,430 The uniform program pushes you towards the other kind of sex. 739 00:37:04,430 --> 00:37:07,210 And you don't also reduce your sex so much. 740 00:37:07,210 --> 00:37:10,590 But the both program, the can't substitute, because 741 00:37:10,590 --> 00:37:13,160 neither of them are so great as an option. 742 00:37:13,160 --> 00:37:17,270 So your only option is either to reduce sex or to increase 743 00:37:17,270 --> 00:37:25,920 protection or both, which is why we get the biggest effect 744 00:37:25,920 --> 00:37:31,040 on HSV2, but we don't get the biggest effect on pregnancy. 745 00:37:31,040 --> 00:37:34,650 Because we don't have this switch towards casual sex, 746 00:37:34,650 --> 00:37:37,703 which really doesn't have very many pregnancies at all. 747 00:37:40,610 --> 00:37:43,540 So that, I think, is what is going on, which is we get 748 00:37:43,540 --> 00:37:46,630 exactly these patterns of results. 749 00:37:46,630 --> 00:37:46,790 Yeah? 750 00:37:46,790 --> 00:37:47,662 AUDIENCE: Sorry. 751 00:37:47,662 --> 00:37:51,616 So you're saying that you don't see this as all with 752 00:37:51,616 --> 00:37:53,065 respect to pregnancy? 753 00:37:53,065 --> 00:37:54,865 You don't necessarily have a shift 754 00:37:54,865 --> 00:37:56,870 towards casual sex there? 755 00:37:56,870 --> 00:38:00,150 PROFESSOR: So this is what we have for pregnancy. 756 00:38:00,150 --> 00:38:02,830 For pregnancy, with the both, they stay 757 00:38:02,830 --> 00:38:05,850 like at the same level. 758 00:38:05,850 --> 00:38:08,210 So you don't have these big shift. 759 00:38:08,210 --> 00:38:10,460 You really have very few pregnancies in casual 760 00:38:10,460 --> 00:38:11,860 relationships. 761 00:38:11,860 --> 00:38:14,443 AUDIENCE: Does that tell us anything about which one, 762 00:38:14,443 --> 00:38:19,273 uniform or teacher training, which is more [INAUDIBLE] 763 00:38:19,273 --> 00:38:21,095 to be more responsible to? 764 00:38:21,095 --> 00:38:23,760 PROFESSOR: So both of them, in a sense, I think, teacher 765 00:38:23,760 --> 00:38:27,880 training, they were responsive to both of them. 766 00:38:27,880 --> 00:38:31,260 But in the way that I just described, in terms of their 767 00:38:31,260 --> 00:38:32,280 thinking about it. 768 00:38:32,280 --> 00:38:35,420 But in terms of the effect you're trying to get, well, if 769 00:38:35,420 --> 00:38:41,980 you're interested in schooling or in teenage pregnancy, 770 00:38:41,980 --> 00:38:44,500 asking people to stay in school is the best thing. 771 00:38:44,500 --> 00:38:50,000 If you're interested in preventing HIV/AIDS and other 772 00:38:50,000 --> 00:38:52,040 STDs, then you want to add some training. 773 00:38:57,800 --> 00:39:01,790 That doesn't make if for a very, very easy policy lesson, 774 00:39:01,790 --> 00:39:04,380 because that slightly depends on you objective. 775 00:39:04,380 --> 00:39:06,670 But for a very fundamental reason, which is it 776 00:39:06,670 --> 00:39:10,010 corresponds to the way people decide and how they decide. 777 00:39:10,010 --> 00:39:14,780 So it seems the biggest policy answer to that is that you 778 00:39:14,780 --> 00:39:17,120 shouldn't have programs that are lying to people. 779 00:39:17,120 --> 00:39:18,980 It just doesn't help. 780 00:39:18,980 --> 00:39:22,720 So the way the program is structured is to try and make 781 00:39:22,720 --> 00:39:25,110 the girls as scared as possible of casual sex. 782 00:39:25,110 --> 00:39:29,780 But it would be a little bit better to insist on helping 783 00:39:29,780 --> 00:39:32,190 them to find a way to protect themselves, both against 784 00:39:32,190 --> 00:39:34,070 pregnancy and against HSV2. 785 00:39:34,070 --> 00:39:36,324 So I think that's kind of the lesson that comes out of this. 786 00:39:36,324 --> 00:39:37,132 Yep? 787 00:39:37,132 --> 00:39:40,592 AUDIENCE: So along those lines, how do these results 788 00:39:40,592 --> 00:39:43,924 reconcile with the fact that the teacher training doesn't 789 00:39:43,924 --> 00:39:46,066 actually emphasize contraception anymore, like 790 00:39:46,066 --> 00:39:47,732 safe sex at all? 791 00:39:47,732 --> 00:39:49,220 PROFESSOR: Yeah. 792 00:39:49,220 --> 00:39:52,070 These results are a direct consequence of the fact that 793 00:39:52,070 --> 00:39:55,320 the teacher training doesn't influence contraception. 794 00:39:55,320 --> 00:39:58,470 So they won't tell you anything. 795 00:39:58,470 --> 00:40:01,570 It's not an evaluation of teacher training in general. 796 00:40:01,570 --> 00:40:04,650 This is an evaluation of the way it is done in Kenya and in 797 00:40:04,650 --> 00:40:07,700 most of East Africa, which is explicitly 798 00:40:07,700 --> 00:40:09,930 ignoring the condoms. 799 00:40:09,930 --> 00:40:10,230 Yeah? 800 00:40:10,230 --> 00:40:10,620 AUDIENCE: Sorry. 801 00:40:10,620 --> 00:40:14,601 What I mean is like, in this last finding, the argument 802 00:40:14,601 --> 00:40:20,324 goes that, if you have both teacher training and uniforms, 803 00:40:20,324 --> 00:40:26,813 then people are more likely to not be married but have safer 804 00:40:26,813 --> 00:40:28,110 casual sex. 805 00:40:28,110 --> 00:40:29,370 PROFESSOR: And fewer sex in general. 806 00:40:29,370 --> 00:40:32,750 AUDIENCE: And have fewer sexual partners. 807 00:40:32,750 --> 00:40:35,890 PROFESSOR: Less sex, also. 808 00:40:35,890 --> 00:40:36,800 Both things will happen on this one. 809 00:40:36,800 --> 00:40:40,880 Because when you have just teacher training, people have 810 00:40:40,880 --> 00:40:44,480 the option to switch towards the married sex. 811 00:40:44,480 --> 00:40:46,550 When you just have the uniform, people have the 812 00:40:46,550 --> 00:40:49,500 option to switch toward the casual sex. 813 00:40:49,500 --> 00:40:53,020 But here, they don't have the option to switch, because they 814 00:40:53,020 --> 00:40:54,550 are scared of married sex, because 815 00:40:54,550 --> 00:40:55,590 they don't want children. 816 00:40:55,590 --> 00:40:57,380 And they're scared of casual sex, because they 817 00:40:57,380 --> 00:40:59,700 don't want an STD. 818 00:40:59,700 --> 00:41:02,280 And they think the probability is high with casual sex. 819 00:41:02,280 --> 00:41:05,450 So the only thing they can do, because they can't substitute 820 00:41:05,450 --> 00:41:08,810 between these two activities, is either to have less sex or 821 00:41:08,810 --> 00:41:10,970 to use more contraception. 822 00:41:10,970 --> 00:41:14,230 And the teacher training, itself, is not teaching them 823 00:41:14,230 --> 00:41:15,080 about contraception. 824 00:41:15,080 --> 00:41:17,720 But they all know, at some level, that it's available, 825 00:41:17,720 --> 00:41:19,590 and that it prevents both pregnancy 826 00:41:19,590 --> 00:41:21,570 and STDs, to a point. 827 00:41:26,140 --> 00:41:28,500 So this doesn't tell us you shouldn't train the teachers. 828 00:41:28,500 --> 00:41:30,120 I think this tells us you train the teachers with 829 00:41:30,120 --> 00:41:34,140 information that is actually useful, actionable, which 830 00:41:34,140 --> 00:41:36,240 might not be such revolutionary thinking. 831 00:41:36,240 --> 00:41:43,100 But when you spend a lot of time in education/health 832 00:41:43,100 --> 00:41:48,540 circle in Kenya, you realize it's actually a pretty big, 833 00:41:48,540 --> 00:41:49,810 pretty controversial statement. 834 00:41:52,640 --> 00:41:55,700 So Pascaline Dupas, when she was still a Ph.D. student, she 835 00:41:55,700 --> 00:42:00,260 worked in this project, to do something else, which we talk 836 00:42:00,260 --> 00:42:01,990 about also a little bit in the book, which is 837 00:42:01,990 --> 00:42:03,900 the sugar daddy project. 838 00:42:03,900 --> 00:42:06,632 And what the sugar daddy project is, is to tell the 839 00:42:06,632 --> 00:42:12,410 kids that it is not true that sex with older people is safe. 840 00:42:12,410 --> 00:42:16,020 In fact, older people are much more likely to infected with 841 00:42:16,020 --> 00:42:18,710 HIV than younger people. 842 00:42:18,710 --> 00:42:23,590 So that's the information that was given, plus a nice video 843 00:42:23,590 --> 00:42:28,850 about a girl who managed to trap someone who wants to be a 844 00:42:28,850 --> 00:42:31,700 sugar daddy and something like that, a very short, very quick 845 00:42:31,700 --> 00:42:32,660 intervention. 846 00:42:32,660 --> 00:42:35,800 A little film produced by UNICEF and then these 847 00:42:35,800 --> 00:42:39,890 statistics, which is your chance to get an STD. 848 00:42:39,890 --> 00:42:43,450 The infection rate of older men is much higher than the 849 00:42:43,450 --> 00:42:47,240 infection rate of younger men. 850 00:42:47,240 --> 00:42:50,670 And sadly, the infection rate of girls is also higher than 851 00:42:50,670 --> 00:42:53,460 that of boys, for the reason that girls are more likely to 852 00:42:53,460 --> 00:42:56,590 have sex with older men, who are likely to be infected. 853 00:42:56,590 --> 00:42:59,850 So you are telling the girls, you should really have sex 854 00:42:59,850 --> 00:43:04,530 with the boys, because they are safer than the men, not 855 00:43:04,530 --> 00:43:06,710 quite in so many words, but making 856 00:43:06,710 --> 00:43:08,140 the information available. 857 00:43:08,140 --> 00:43:10,390 And you're also telling the boys that these innocent 858 00:43:10,390 --> 00:43:13,600 looking girls actually are probably more 859 00:43:13,600 --> 00:43:15,770 risky than you think. 860 00:43:15,770 --> 00:43:19,790 And this intervention was actually quite effective at 861 00:43:19,790 --> 00:43:25,140 reducing sex with older partners, as indicated by the 862 00:43:25,140 --> 00:43:27,540 number of pregnancies with older dads. 863 00:43:27,540 --> 00:43:33,610 It reduced the pregnancy with older dads by 66%, 2/3. 864 00:43:33,610 --> 00:43:37,430 So this is also another sign of the fact that, if you tell 865 00:43:37,430 --> 00:43:40,820 people the right thing, like people will react to what it 866 00:43:40,820 --> 00:43:41,380 is you tell them. 867 00:43:41,380 --> 00:43:43,320 If you tell people the wrong thing, they're also going to 868 00:43:43,320 --> 00:43:43,870 react to it. 869 00:43:43,870 --> 00:43:46,000 But they'll not react to it in a way that is very. 870 00:43:48,520 --> 00:43:51,490 You don't get that many more HSV2, and you get many more 871 00:43:51,490 --> 00:43:55,060 kids who are married with older men and have 872 00:43:55,060 --> 00:43:56,640 children too early. 873 00:43:56,640 --> 00:44:00,080 So that's, maybe, the biggest lesson from this finding is 874 00:44:00,080 --> 00:44:01,890 you should tell people the truth and assume that they 875 00:44:01,890 --> 00:44:04,690 will be able to make the right decisions. 876 00:44:04,690 --> 00:44:09,470 You don't have to scare them away into doing things. 877 00:44:09,470 --> 00:44:13,585 So that's for that. 878 00:44:17,510 --> 00:44:22,410 So I guess the key lesson is fertility 879 00:44:22,410 --> 00:44:23,610 decisions are like others. 880 00:44:23,610 --> 00:44:25,740 We need to try and understand them. 881 00:44:25,740 --> 00:44:29,380 And this is one area where policies have been based on 882 00:44:29,380 --> 00:44:31,640 the assumption that people are idiots. 883 00:44:31,640 --> 00:44:34,440 And that's, I think, a very bad assumption in general and 884 00:44:34,440 --> 00:44:36,790 particularly, perhaps, in this area, because it's a very 885 00:44:36,790 --> 00:44:40,210 important decision that people are much more likely to take, 886 00:44:40,210 --> 00:44:43,290 with a lot of thinking and with a lot of you. 887 00:44:43,290 --> 00:44:47,180 So that brings us, pretty naturally, into the next 888 00:44:47,180 --> 00:44:52,710 topic, which is, if it's a decision, inherently, 889 00:44:52,710 --> 00:44:57,370 fertility decisions are decisions that are made by a 890 00:44:57,370 --> 00:45:00,300 man and a woman. 891 00:45:00,300 --> 00:45:03,000 But if we think about it, a lot of the other decisions we 892 00:45:03,000 --> 00:45:05,950 talked about in this class are also decisions that are being 893 00:45:05,950 --> 00:45:07,740 made by a man and a woman. 894 00:45:07,740 --> 00:45:11,620 And we completely ignored that our discussions so far. 895 00:45:11,620 --> 00:45:15,730 I've always talked about parents decide how many, 896 00:45:15,730 --> 00:45:19,350 whether to immunize their kids, how much to feed them, 897 00:45:19,350 --> 00:45:22,740 how many years of education they should get, et cetera. 898 00:45:22,740 --> 00:45:26,370 And in this, I've ignored the dynamic of the family. 899 00:45:26,370 --> 00:45:28,660 So now I want to spend a bit of time-- 900 00:45:28,660 --> 00:45:31,190 not much, but the rest of this lecture-- 901 00:45:31,190 --> 00:45:33,820 to talk about family decisions. 902 00:45:57,990 --> 00:46:02,120 So the question, here, we're going to ask is, how do 903 00:46:02,120 --> 00:46:03,370 families decide? 904 00:46:06,550 --> 00:46:09,190 So when we mean family, we mean who? 905 00:46:09,190 --> 00:46:09,940 Like who decides? 906 00:46:09,940 --> 00:46:10,820 The dad decides. 907 00:46:10,820 --> 00:46:11,650 The mom decides. 908 00:46:11,650 --> 00:46:12,720 The mother-in-law decides. 909 00:46:12,720 --> 00:46:16,590 The kids, what do kids have to play? 910 00:46:16,590 --> 00:46:19,180 They are relevant to all sorts of decisions that are taken 911 00:46:19,180 --> 00:46:21,390 jointly by parents. 912 00:46:21,390 --> 00:46:23,790 Until now, we've ignored it. 913 00:46:23,790 --> 00:46:26,710 We've assumed that there was something called, a family. 914 00:46:26,710 --> 00:46:29,710 And yet the family, we know, is a group. 915 00:46:29,710 --> 00:46:35,580 So in what case would it be legitimate to ignore the fact 916 00:46:35,580 --> 00:46:37,600 that the family is a group of people? 917 00:46:37,600 --> 00:46:40,790 There are two cases where we could say, well, we know it's 918 00:46:40,790 --> 00:46:42,040 a group, but we can ignore it. 919 00:46:45,432 --> 00:46:47,420 AUDIENCE: I would think all families are the same? 920 00:46:47,420 --> 00:46:48,780 PROFESSOR: All family members. 921 00:46:48,780 --> 00:46:51,310 Like if we think that all family members are the same, 922 00:46:51,310 --> 00:46:53,230 they all have the same preferences, so they all want 923 00:46:53,230 --> 00:46:54,580 the same thing. 924 00:46:54,580 --> 00:46:58,850 So anybody who has been part of a family, does this sound 925 00:46:58,850 --> 00:47:01,810 like a good approximation to a family, the 926 00:47:01,810 --> 00:47:03,456 way a family functions? 927 00:47:03,456 --> 00:47:04,450 AUDIENCE: No. 928 00:47:04,450 --> 00:47:04,970 PROFESSOR: Not really. 929 00:47:04,970 --> 00:47:06,570 I think it's not the case that there are no 930 00:47:06,570 --> 00:47:07,560 conflicts within families. 931 00:47:07,560 --> 00:47:09,700 People have different views about how 932 00:47:09,700 --> 00:47:11,160 everything should be done. 933 00:47:11,160 --> 00:47:14,850 And the second way in which we could ignore the fact that 934 00:47:14,850 --> 00:47:16,625 it's a collection of individuals is? 935 00:47:19,385 --> 00:47:22,300 AUDIENCE: [INAUDIBLE]? 936 00:47:22,300 --> 00:47:24,380 PROFESSOR: If one person made all the decisions. 937 00:47:24,380 --> 00:47:29,580 So we had the mother or the father is like the person who 938 00:47:29,580 --> 00:47:32,220 makes all the decisions for everyone, possibly taking into 939 00:47:32,220 --> 00:47:35,600 account people's preferences, but not giving them a voice in 940 00:47:35,600 --> 00:47:36,990 how the decision is being made. 941 00:47:40,450 --> 00:47:42,740 Does that sound a plausible model of the family? 942 00:47:45,900 --> 00:47:47,600 AUDIENCE: At certain ages. 943 00:47:47,600 --> 00:47:48,040 PROFESSOR: Sorry? 944 00:47:48,040 --> 00:47:48,730 AUDIENCE: At certain ages. 945 00:47:48,730 --> 00:47:52,125 PROFESSOR: At certain ages, of course, like the infant is not 946 00:47:52,125 --> 00:47:55,240 really in control, so someone decides for them. 947 00:47:58,590 --> 00:47:59,020 Norm? 948 00:47:59,020 --> 00:48:02,124 AUDIENCE: Probably on the gender breakdown, so the guys 949 00:48:02,124 --> 00:48:04,170 will decide, or the male in the family will 950 00:48:04,170 --> 00:48:05,590 decide for the females. 951 00:48:05,590 --> 00:48:06,070 PROFESSOR: Right. 952 00:48:06,070 --> 00:48:11,380 So it might be that in some societies, there is really a 953 00:48:11,380 --> 00:48:14,540 lot of control by the guys in the family, as you say. 954 00:48:17,970 --> 00:48:20,500 But that will depend. 955 00:48:20,500 --> 00:48:24,100 Their society's say, you take Pakistan or part of India, 956 00:48:24,100 --> 00:48:27,390 which will be more patriarchal than Africa, where the African 957 00:48:27,390 --> 00:48:28,880 family is like-- 958 00:48:28,880 --> 00:48:30,810 it depends where, of course-- 959 00:48:30,810 --> 00:48:32,570 where there is more equality between genders. 960 00:48:32,570 --> 00:48:35,500 So no one, maybe, is like in a position to 961 00:48:35,500 --> 00:48:37,390 really, fully decide. 962 00:48:37,390 --> 00:48:40,680 So maybe the assumption that one person decides for 963 00:48:40,680 --> 00:48:43,820 everyone is more plausible in some cases, but we feel that 964 00:48:43,820 --> 00:48:46,480 it's not going to be always true. 965 00:48:46,480 --> 00:48:50,300 And in a lot of cases, the family is not one person. 966 00:48:50,300 --> 00:48:52,660 The family is a collection of people. 967 00:48:52,660 --> 00:48:58,320 And they are trying to fight it out. 968 00:48:58,320 --> 00:49:01,190 So women and men have different preferences. 969 00:49:01,190 --> 00:49:03,910 For example, one area where they have very different 970 00:49:03,910 --> 00:49:10,270 preferences is how many children they want to have. 971 00:49:10,270 --> 00:49:12,600 So, for example, there was a study in Zambia-- that I'm 972 00:49:12,600 --> 00:49:15,240 going to show you in a moment-- showing that women 973 00:49:15,240 --> 00:49:18,660 want to have much fewer children than men. 974 00:49:18,660 --> 00:49:22,810 So one reason, of course, would be that it's Zambia and 975 00:49:22,810 --> 00:49:26,390 the probability of dying in childbirth is not 0, while if 976 00:49:26,390 --> 00:49:29,090 you're the woman, you're more likely to be the one who is 977 00:49:29,090 --> 00:49:30,340 paying the cost of that. 978 00:49:32,960 --> 00:49:36,000 So as Norm mentioned, the family is patriarchal in many 979 00:49:36,000 --> 00:49:38,990 developing countries, so maybe men take lot of decisions. 980 00:49:38,990 --> 00:49:42,050 But even there, you have, usually, a little sphere, like 981 00:49:42,050 --> 00:49:44,500 how to buy the food, et cetera, where it's really the 982 00:49:44,500 --> 00:49:46,680 woman who decides and where bargaining 983 00:49:46,680 --> 00:49:49,400 could be taking place. 984 00:49:49,400 --> 00:49:52,050 So how do we test this? 985 00:49:52,050 --> 00:49:53,270 How do we know? 986 00:49:53,270 --> 00:49:55,840 How we can we try and see whether the family is 987 00:49:55,840 --> 00:49:59,680 patriarchal, is unitary, either everybody agrees or 988 00:49:59,680 --> 00:50:02,300 someone decides for everyone? 989 00:50:02,300 --> 00:50:06,050 So if the family behaves like one individual, we could 990 00:50:06,050 --> 00:50:08,020 analyze the decision problem of a family-- 991 00:50:08,020 --> 00:50:10,920 for example, how much food to buy, whether to send the kids 992 00:50:10,920 --> 00:50:13,790 to school, how many children to how, et cetera-- 993 00:50:13,790 --> 00:50:16,770 like the problem of one single person. 994 00:50:16,770 --> 00:50:19,440 So the only thing that should matter to a single person, 995 00:50:19,440 --> 00:50:22,760 and, therefore, to the family in this context, is the 996 00:50:22,760 --> 00:50:27,730 family's overall income, the prices of things-- 997 00:50:27,730 --> 00:50:32,280 how expensive it is to buy whichever types of thing, 998 00:50:32,280 --> 00:50:35,320 school fees, et cetera-- 999 00:50:35,320 --> 00:50:41,310 and information that you have, to everyone in the family. 1000 00:50:41,310 --> 00:50:42,980 But if the family is not unitary. 1001 00:50:42,980 --> 00:50:46,380 It's what we call this model, where one person decides for 1002 00:50:46,380 --> 00:50:47,900 everyone or everybody agrees. 1003 00:50:47,900 --> 00:50:51,640 We call this the unitary model, like there is unity in 1004 00:50:51,640 --> 00:50:52,820 the family. 1005 00:50:52,820 --> 00:50:55,780 So if the family is not unitary, than other things are 1006 00:50:55,780 --> 00:50:57,710 going to start to matter. 1007 00:50:57,710 --> 00:51:00,740 So, for example, it's not only the overall income that is 1008 00:51:00,740 --> 00:51:04,730 going to matter, but who brings what share. 1009 00:51:04,730 --> 00:51:09,680 Because they can now, say, well, if you don't do what I 1010 00:51:09,680 --> 00:51:11,690 want to do, then I'm just going to take my 1011 00:51:11,690 --> 00:51:13,430 income and go away. 1012 00:51:13,430 --> 00:51:16,220 So now it's not the overall income that's going to matter, 1013 00:51:16,220 --> 00:51:18,540 it's the share of the income. 1014 00:51:18,540 --> 00:51:21,820 So again, we could see men taking much more decisions, 1015 00:51:21,820 --> 00:51:24,670 that's because they are bring more of the income. 1016 00:51:24,670 --> 00:51:30,300 And we could see, maybe, that when woman bring more income, 1017 00:51:30,300 --> 00:51:33,720 they would be able to take more decisions. 1018 00:51:33,720 --> 00:51:36,910 The second thing that might matter is private information. 1019 00:51:36,910 --> 00:51:39,480 So if you know that the family is not unitary, and you find 1020 00:51:39,480 --> 00:51:42,810 something out, you might keep it for you. 1021 00:51:42,810 --> 00:51:46,160 Because that gives you some power to make some decisions. 1022 00:51:46,160 --> 00:51:49,770 So let's look at whether we find evidence that private 1023 00:51:49,770 --> 00:51:52,150 information, that is information that one member 1024 00:51:52,150 --> 00:51:55,630 has and not the other, makes a difference in decisions and 1025 00:51:55,630 --> 00:51:58,610 whether it's who brings the income to 1026 00:51:58,610 --> 00:52:01,050 family matters as well. 1027 00:52:01,050 --> 00:52:02,300 Let's look at these two things. 1028 00:52:04,930 --> 00:52:08,970 So private information, let's look at fertility. 1029 00:52:08,970 --> 00:52:13,060 So I was telling this data from Zambia showing that women 1030 00:52:13,060 --> 00:52:16,010 want to have fewer kids than men. 1031 00:52:16,010 --> 00:52:18,630 And in fact, they report hiding contraceptives from 1032 00:52:18,630 --> 00:52:21,280 their husbands. 1033 00:52:21,280 --> 00:52:26,140 So in that case, you could ask a question. 1034 00:52:26,140 --> 00:52:30,990 If I provide information to women about contraceptives, 1035 00:52:30,990 --> 00:52:34,150 would it make a difference if I provide it to the woman 1036 00:52:34,150 --> 00:52:39,260 alone or if I provide it together to women and men? 1037 00:52:39,260 --> 00:52:43,050 So under the unitary model, what would we expect? 1038 00:52:43,050 --> 00:52:47,750 AUDIENCE: Whoever the head of the household is, [INAUDIBLE]. 1039 00:52:47,750 --> 00:52:49,620 PROFESSOR: Yeah, whoever we give it to, even if it's not 1040 00:52:49,620 --> 00:52:51,270 the head, we give it to the woman, she's 1041 00:52:51,270 --> 00:52:52,780 going to tell the husband. 1042 00:52:52,780 --> 00:52:54,330 So it should make no difference. 1043 00:52:54,330 --> 00:52:56,785 Or maybe it would even have a bigger effect to give it to 1044 00:52:56,785 --> 00:52:59,390 the men, because they are more likely to understand, they're 1045 00:52:59,390 --> 00:53:01,690 more likely to be involved, et cetera. 1046 00:53:01,690 --> 00:53:05,100 So a lot of family planning policies are now saying, we 1047 00:53:05,100 --> 00:53:06,980 have to involve the men. 1048 00:53:06,980 --> 00:53:08,500 Because after all, the men have something to do with 1049 00:53:08,500 --> 00:53:09,140 contraception. 1050 00:53:09,140 --> 00:53:09,830 That's the idea. 1051 00:53:09,830 --> 00:53:11,870 Let's involve the men, and they would be more likely to 1052 00:53:11,870 --> 00:53:13,210 be informed. 1053 00:53:13,210 --> 00:53:23,540 On the other hand, if it is the case that women are more 1054 00:53:23,540 --> 00:53:27,770 likely to want fewer children, then involving only the woman, 1055 00:53:27,770 --> 00:53:32,730 it gives her an option to hide the contraception from her 1056 00:53:32,730 --> 00:53:34,280 husband and, therefore, potentially, 1057 00:53:34,280 --> 00:53:35,430 to have fewer kids. 1058 00:53:35,430 --> 00:53:37,050 So it could really cut both ways. 1059 00:53:37,050 --> 00:53:39,910 In the unitary model, you would expect no difference or 1060 00:53:39,910 --> 00:53:42,920 maybe doing it to both together would be better. 1061 00:53:42,920 --> 00:53:47,740 Or it would have a larger effect, because the 1062 00:53:47,740 --> 00:53:49,190 information would be more likely to be 1063 00:53:49,190 --> 00:53:50,730 diffused in the family. 1064 00:53:50,730 --> 00:53:52,750 Under the non-unitary model, maybe the woman 1065 00:53:52,750 --> 00:53:54,750 would matter more. 1066 00:53:54,750 --> 00:53:56,280 So what do you think is happening? 1067 00:53:59,220 --> 00:54:00,470 What's your bet? 1068 00:54:03,502 --> 00:54:05,306 AUDIENCE: The women use it more when their 1069 00:54:05,306 --> 00:54:06,460 husbands don't know. 1070 00:54:06,460 --> 00:54:07,440 PROFESSOR: Right. 1071 00:54:07,440 --> 00:54:11,210 What they find is that women use it much more when their 1072 00:54:11,210 --> 00:54:12,180 husband doesn't how. 1073 00:54:12,180 --> 00:54:12,960 So they did that? 1074 00:54:12,960 --> 00:54:17,620 Well, they ran an experiment where they provided people a 1075 00:54:17,620 --> 00:54:20,480 voucher to skip the queue. 1076 00:54:20,480 --> 00:54:22,780 Because in Zambia, in principle, you are allowed to 1077 00:54:22,780 --> 00:54:25,030 have contraception, et cetera. 1078 00:54:25,030 --> 00:54:27,630 But in practice, it takes a long time. 1079 00:54:27,630 --> 00:54:28,510 The nurse calls you. 1080 00:54:28,510 --> 00:54:31,430 Once you've come, if you don't have your clothes for an 1081 00:54:31,430 --> 00:54:32,880 examination. 1082 00:54:32,880 --> 00:54:35,730 So people don't really go to the family planning. 1083 00:54:35,730 --> 00:54:40,850 So here, it's a voucher for first class treatment, no need 1084 00:54:40,850 --> 00:54:46,170 to bring your clothes and a good, nice setting. 1085 00:54:46,170 --> 00:54:49,350 So they did that. 1086 00:54:49,350 --> 00:54:50,510 I don't know if it was half. 1087 00:54:50,510 --> 00:54:54,550 But the separated the group in two, one where they gave the 1088 00:54:54,550 --> 00:54:57,750 voucher to the wife, alone, and one where they gave it to 1089 00:54:57,750 --> 00:54:59,000 wife and husband together. 1090 00:55:01,410 --> 00:55:04,190 And so the question is whether women are more likely to take 1091 00:55:04,190 --> 00:55:08,440 care of their contraception, whether they're more likely to 1092 00:55:08,440 --> 00:55:10,890 be able to control their fertility in one 1093 00:55:10,890 --> 00:55:13,130 case than in the other. 1094 00:55:13,130 --> 00:55:18,050 So that's what the find for the take-up of the voucher. 1095 00:55:18,050 --> 00:55:21,850 So the first two columns are whether or not 1096 00:55:21,850 --> 00:55:23,970 you redeemed a voucher. 1097 00:55:23,970 --> 00:55:29,200 And the second two are whether or not you get an injection of 1098 00:55:29,200 --> 00:55:33,670 contraception, which is a method of contraception that 1099 00:55:33,670 --> 00:55:35,770 you can hide easily. 1100 00:55:35,770 --> 00:55:37,580 Because you got the injection. 1101 00:55:37,580 --> 00:55:39,940 No one knows that you are not fertile for 1102 00:55:39,940 --> 00:55:42,180 the next three months. 1103 00:55:42,180 --> 00:55:45,590 And what you find is that people are more likely to 1104 00:55:45,590 --> 00:55:48,880 redeem the voucher in the individual treatment than in 1105 00:55:48,880 --> 00:55:50,920 the couple treatment . 1106 00:55:50,920 --> 00:55:54,640 And they are more likely to receive the concealable method 1107 00:55:54,640 --> 00:55:59,240 in the individual treatment than in a couple treatment. 1108 00:55:59,240 --> 00:56:05,220 So the effects are of the order of, let's say, 55% 1109 00:56:05,220 --> 00:56:08,200 received a voucher in individual versus 42%. 1110 00:56:08,200 --> 00:56:11,800 So these are not negligible effects. 1111 00:56:11,800 --> 00:56:15,010 So it seems that it matters how you provide information. 1112 00:56:15,010 --> 00:56:20,260 It matters also for the number of kids that people have. 1113 00:56:20,260 --> 00:56:24,270 So this is the fraction of births, in the months 1114 00:56:24,270 --> 00:56:28,450 following the treatment, among women, who said, at baseline, 1115 00:56:28,450 --> 00:56:30,760 that they did not want any more births. 1116 00:56:30,760 --> 00:56:36,000 So you can think of them as unwanted pregnancies. 1117 00:56:36,000 --> 00:56:38,820 So six months later, of course, you 1118 00:56:38,820 --> 00:56:40,070 should see no effect. 1119 00:56:46,045 --> 00:56:48,650 The couple treatment, which is the red line, is a little 1120 00:56:48,650 --> 00:56:51,410 below, but that's not due to the treatment, because it's 1121 00:56:51,410 --> 00:56:53,990 six months later. 1122 00:56:53,990 --> 00:56:57,000 And then at nine months, the couple cross. 1123 00:56:57,000 --> 00:57:01,920 And the red line is above the blue line for a few months, 1124 00:57:01,920 --> 00:57:05,950 which reflects about three months of Depo-Provera use. 1125 00:57:05,950 --> 00:57:08,160 And then what happened is that the effect on that, it 1126 00:57:08,160 --> 00:57:09,620 disappears subsequently. 1127 00:57:09,620 --> 00:57:12,420 And it goes up a lot for both. 1128 00:57:12,420 --> 00:57:17,740 And the reason is that there was a policy change in Zambia. 1129 00:57:17,740 --> 00:57:21,040 All of the injectable were taken away from the market. 1130 00:57:21,040 --> 00:57:26,310 So any effect of the treatment goes away, because after your 1131 00:57:26,310 --> 00:57:29,230 first injection, you could not get the second one. 1132 00:57:29,230 --> 00:57:36,860 So we should expect an effect in the 9 to 14 month, after 1133 00:57:36,860 --> 00:57:39,950 the treatment, which is exactly what we find. 1134 00:57:39,950 --> 00:57:45,310 So if we look at the number of unwanted births that were 1135 00:57:45,310 --> 00:57:47,140 prevented-- 1136 00:57:47,140 --> 00:57:48,630 I actually don't remember exactly the number. 1137 00:57:48,630 --> 00:57:50,950 But it's a fair number of unwanted births that were 1138 00:57:50,950 --> 00:57:56,750 prevented by going from couple treatment to 1139 00:57:56,750 --> 00:57:57,780 the woman only treatment. 1140 00:57:57,780 --> 00:58:01,000 So that's the first indication that the family is not 1141 00:58:01,000 --> 00:58:04,380 unitary, because who you are giving information to changes 1142 00:58:04,380 --> 00:58:05,210 the outcome. 1143 00:58:05,210 --> 00:58:06,850 So people hide information. 1144 00:58:10,560 --> 00:58:13,560 The second information is what economists 1145 00:58:13,560 --> 00:58:15,870 call bargaining power. 1146 00:58:15,870 --> 00:58:18,650 So even when the information set is the same for both 1147 00:58:18,650 --> 00:58:20,560 partners, they might start to bargain over a family 1148 00:58:20,560 --> 00:58:25,380 decision, who works, how hard, how many children to have, how 1149 00:58:25,380 --> 00:58:27,710 much to feed the children, how much to educate the 1150 00:58:27,710 --> 00:58:29,520 children, et cetera. 1151 00:58:29,520 --> 00:58:32,860 So the bargaining power is your ability to weigh-in on 1152 00:58:32,860 --> 00:58:35,280 the family decision, if the family decision are the 1153 00:58:35,280 --> 00:58:38,560 process of these discussions. 1154 00:58:38,560 --> 00:58:41,630 So what would affect a woman's bargaining power, other than 1155 00:58:41,630 --> 00:58:42,920 the information she has? 1156 00:58:46,310 --> 00:58:49,020 So we already mentioned an income. 1157 00:58:49,020 --> 00:58:51,992 What else could be there? 1158 00:58:51,992 --> 00:58:54,990 If you have good eyesight, you can read it on the slide, but 1159 00:58:54,990 --> 00:58:56,670 you can also think about it instead. 1160 00:59:00,135 --> 00:59:03,590 AUDIENCE: Maybe divorce policy? 1161 00:59:03,590 --> 00:59:04,760 PROFESSOR: Absolutely right. 1162 00:59:04,760 --> 00:59:07,400 Divorce policy will affect bargaining power. 1163 00:59:07,400 --> 00:59:08,863 And how does it work? 1164 00:59:08,863 --> 00:59:11,761 AUDIENCE: Depending on how they allocate the family 1165 00:59:11,761 --> 00:59:15,142 resources after divorce, I don't know there. 1166 00:59:15,142 --> 00:59:18,523 But I know in the past, in the US, they would allocate it 1167 00:59:18,523 --> 00:59:19,972 mostly to the man. 1168 00:59:19,972 --> 00:59:21,920 And now, in a lot of states, it's 50-50. 1169 00:59:21,920 --> 00:59:22,410 PROFESSOR: Exactly. 1170 00:59:22,410 --> 00:59:24,600 If you live in a divorce regime, where all of the 1171 00:59:24,600 --> 00:59:28,060 assets, and maybe the control over the children, goes to the 1172 00:59:28,060 --> 00:59:31,910 man, then for a woman to divorce is a disaster. 1173 00:59:31,910 --> 00:59:34,080 So when the take decisions, when they bargain, they are 1174 00:59:34,080 --> 00:59:36,980 thinking, what is my outside option? 1175 00:59:36,980 --> 00:59:40,520 And if I'm thinking my outside option is absolutely horrible, 1176 00:59:40,520 --> 00:59:42,640 then I have to agree to whatever 1177 00:59:42,640 --> 00:59:44,670 the husband is proposing. 1178 00:59:44,670 --> 00:59:48,070 So the divorce laws are going to make a difference. 1179 00:59:48,070 --> 00:59:52,620 How much of the commons assets can the woman get back, and 1180 00:59:52,620 --> 00:59:55,020 how much control of kids could she have? 1181 00:59:55,020 --> 00:59:58,430 And in fact, it's been shown, in the US, that when the 1182 00:59:58,430 --> 01:00:01,150 change that you talked about was implemented, it was 1183 01:00:01,150 --> 01:00:03,860 implemented in different states at different times. 1184 01:00:03,860 --> 01:00:07,770 So you can trace what happened to women, in particular, to 1185 01:00:07,770 --> 01:00:10,230 decisions that are more in the interest of the woman. 1186 01:00:10,230 --> 01:00:13,190 You can see that the decisions, within the 1187 01:00:13,190 --> 01:00:16,390 household, improved in favor of the woman when the divorce 1188 01:00:16,390 --> 01:00:20,500 law improved to give women more rights. 1189 01:00:20,500 --> 01:00:22,628 So divorce law is one. 1190 01:00:22,628 --> 01:00:25,098 AUDIENCE: Actually, like health? 1191 01:00:25,098 --> 01:00:28,925 So if the husband is in really bad health, the woman probably 1192 01:00:28,925 --> 01:00:31,520 has a lot of power in the family. 1193 01:00:31,520 --> 01:00:33,090 PROFESSOR: If the husband has bad health, the 1194 01:00:33,090 --> 01:00:34,340 woman has more power. 1195 01:00:34,340 --> 01:00:36,260 Yes, that could be one. 1196 01:00:36,260 --> 01:00:39,490 Maybe because his outside option is now very bad, 1197 01:00:39,490 --> 01:00:42,620 because A, he couldn't get married to someone else, 1198 01:00:42,620 --> 01:00:43,460 because he is so sick. 1199 01:00:43,460 --> 01:00:50,330 B, he couldn't get by himself, so he has to kind of be nice 1200 01:00:50,330 --> 01:00:53,570 so that she will continue to take care of him. 1201 01:00:53,570 --> 01:00:55,170 AUDIENCE: So this slightly related to a lot of them. 1202 01:00:57,825 --> 01:01:01,730 If you came from a wealthy family or came from a family 1203 01:01:01,730 --> 01:01:05,322 with a status, the husband couldn't get anything from the 1204 01:01:05,322 --> 01:01:06,768 family, but just so as not to embarrass 1205 01:01:06,768 --> 01:01:08,220 himself in the society? 1206 01:01:08,220 --> 01:01:08,690 PROFESSOR: Yes, exactly. 1207 01:01:08,690 --> 01:01:12,560 So the family you're coming from will influence bargaining 1208 01:01:12,560 --> 01:01:17,550 power, because, maybe, not wanting to embarrass himself. 1209 01:01:17,550 --> 01:01:20,500 Also, because it could be a threat that my very strong 1210 01:01:20,500 --> 01:01:24,060 brothers are going to break your kneecaps if you don't 1211 01:01:24,060 --> 01:01:25,650 behave with me. 1212 01:01:25,650 --> 01:01:28,070 So that creates like an outside option. 1213 01:01:28,070 --> 01:01:31,350 Also, in case of divorce, again, it gives you a chance 1214 01:01:31,350 --> 01:01:34,350 to come back to something, whereas if you have no family, 1215 01:01:34,350 --> 01:01:36,100 then you have no options. 1216 01:01:36,100 --> 01:01:40,030 So that's very good, your family, 1217 01:01:40,030 --> 01:01:41,580 where people came from. 1218 01:01:41,580 --> 01:01:42,938 What else could we have? 1219 01:01:42,938 --> 01:01:46,354 AUDIENCE: The relative level of education [INAUDIBLE]. 1220 01:01:46,354 --> 01:01:50,746 If she's just been to high school, [INAUDIBLE] 1221 01:01:50,746 --> 01:01:53,186 make that economic decision. 1222 01:01:53,186 --> 01:01:58,554 She may not have the mental leverage to say, what I'm 1223 01:01:58,554 --> 01:02:00,506 saying makes more sense. 1224 01:02:00,506 --> 01:02:01,030 PROFESSOR: Yes. 1225 01:02:01,030 --> 01:02:01,820 So that's very good. 1226 01:02:01,820 --> 01:02:05,760 So the relative level of education will matter, both 1227 01:02:05,760 --> 01:02:07,820 because, again, it creates some different outside 1228 01:02:07,820 --> 01:02:09,750 options, because with a high school dropout, 1229 01:02:09,750 --> 01:02:11,040 what would you do? 1230 01:02:11,040 --> 01:02:12,820 You're kind of dependent. 1231 01:02:12,820 --> 01:02:16,230 And the other point you're making, which is not about 1232 01:02:16,230 --> 01:02:18,940 outside options, but even your ability to bargain. 1233 01:02:18,940 --> 01:02:21,510 You might not feel confident. 1234 01:02:21,510 --> 01:02:23,810 When someone tells you that I know better, you might feel 1235 01:02:23,810 --> 01:02:26,080 like, yeah, maybe you know better. 1236 01:02:26,080 --> 01:02:27,791 Good. 1237 01:02:27,791 --> 01:02:31,767 AUDIENCE: Is it possible to consider that a woman could 1238 01:02:31,767 --> 01:02:35,246 tell her husband, we won't have intercourse if you don't 1239 01:02:35,246 --> 01:02:37,750 do this and that. 1240 01:02:37,750 --> 01:02:37,980 PROFESSOR: Yeah. 1241 01:02:37,980 --> 01:02:45,430 Not only is it possible, but I think that's a pretty frequent 1242 01:02:45,430 --> 01:02:47,440 bargaining technique. 1243 01:02:47,440 --> 01:02:52,830 So, in that case, if that's the bargaining technique, then 1244 01:02:52,830 --> 01:02:56,830 you would think how attractive each partner is and how long 1245 01:02:56,830 --> 01:02:58,890 they've been in a relationship and things like that 1246 01:02:58,890 --> 01:02:59,720 are going to matter. 1247 01:02:59,720 --> 01:03:02,010 Because the level of the threat is also whether or not 1248 01:03:02,010 --> 01:03:06,170 you think it's a problem or you could figure something 1249 01:03:06,170 --> 01:03:07,420 else out if necessary. 1250 01:03:11,610 --> 01:03:14,338 What else would we have? 1251 01:03:20,430 --> 01:03:23,920 Well, one more is your ability to earn an independent income. 1252 01:03:23,920 --> 01:03:26,760 So that's related to the point you're making about education. 1253 01:03:26,760 --> 01:03:33,660 But it could also be just labor market opportunities for 1254 01:03:33,660 --> 01:03:34,630 men and women. 1255 01:03:34,630 --> 01:03:38,280 So for example, if you live in a society where women cannot 1256 01:03:38,280 --> 01:03:42,680 get jobs, because say there is [INAUDIBLE], then what will 1257 01:03:42,680 --> 01:03:45,040 the woman do without husbands? 1258 01:03:45,040 --> 01:03:46,350 There would be a big problem. 1259 01:03:46,350 --> 01:03:51,370 Whereas if you live in a society like here, where you 1260 01:03:51,370 --> 01:03:57,580 could get a job, if you needed to, then that changes your 1261 01:03:57,580 --> 01:04:00,950 level of outside options, and, therefore, changes the way the 1262 01:04:00,950 --> 01:04:03,010 negotiations would be made. 1263 01:04:03,010 --> 01:04:06,120 Another one is the marriage market. 1264 01:04:06,120 --> 01:04:09,130 Is it possible for a woman to get married or for a man to 1265 01:04:09,130 --> 01:04:11,310 get married after a divorce? 1266 01:04:11,310 --> 01:04:13,190 There are some places where it's harder. 1267 01:04:13,190 --> 01:04:19,150 What is the gender ratio, how many women there are for men? 1268 01:04:19,150 --> 01:04:23,350 If you are in, for example, an immigrant community, where the 1269 01:04:23,350 --> 01:04:27,540 ratio of men to woman is very skewed, because a lot of men 1270 01:04:27,540 --> 01:04:32,190 have come, new men have come, then the woman can threaten to 1271 01:04:32,190 --> 01:04:34,090 get married to any of those available 1272 01:04:34,090 --> 01:04:35,990 boys within the community. 1273 01:04:35,990 --> 01:04:38,250 That would increase their bargaining power within the 1274 01:04:38,250 --> 01:04:42,790 family compared to a situation which is more even. 1275 01:04:42,790 --> 01:04:47,352 People said that about China, today, that one kind of 1276 01:04:47,352 --> 01:04:51,130 strange consequences of the fact that the gender ratio are 1277 01:04:51,130 --> 01:04:54,140 completely skewed, due to selective abortion and things 1278 01:04:54,140 --> 01:04:57,040 like, it means that women, actually, are in short supply. 1279 01:04:57,040 --> 01:05:02,090 And therefore, it's getting easier to marry them. 1280 01:05:02,090 --> 01:05:05,210 And they have a lot of bargaining power within the 1281 01:05:05,210 --> 01:05:09,090 family, because there's no substitute for them. 1282 01:05:09,090 --> 01:05:11,450 So all of these things would matter. 1283 01:05:11,450 --> 01:05:18,120 So let's focus on one, for the purpose of this, which is how 1284 01:05:18,120 --> 01:05:21,500 much money women and men bring to the family, and, in 1285 01:05:21,500 --> 01:05:24,810 particular, how much money they would have access to 1286 01:05:24,810 --> 01:05:26,190 should they leave the marriage. 1287 01:05:29,080 --> 01:05:32,640 That's going to matter for family decisions. 1288 01:05:32,640 --> 01:05:37,160 So policy makers are very convinced it's the case. 1289 01:05:37,160 --> 01:05:40,460 So for example, all the cash transfer programs that exist 1290 01:05:40,460 --> 01:05:45,160 in Latin America, to encourage schooling and health-- 1291 01:05:45,160 --> 01:05:47,685 like the PROGRESA program in Mexico, that we discussed in 1292 01:05:47,685 --> 01:05:49,220 the case of education-- 1293 01:05:49,220 --> 01:05:51,570 they tend to always give the money to women. 1294 01:05:51,570 --> 01:05:55,190 On the ground that A, women will make a bigger use of the 1295 01:05:55,190 --> 01:05:57,830 money, and B, that's going to increase their bargaining 1296 01:05:57,830 --> 01:06:00,410 power and, therefore, increase all decisions. 1297 01:06:00,410 --> 01:06:05,720 So there is very benign sexism going on around in the policy 1298 01:06:05,720 --> 01:06:08,900 world that women are generally better people, which I don't 1299 01:06:08,900 --> 01:06:10,160 necessarily disagree with. 1300 01:06:10,160 --> 01:06:14,820 But it should be looked into a little bit. 1301 01:06:14,820 --> 01:06:17,680 So another one that is very much inspired by this is 1302 01:06:17,680 --> 01:06:18,800 microcredit. 1303 01:06:18,800 --> 01:06:21,000 Microcredit claims, the world-wide, 1304 01:06:21,000 --> 01:06:22,780 are almost all women. 1305 01:06:22,780 --> 01:06:25,370 Part of the reason is that women are believed to be 1306 01:06:25,370 --> 01:06:27,350 better at repaying. 1307 01:06:27,350 --> 01:06:29,530 But part of the reason is also that it was believed that it 1308 01:06:29,530 --> 01:06:33,360 was a way to introduce some changes, within their 1309 01:06:33,360 --> 01:06:35,780 families, by giving power to the family. 1310 01:06:35,780 --> 01:06:39,470 So does it matter who gets money or doesn't get money? 1311 01:06:39,470 --> 01:06:43,000 I want to give one example to prove my point that women are 1312 01:06:43,000 --> 01:06:45,760 better people than men. 1313 01:06:45,760 --> 01:06:50,840 Looking at one example where money went into the household, 1314 01:06:50,840 --> 01:06:56,560 either with the men or with women, which is the end of 1315 01:06:56,560 --> 01:06:59,930 Apartheid in South Africa. 1316 01:06:59,930 --> 01:07:02,150 They took a pension program, which existed for white 1317 01:07:02,150 --> 01:07:05,270 people, which was, for white people, a very small pension 1318 01:07:05,270 --> 01:07:10,480 program, in that it was not a lot of money for them. 1319 01:07:10,480 --> 01:07:13,910 And also, almost no one was eligible, because you couldn't 1320 01:07:13,910 --> 01:07:16,070 be eligible if you had a private pension. 1321 01:07:16,070 --> 01:07:18,050 And most people have private pensions. 1322 01:07:18,050 --> 01:07:21,820 So it was smallish program on the side, a little bit like we 1323 01:07:21,820 --> 01:07:24,700 have here with minimum, old-age security. 1324 01:07:24,700 --> 01:07:28,640 But at the end of Apartheid, the new government had to make 1325 01:07:28,640 --> 01:07:31,970 a decision about, should we can the program or should we 1326 01:07:31,970 --> 01:07:33,870 expand it to the black South Africans? 1327 01:07:33,870 --> 01:07:36,970 Of course, they're not going to keep two separate programs. 1328 01:07:36,970 --> 01:07:41,380 And they decided to expand it, because it was believed, 1329 01:07:41,380 --> 01:07:44,600 rightly, that the older people, who would be eligible 1330 01:07:44,600 --> 01:07:47,240 for this program, are precisely the people who 1331 01:07:47,240 --> 01:07:50,790 suffered the most from Apartheid, for 1332 01:07:50,790 --> 01:07:52,570 their entire lifetime. 1333 01:07:52,570 --> 01:07:54,210 So they expanded the program. 1334 01:07:54,210 --> 01:07:58,650 And every man above 65 and every woman about 60 is 1335 01:07:58,650 --> 01:08:00,770 eligible for the program unless they 1336 01:08:00,770 --> 01:08:01,560 have a private pension. 1337 01:08:01,560 --> 01:08:04,220 And very few of the black people had a private pension. 1338 01:08:04,220 --> 01:08:07,160 So about 85 to 90% percent of people who are eligible 1339 01:08:07,160 --> 01:08:08,950 actually get the program. 1340 01:08:08,950 --> 01:08:11,200 So the old-age pension program, which is still there 1341 01:08:11,200 --> 01:08:13,800 today, it's a huge, huge deal. 1342 01:08:13,800 --> 01:08:15,220 It's a lot of money. 1343 01:08:15,220 --> 01:08:18,470 And it's very important for the families. 1344 01:08:18,470 --> 01:08:22,880 And, of course, a lot of the older people don't live alone. 1345 01:08:22,880 --> 01:08:25,370 They live with children or they live with grandchildren. 1346 01:08:25,370 --> 01:08:28,460 And in particular, in South Africa, it is quite frequent 1347 01:08:28,460 --> 01:08:32,680 that you have skip-generation households, where you have a 1348 01:08:32,680 --> 01:08:36,210 grandparent, no parent, and the kids. 1349 01:08:36,210 --> 01:08:40,290 This is for two reasons, one is HIV/AIDS, which has killed 1350 01:08:40,290 --> 01:08:42,460 a lot of the middle generation. 1351 01:08:42,460 --> 01:08:46,979 And the other is the way the society was structured under 1352 01:08:46,979 --> 01:08:49,870 Apartheid, people couldn't bring their family when they 1353 01:08:49,870 --> 01:08:51,529 went to work in the city. 1354 01:08:51,529 --> 01:08:56,739 So they went alone, and they left the kids behind with the 1355 01:08:56,739 --> 01:08:57,689 grandparents. 1356 01:08:57,689 --> 01:09:01,520 So for this reason, about a quarter of kids, aged 0 to 5, 1357 01:09:01,520 --> 01:09:03,550 live with a grandparent. 1358 01:09:03,550 --> 01:09:05,314 And that grandparent can either be a man 1359 01:09:05,314 --> 01:09:08,170 or it can be a woman. 1360 01:09:08,170 --> 01:09:10,540 And the question we can ask is, does it make a difference, 1361 01:09:10,540 --> 01:09:13,910 when this pension program comes in, in 1993, if you 1362 01:09:13,910 --> 01:09:17,310 happen to live with no one or with a grandparent, who 1363 01:09:17,310 --> 01:09:21,779 becomes eligible, or with a grandfather or a grandmother? 1364 01:09:21,779 --> 01:09:29,060 So of course, one problem is that the families who have a 1365 01:09:29,060 --> 01:09:33,359 grandfather are different from the family who don't have a 1366 01:09:33,359 --> 01:09:36,160 grandfather. 1367 01:09:36,160 --> 01:09:47,204 So what we can look at is, let's take a sample of kids 1368 01:09:47,204 --> 01:09:54,980 who live with a grandfather, but let's compare the one who 1369 01:09:54,980 --> 01:09:59,030 lives with a grandfather who is, let's say, 55, versus one 1370 01:09:59,030 --> 01:10:04,920 who is 65 or a grandfather who is 50 to 64 versus 66. 1371 01:10:04,920 --> 01:10:07,780 So in both cases, there is a grandfather. 1372 01:10:07,780 --> 01:10:10,090 But one is just short of being eligible. 1373 01:10:10,090 --> 01:10:12,770 And one is just eligible. 1374 01:10:12,770 --> 01:10:15,880 So other characteristics of the family should be more or 1375 01:10:15,880 --> 01:10:16,990 less constant. 1376 01:10:16,990 --> 01:10:19,360 It just happened that your age is a good thing. 1377 01:10:19,360 --> 01:10:21,600 And of course, people couldn't really predict that this 1378 01:10:21,600 --> 01:10:25,780 pension was coming, so the kid lives with a grandfather and, 1379 01:10:25,780 --> 01:10:28,920 oh, he becomes just eligible or is not yet eligible. 1380 01:10:28,920 --> 01:10:32,520 The same thing with the grandmother but around 60. 1381 01:10:32,520 --> 01:10:34,290 So this is what this is doing. 1382 01:10:34,290 --> 01:10:36,360 This is looking at weight for height. 1383 01:10:36,360 --> 01:10:38,485 So this is short trend measure of nutrition. 1384 01:10:43,095 --> 01:10:44,880 As soon as you start eating better, you 1385 01:10:44,880 --> 01:10:47,190 weight for height improves. 1386 01:10:47,190 --> 01:10:52,510 And in orange, we have the boys, and, in 1387 01:10:52,510 --> 01:10:54,980 yellow, we have the girls. 1388 01:10:54,980 --> 01:10:58,010 And you can see that. 1389 01:10:58,010 --> 01:10:59,000 So what can you see? 1390 01:10:59,000 --> 01:11:00,743 Can you interpret this? 1391 01:11:03,550 --> 01:11:07,644 AUDIENCE: So this bar is higher for eligible 1392 01:11:07,644 --> 01:11:12,370 grandmothers that took all their grandsons in. 1393 01:11:12,370 --> 01:11:13,200 PROFESSOR: [INAUDIBLE] 1394 01:11:13,200 --> 01:11:15,088 AUDIENCE: Heavier grandsons and granddaughters. 1395 01:11:15,088 --> 01:11:19,336 And my guess is that, because they're eligible earlier, they 1396 01:11:19,336 --> 01:11:22,130 have more to give earlier. 1397 01:11:22,130 --> 01:11:23,090 PROFESSOR: So this is right after 1398 01:11:23,090 --> 01:11:26,660 eligibility in both cases. 1399 01:11:26,660 --> 01:11:28,710 So we don't have a longer time. 1400 01:11:28,710 --> 01:11:29,080 Yeah? 1401 01:11:29,080 --> 01:11:32,050 AUDIENCE: Well, it's just to show that grandmothers seem to 1402 01:11:32,050 --> 01:11:34,525 care more about their grandchildren than 1403 01:11:34,525 --> 01:11:37,495 grandfathers do when they become eligible. 1404 01:11:37,495 --> 01:11:40,465 And then you can also see the bias. 1405 01:11:40,465 --> 01:11:44,920 So for grandmothers, it seems like they treat girls better. 1406 01:11:44,920 --> 01:11:48,880 And grandfathers also treat girls better. 1407 01:11:48,880 --> 01:11:50,960 PROFESSOR: So, basically, we don't have much for the 1408 01:11:50,960 --> 01:11:52,950 grandfathers. 1409 01:11:52,950 --> 01:11:56,080 So the grandfather doesn't seem to be using the money to 1410 01:11:56,080 --> 01:11:57,760 be feeding the kids. 1411 01:11:57,760 --> 01:12:01,210 And we have up a much larger effect for the grandmothers, 1412 01:12:01,210 --> 01:12:04,040 particularly for the girls. 1413 01:12:04,040 --> 01:12:06,750 And some effects for the boys, but only half the size. 1414 01:12:06,750 --> 01:12:10,102 AUDIENCE: Now why is it the case that the grandsons are 1415 01:12:10,102 --> 01:12:15,310 worse off when their eligible grandfathers [INAUDIBLE]? 1416 01:12:15,310 --> 01:12:17,650 PROFESSOR: So if we had the confidence interval bar on 1417 01:12:17,650 --> 01:12:20,310 this graph, you would see that this is not significant. 1418 01:12:20,310 --> 01:12:23,830 So this is kind of same-ish. 1419 01:12:23,830 --> 01:12:26,330 No eligible grandparents and the bars for eligible 1420 01:12:26,330 --> 01:12:29,340 grandfather are around the same magnitude. 1421 01:12:29,340 --> 01:12:32,010 What is really different is the girls with the 1422 01:12:32,010 --> 01:12:33,620 grandmother. 1423 01:12:33,620 --> 01:12:36,584 And the boys to some extent, but it's about half the size. 1424 01:12:36,584 --> 01:12:39,795 AUDIENCE: So could there be an effect in that, when you have 1425 01:12:39,795 --> 01:12:41,928 an eligible grandfather, they're older, so that they'll 1426 01:12:41,928 --> 01:12:43,560 have to spend more of their pension on 1427 01:12:43,560 --> 01:12:45,340 like their own stuff? 1428 01:12:45,340 --> 01:12:45,980 PROFESSOR: Yes. 1429 01:12:45,980 --> 01:12:50,280 So that could be that. 1430 01:12:50,280 --> 01:12:51,170 The grandfather is older. 1431 01:12:51,170 --> 01:12:55,050 And also, men live less longer than women. 1432 01:12:55,050 --> 01:13:01,960 So actually, a 65-year-old man is quite a bit older than a 1433 01:13:01,960 --> 01:13:03,280 60-year-old woman. 1434 01:13:03,280 --> 01:13:05,630 So that could be a reason. 1435 01:13:05,630 --> 01:13:08,130 So we can start looking at this kind 1436 01:13:08,130 --> 01:13:09,700 of stuff, for example. 1437 01:13:09,700 --> 01:13:12,400 Also, they might want to save more of their pension if it's 1438 01:13:12,400 --> 01:13:15,670 a grandfather, because he's going to die very soon. 1439 01:13:15,670 --> 01:13:18,550 So it's more like a transitory income shock than a permanent 1440 01:13:18,550 --> 01:13:19,090 income shock. 1441 01:13:19,090 --> 01:13:21,380 So you might want to say, let's save it. 1442 01:13:21,380 --> 01:13:23,850 So we looked at saving, too, to control. 1443 01:13:23,850 --> 01:13:26,370 And we couldn't find a difference in savings. 1444 01:13:26,370 --> 01:13:27,610 The savings seem to be the same. 1445 01:13:27,610 --> 01:13:34,560 But you're right, it could be explained by that, that the 1446 01:13:34,560 --> 01:13:36,210 grandfathers need to be spending the money on 1447 01:13:36,210 --> 01:13:38,408 themselves, where the grandmother doesn't need to. 1448 01:13:38,408 --> 01:13:41,645 AUDIENCE: Also, could it potentially be the case that 1449 01:13:41,645 --> 01:13:45,380 the grandmothers may be alone in the household, with their 1450 01:13:45,380 --> 01:13:48,119 grandchildren, therefore they would not have to spend 1451 01:13:48,119 --> 01:13:50,360 anything on other people. 1452 01:13:50,360 --> 01:13:52,850 Whereas, maybe in the culture, it's more likely that a 1453 01:13:52,850 --> 01:13:56,114 grandfather has to take care, also, of his wife, therefore 1454 01:13:56,114 --> 01:13:59,036 when he becomes eligible, there is a greater fraction of 1455 01:13:59,036 --> 01:14:01,471 his income that also goes to a wife. 1456 01:14:01,471 --> 01:14:02,445 PROFESSOR: That's a good question. 1457 01:14:02,445 --> 01:14:03,730 That can be tested. 1458 01:14:03,730 --> 01:14:06,080 Because you can look at all of the family size 1459 01:14:06,080 --> 01:14:07,360 configurations. 1460 01:14:07,360 --> 01:14:12,060 And you seem to find these difference regardless of who 1461 01:14:12,060 --> 01:14:12,840 is who is who. 1462 01:14:12,840 --> 01:14:16,400 What is interesting, which helps answering your question, 1463 01:14:16,400 --> 01:14:20,230 is the grandmother could be the mother's mother or could 1464 01:14:20,230 --> 01:14:22,460 be the father's mother. 1465 01:14:22,460 --> 01:14:26,000 So you can look at, the pension goes to the mother's 1466 01:14:26,000 --> 01:14:29,670 mother or to mother's father or to father's mother or to 1467 01:14:29,670 --> 01:14:30,920 the father's father. 1468 01:14:33,900 --> 01:14:36,280 Which one do you think matters? 1469 01:14:36,280 --> 01:14:39,920 Under your hypothesis, if it's only due to the health issue, 1470 01:14:39,920 --> 01:14:42,930 what should we see? 1471 01:14:42,930 --> 01:14:45,780 That mother's mothers and mother's fathers should be the 1472 01:14:45,780 --> 01:14:48,960 same and then the father would be lower, because they need 1473 01:14:48,960 --> 01:14:50,600 more money. 1474 01:14:50,600 --> 01:14:52,800 And then, in practice, what do we see? 1475 01:14:52,800 --> 01:14:57,020 These high bar is only their mother's mother. 1476 01:14:57,020 --> 01:15:00,250 So it's only the pension translating to higher health 1477 01:15:00,250 --> 01:15:04,310 of the kids if it's a maternal grandmother of a girl. 1478 01:15:04,310 --> 01:15:06,960 So that seemed to have something to do with these 1479 01:15:06,960 --> 01:15:08,830 gender lines. 1480 01:15:08,830 --> 01:15:11,130 And that doesn't mean that the grandmother doesn't do other 1481 01:15:11,130 --> 01:15:12,210 things for the kids. 1482 01:15:12,210 --> 01:15:16,220 But in terms of how they decide to feed the kids, it's 1483 01:15:16,220 --> 01:15:20,040 really the identity of the kid that matters a lot. 1484 01:15:20,040 --> 01:15:28,750 So let me show you one last thing, which is these results 1485 01:15:28,750 --> 01:15:32,330 seem to show, relatively clearly, that the family is 1486 01:15:32,330 --> 01:15:33,470 not unitary. 1487 01:15:33,470 --> 01:15:37,880 So who brings in the resources matters. 1488 01:15:37,880 --> 01:15:40,310 They actually do not show that women are better people than 1489 01:15:40,310 --> 01:15:41,940 men, because we don't know what men do. 1490 01:15:41,940 --> 01:15:44,920 Maybe they spend more on the education of the children. 1491 01:15:44,920 --> 01:15:46,756 I only showed you a very partial 1492 01:15:46,756 --> 01:15:49,330 window into the family. 1493 01:15:49,330 --> 01:15:51,940 So the only conclusion we can take from this is that 1494 01:15:51,940 --> 01:15:53,470 families are not unitary. 1495 01:15:53,470 --> 01:15:55,220 So the next question we might want to ask 1496 01:15:55,220 --> 01:15:56,950 is, are they efficient? 1497 01:15:56,950 --> 01:16:01,170 In that, they are bargaining with each other, but are they 1498 01:16:01,170 --> 01:16:03,790 bargaining in a way that they are first maximizing the size 1499 01:16:03,790 --> 01:16:07,650 of the pie and then deciding to split the pie? 1500 01:16:07,650 --> 01:16:11,640 Or are they bargaining in a way that the women would be 1501 01:16:11,640 --> 01:16:15,230 willing to reduce the pie as long as she could get a higher 1502 01:16:15,230 --> 01:16:16,110 share of it? 1503 01:16:16,110 --> 01:16:18,230 You understand the difference? 1504 01:16:18,230 --> 01:16:20,950 So two people who would just meet and who don't know each 1505 01:16:20,950 --> 01:16:23,860 other, will tend to interact with each other like in the 1506 01:16:23,860 --> 01:16:26,020 prisoner's dilemma, that you might have seen in game 1507 01:16:26,020 --> 01:16:29,420 theory, which is they will try to grab as much of the pie as 1508 01:16:29,420 --> 01:16:32,730 they can, even if it makes the pie lower. 1509 01:16:32,730 --> 01:16:36,350 But two people who know each other very well, who are in it 1510 01:16:36,350 --> 01:16:38,850 for the longer run, might decide that cooperating to 1511 01:16:38,850 --> 01:16:42,020 make the size of the pie the biggest possible is the first 1512 01:16:42,020 --> 01:16:45,590 thing they should do and then think about how to share it. 1513 01:16:45,590 --> 01:16:48,870 And this is actually an idea we can test, because 1514 01:16:48,870 --> 01:16:51,490 households are engaged not only in consumption but also 1515 01:16:51,490 --> 01:16:53,710 in production. 1516 01:16:53,710 --> 01:16:58,100 So if the direct analogy is to first maximize the size of the 1517 01:16:58,100 --> 01:17:02,880 pie and then decide how you're going divide it, are they 1518 01:17:02,880 --> 01:17:06,530 maximizing the production that they can have. 1519 01:17:06,530 --> 01:17:10,250 And one way to look at that is farming, because in a lot of 1520 01:17:10,250 --> 01:17:14,570 countries women and men grow different plots of land. 1521 01:17:14,570 --> 01:17:19,010 So in Burkina Faso, the study, by Chris Udry, looking at 1522 01:17:19,010 --> 01:17:22,600 households in which each member is farming their own 1523 01:17:22,600 --> 01:17:24,490 little plot. 1524 01:17:24,490 --> 01:17:28,650 And an efficient household should first use the inputs 1525 01:17:28,650 --> 01:17:32,390 the most efficiently possible and then decide on the basis 1526 01:17:32,390 --> 01:17:36,120 of people's bargaining power how to share the product. 1527 01:17:39,350 --> 01:17:42,450 Once we condition for the quality of the plot and the 1528 01:17:42,450 --> 01:17:45,150 crop planted, we shouldn't see that the women get fewer 1529 01:17:45,150 --> 01:17:50,400 fertilizer or less labor or less input or less seeds. 1530 01:17:50,400 --> 01:17:55,330 And in fact, instead this paper tested it and found that 1531 01:17:55,330 --> 01:17:59,380 women get significantly less of every input, including 1532 01:17:59,380 --> 01:18:00,540 fertilizer. 1533 01:18:00,540 --> 01:18:03,190 And fertilizer has very concave returns. 1534 01:18:03,190 --> 01:18:06,200 So there is really no point of putting more fertilizer on a 1535 01:18:06,200 --> 01:18:09,600 field of the men, when you could put a little bit on the 1536 01:18:09,600 --> 01:18:12,520 field of a woman, from the point of view of family 1537 01:18:12,520 --> 01:18:13,540 production. 1538 01:18:13,540 --> 01:18:14,870 And that's not what you see. 1539 01:18:14,870 --> 01:18:17,620 You see all of the fertilizer being spent on the men's field 1540 01:18:17,620 --> 01:18:19,570 and very little on the women's fields. 1541 01:18:19,570 --> 01:18:23,520 Which suggests that the family is not efficient, because they 1542 01:18:23,520 --> 01:18:26,610 could become richer by just reallocating resources among 1543 01:18:26,610 --> 01:18:27,890 themselves. 1544 01:18:27,890 --> 01:18:30,650 So this suggests that the family is not only not 1545 01:18:30,650 --> 01:18:37,410 unitary, but it is also not fully efficient. 1546 01:18:37,410 --> 01:18:39,850 Which leads us to this question, what role does the 1547 01:18:39,850 --> 01:18:41,460 family play? 1548 01:18:41,460 --> 01:18:43,750 What is the family inefficient? 1549 01:18:43,750 --> 01:18:46,550 And what implications does have to policy, which we'll 1550 01:18:46,550 --> 01:18:50,630 take up after you become all refreshed from the vacation.