Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=

# What Drives Child Penalties? Evidence from a Global Organization

Authors: Jessica Bracco, Yves Froude, Maria F. Pinto, Bob Rijkers, Michele Valsecchi  
Discussant: None  
Video: https://www.youtube.com/watch?v=1kb3a99sA-E&t=14221s


## Talk (03:57:01 – 04:10:07)

[03:57:01] [applause] >> Great. Thank you so much, Rachel. Um, >> let's see.  
[03:57:15] >> Actually, do you know what? >> Do you know his name?  
[03:57:19] >> I don't know his name. >> Hello, my name is Bob Rikers.  
[03:57:22] >> Oh, there we go. >> I'm uh very excited to be here today.  
[03:57:26] So, thanks for including us in the program. And I'm going to present a paper on child penalties which is joint work with Jessica Proco who's in the audience ER Florentia Pinto and Michaela Fzi and the motivation for this work is that  
[03:57:41] you know as we all know child penalties are extremely prevalent and widely documented but there's comparatively less evidence on how they emerge within organizations. So a lot of work shows that models are just labor supply and  
[03:57:54] may switch employers but they may arise even for women who remain full-time employed with the same employer through different mechanisms. It's possible that the tasks they perform change. It is  
[03:58:08] possible that their output is impacted by parenthood. It's also possible that they choose different work arrangements and different career trajectories.  
[03:58:18] And so what we're going to do in this paper is use unusually detailed administrative data from my employer, the World Bank Group, to shed light on some of these mechanisms. And in particular, uh I hope that by the end  
[03:58:33] I've convinced you that there is evidence of a within organization tradeoff between sorting, sorry, between flexibility and uh career advancement.  
[03:58:46] And in fact, women sort into jobs that are less conducive to promotions but more compatible with caregiving responsibilities.  
[03:58:56] So, let me preview the main findings. Like in many other uh instances, we find that there's a child penalty for mothers but none for fathers.  
[03:59:07] This penalty is an advancement penalty. It arises almost entirely because women get passed over for for promotions. It's small relative to to other settings.  
[03:59:20] It's roughly 3.1% in total uh across the the first six years after child birth.  
[03:59:27] And we also are going to show and this is perhaps more novel is that after child birth, mothers take on fewer leadership responsibilities. They're less likely to initiate development projects.  
[03:59:40] they're less likely to supervise external consultants and they reduce their travel much more sharply than than men do. But if we look at the quality of realized output, in fact, we find no evidence of significant deterioration  
[03:59:55] therein. So it seems that responsibilities shift more than than performance. And then third, and I'm going to show you that, you know, mod's career mobility slows down and  
[04:00:08] increasingly they sort away from jobs where promotions are very prevalent into jobs where homebased work is more prevalent and traveling is less frequent.  
[04:00:22] So, I'm going to focus on the World Bank Group. It's a global employer. For those of you who don't know, it's the largest [laughter] multilateral development lenders and you know most of our financing is provided through through projects and that's  
[04:00:35] that's a very important component of our work. Of course also it's advantageous that we have standardized grades pay and a common promotion system and in fact earnings can rise basically through two channels. One is every year all of us  
[04:00:48] get a performance rating and and based on that rating which is you know varies from one to five like you get a salary increase that's permanent and secondly in some years some of us get promoted to higher grades. There's almost no  
[04:01:02] part-time work and there's no compensation for overtime. So we don't have to worry that the changes in earnings we're going to document reflect changes in hours worked. So then we're going to put together like extremely rich administrative records from for a  
[04:01:16] 36 year period. And we observe sort of the job people work in, how much they earn, like what ratings they get for a subset of this period each each year, their family structure that we use sort  
[04:01:29] of to deduce information on whether they have a child or not. And perhaps more novel, we're going to also link these data to pro sorry to project outcomes.  
[04:01:38] So we know what projects uh staff are managing and then whether those project eventually are successful or whether they incur delays for for people who work in research. We're also going to look at like do they produce working papers? How often are those working  
[04:01:53] papers downloaded? How much do people travel? Uh where do they travel to? And and for the last 10 years we can also see do they work from home? Yes or no.  
[04:02:04] uh and so we're going to focus on on staff based in in headquarters who we can observe both before and after uh child birth. And what's also important is that the World Bank group is explicitly committed to to gender equity. And in fact, in the paper, we show that there's very little evidence  
[04:02:19] of sort of differential evaluation based on observable performance by managers.  
[04:02:23] So we don't have to be concerned here uh that sort of discrimination or or the preference of the employer or a big generation of of the gender gaps I'm going to document.  
[04:02:34] So the methodology we use is extremely standard. We're going to use an event study around the birth of the first uh children that we're going to estimate separately for fathers and mothers using as a control group people who will  
[04:02:49] eventually become parents. So that identification comes solely from variation in the timing of of the uh birth of the first children. And we're going to scale outcomes  
[04:03:00] uh by their sort of pre mean and express them in percentage changes to facilitate uh interpretation.  
[04:03:10] I should also say that we find very little evidence of differential treatment attrition of parents versus non-parents based on observables.  
[04:03:20] So our first main finding is that like in many other organizations there is a child penalty at the World Bank for for mothers. So this event study shows you the evolution of of parents earnings after they have their first child in  
[04:03:33] blue for men and in orange for women. And clearly there's no impact for for fathers but women's earnings seem to decline. And on average this is a  
[04:03:44] reduction of 3.1% um in the first six years after first having a child. So how is this penalty generated? Well, remember that earnings can increase in two ways. One is promotions, the other  
[04:03:57] is sort of your annual pay increases. In fact, it seems that most of the action is coming through promotions. So, this graph shows you the evolution of great progression. Again, using the same framework that we used to look at  
[04:04:10] earnings and you see that men's great progression. So, men's promotions are not impacted whereas women seem to suffer quite a big promotion penalty. In fact, if you look year by year, their likelihood of getting promoted drops by more than 50% and after six years,  
[04:04:25] they're roughly a quarter of a grade behind men.  
[04:04:30] Sorry, behind women who don't haven't had a child yet. Then if we look at performance ratings, we see that there is a penalty in terms of annual performance ratings both for men and women. But actually for women, the these  
[04:04:45] ratings recover after about four years. uh whereas for men they don't. So, you know, whatever penalty [laughter] there is, we have to bit be a bit careful here because they're clearly pre-trend. So, let's let's perhaps not  
[04:04:58] put too much weight on on these results. But it seems unlikely that these performance ratings can explain this this gap that we see between men and women.  
[04:05:06] So, what do women do after they first have a child? Well, if we look at sort of the responsibilities they take on and we look at like whether or not they initiate a development project and that's a big deal at the World Bank, right? So, and it's also something that's highly costly. It takes a lot of  
[04:05:20] time. You have to like interact with client governments. You have to convince internal management that sort of this is a good idea. Then you have to like meet all sorts of safeguards. It generates a lot of visibility. It's highly valued in terms of like your your career  
[04:05:34] progression, but it's extremely time inensive and can be a bit stressful. And you see that women do much less of this after child birth. Whereas if anything by man for men the likelihood of of bringing a project to the board seems to go up. [clears throat]  
[04:05:49] Similarly if we look another at another leadership responsibility notably managing external consultants we also see that women do less of this after childirth whereas for men if anything  
[04:06:02] the opposite pattern prevails. But does this impact sort of the does motherhood or parenthood impact the quality of realized output? Well, it's hard to say but you know the few proxies  
[04:06:16] we have for this suggest that the answer is no. So if you look at project outcome ratings so are these projects that women manage are they eventually successful?  
[04:06:26] Um we see little evidence of a significant impact or significant decline in in the success rate of these projects. We also don't find any impacts on delays. When we look at research productivity, we don't find any impact on either the number of new working  
[04:06:41] papers produced or like research impact as proxy by the downloads of those working papers shown uh on the right hand side. So it really seems that sort of the task that that parents undertake shift but you know we don't find sort of  
[04:06:55] any strong evidence of impacts on on quality.  
[04:06:59] We do see however that women start to women in particular but all parents start to work very differently. So World Bank jobs tend to be extraordinarily travel intensive. The average staff member spends more than a month a year traveling.  
[04:07:13] And you see that sort of after giving um getting a child both fathers and mothers reduce their travel but women do so much more than men. And in fact, you know, women reduce their travel by  
[04:07:27] roughly more than two weeks, men by roughly one one week on average. They also simultaneously shift towards working more from home, but that shift is not sort of statistically significant.  
[04:07:41] And so that brings me sort of to to the third and final finding is that as sort of family demands increase. We see that you know both fathers and mothers  
[04:07:54] sort out of promotion intensive jobs. So these are jobs sort of where the rate of at which we're promoted uh uh tends to be above the median. And this shift again is sharper for mothers than it is  
[04:08:07] for fathers. So where do they sort to? Well towards jobs that offer more flexibility. So models in particular are more likely to sort into jobs that are  
[04:08:20] amenable to being uh performed from home. They also sort out of jobs that are highly travel inensive and these happen to be sort of the jobs where  
[04:08:33] child penalties are comparatively low. So this evidence is very much consistent with a within organization tradeoff between advancement and and flexibility.  
[04:08:46] So to summarize, so I hope I've convinced you that sort of at least at the World Bank, the motherhood penalty is primarily a career progression  
[04:08:57] penalty. So women seem to get promoted at a much slower rate.  
[04:09:05] the tasks they take on and how they work also seems to change and in particular women are less likely to undertake leadership responsibilities that tend to be quite visible uh than than men. So  
[04:09:19] for men actually there's no change in in the leadership responsibilities they they perform.  
[04:09:26] Both men and women reduce travel but women do so at a much higher rate than men. uh at the same time we don't detect any evidence that the quality of the outputs that they continue to produce is meaningfully impacted by by parenthood.  
[04:09:41] And and then third uh we show evidence that's consistent with sort of sorting into jobs that are more compatible with caregiving. Um but this comes at the  
[04:09:54] expense of slower career advancement. So that's it.  
[04:10:02] >> [applause] >> All right. Thank you so much, Bob.