 So this is working all fine, so I'm Nekita, I'm from the Indian Statistical Institute and thank you very much for giving the opportunity to present this work which is about the role of digital job search technology in enabling access to work opportunities. And before I start, you know, with the evidence, I want to give you a context of where this work is really based. So in India what we have experienced is a significant improvement when we look at the GDP and the socioeconomic other parameters. But when it comes to the female labour force participation, it continues to remain very low and it is very, it's not just low, it has remained stagnant. Another key feature is that the urban female labour force participation is very low compared to that of the rural female labour force participation and for this reason we are going to be focusing on the urban sample. So the big question really is why are so few women in the labour market, right? So is it that they don't want to join the labour market? But the answer to that is no because what we see is that there is a significant latent demand for work by women. If we look at the national sample survey, we see that 33% of the women that are currently not in the labour market, they want to be, they are looking for work and they want to find work. Also another reason why women are not in the labour force is coming from the fact that women have preference for work which is located close to where they are staying. And this preference for work is coming from the internalization of the fact that women have limited physical mobility, they are constrained by the social norms and the home production responsibilities which is majorly borne by the women in these households. Also there are safety concerns associated with women who have to travel longer distances and more so when there are very few other women that are travelling with them. Additionally when it comes to awareness and information about these kind of job opportunity that is missing especially for women because they do not have access to weak ties and these weak ties are very important when it comes to referral for job opportunities and also getting access to information about the presence of these job opportunities. And it is given this background that digital labour market platforms can really play a role because what they can do is reduce the job search cost. And since we expect the job search cost to be higher for women, we expect the benefits coming from adoption of these digital job market platforms to be higher for women. And these platforms, they reduce the matching frictions and therefore can be really beneficial in not only matching people who are not in the labour market but improving the quality of the match for people who are already in the labour market. But this does not go without a caveat which is that benefits may not actually accrue to women. And the reason for that again is starting with the first difference that these gender differences they are not only in terms of the labour force participation even when it comes to the adoption of these technologies and access to information about these new technology platforms that can itself be gendered. Now there is additionally a possibility of male backlash especially in the Indian context where there is a prevalent norm wherein men are the ones who are the breadwinners of the household while women take care of the children. Another thing that has been brought about with the recent strand of literature is about the joint decision making. So if husband and wife are jointly deciding who should be going out and working in the labour market and who stays back and takes care of the children and the domestic chores given that women have higher returns to home production compared to the returns in the job market and in fact that was mentioned in the presentation about the gender wage gap. So that might actually lead to women staying back and withdrawing from the labour market. And this is where the innovation that we have tried in this work is to bring in the role of networks of women because if women are treated along with their networks and they are jointly trying to overcome the restrictions that are put in by these social norms it is going to be easier to do it when they are doing it along with the peers that they are regularly interacting compared to doing it independently or in silos. But here again the problem is that when we talk about the social networks of women most of them are home bound and it is possible that these networks instead of acting as an enabler might in fact further constrain their labour force participation by reinforcing the prevalent social norms. And this is what we are trying to do in this study which is joint work with Farzana Fridhi at Indian Statistical Institute and Amrita Dhillon and Sanchari Roy at King's College London. And the idea is really to have you know what we did was collected experimental evidence by a randomized controlled trial that was done in urban India. And the idea is to understand how getting access to job search technology can improve you know access to information about the existing jobs and improve the quality of the matches. And the idea is to also look at the interaction how the social norms can play a role especially from a gender perspective. The research questions that we are interested in addressing here is whether whether access to the technology can enhance the labour market participation. The role that the networks can play and how the social norms interact with these networks. But for today's presentation I'm going to be focusing on the first bullet point here which is how do access to information about the technology help in enhancing the labour force participation of women as well as men. And how we are doing this is designing an intervention that is able to address the demand as well as the supply side factors in the labour market by matching them to potential employers that are locally available. So these are the blue collar jobs that are available close to where these people are staying. So these are local jobs and the way this platform works is it's very similar to Uber because depending on where the demand for labour is and the supply for labour is it matches them on the basis of that. And the way this intervention is working is by reducing the job search cost because the people who are going to be onboarding the portal they don't have to pay any fee to be on the portal and are matched to potential jobs. Now coming to the study sample so I'm not going to go into the detail of the randomization and how you know how we are going about with the sampling strategy here but just to give you an idea of where the sample is coming from we are focused on the poor urban households from Delhi and here is so this is the map of Delhi and we are looking at five districts in Delhi and from these five districts we have a sample of 108 polling stations that have been randomly selected not going to do the details of how we select them but these 108 polling stations were then assigned to one of the three treatment arms that we have and this gives us 36 clusters for each treatment arm. So the first treatment is having the matched husband and wife there being given the information about the portal they were informed that this is the portal which is going to match you to potential jobs are you interested in onboarding the portal so that's the first step that we're interested in whether they are interested in being part of these digital technology in the first place. The second thing was whether they actually registered after showing interest in the portal and on registration their preferences of the kind of jobs that they're looking for was recorded and they were matched to potential jobs and the next stage was to understand what kind of jobs they were matched to whether they accepted the jobs if they're not accepting the jobs why is it that they're not accepting these jobs. The second treatment arm brings in the network wherein we are giving this information to the matched husband and wife pair along with two of the wife's self-reported friends so this this is her social network and the idea is to harness the role of social network in not only adoption of the technology but also the usage of this new technology and in both these treatment arms we are going to be comparing the impact related to the peer control group where we are just collecting information on their labor market participation without giving them any information about the job portal. In our sample we have focused exclusively on married couples because the idea is to understand the intra household dynamics so we have the matched husband and wife pair and in this match husband and wife pair we have exclusively focused on the age group of 18 to 45 years because this is the age bracket where you know they are more active in the labor market and also women who are married women in this age bracket they're also going to have domestic job burdens including child care and elderly care responsibilities. Now given this sample coming to you a brief of the timeline of the study we started the data collection and the baseline collection in mid 2019 and we rolled out the intervention by end of 2019 and right after that we had the nationwide pandemic lockdown and as you all know India had one of the worst cases and in fact the strictest lockdown so when we started with the first end line which was six months after the lockdown was just uplifted in August 2020 we are not finding any significant impacts here because they were no new jobs that were created during this phase and we are for today's presentation I'm going to be exclusively focusing on the second end line which was performed after one year of rolling out the intervention and the intervention was you know giving information about the portal. Okay coming to the estimation strategy we are following a very standard estimation strategy and the two variables are the two coefficients that you should be looking out for are the betas here because they are going to give us the impact of the first treatment which is just the husband and wife pair without the network treatment and the second treatment which is the husband and wife along with the network and in all our specifications we are having controls for the baseline characteristics because this is the end code specification wherein we want to control for any differences at the baseline also given that this is a randomized control trial we have we have the balance checks and everything but I'm not going to go into those details for this presentation but what I do want to you know highlight here is in the what is the baseline characteristics of the sample that we are looking at and what we see is that 96% of the husbands in our sample are already working and only 24% of the wife in our sample are working so most of the women who we are serving they are not in that they are not currently engaged in the labor market and after 24% who are working we most of them are engaged in self-employed work so the self-employment is you know the work which they can do from that households right so most of them are operating from the premises of the household and this is again comes you know this again comes in when we look at the job preferences that they have so it's important and I'll you know point out why it's important to focus on the self-employment here so what we did was explicitly asking the husband as well as the wife about the kind of jobs that they want the women to do so the husband was asked the job that he wants the wife to be engaged in and the wife was asked about the job she wants to be engaged in and we see that there is you know there's a good symmetry in terms of their responses because both of them want to be engaged in solid work and home based work so this is why I was you know focusing on the self-employment you know because the current ongoing norm is to do work from where these most of these women are based that is acceptable so this is the ongoing norm in fact when we look at the people who don't want to do any work right so that is very close to zero so this means that there is a significant latent demand for work okay now coming to the results and so right now I'm going to be focusing only on the results on employment and the different labor market outcomes that we have and as I had mentioned before we don't find anything when we look at end line one when we look at in line two what we are finding is that for the husbands the probability of being in the labor market is going up by 4.4 point with no similar impact for the wife or for the treatment without the network so this impact is observed only when they are treated when the husband is treated with the network of the wife another interesting point to note here is that when we compare the coefficients when you compare the impact for treatment to versus treatment one we find that there is a significant difference between these coefficients for both the husband and the wife so clearly there are some network effects that are ongoing here now coming to the type of employment while I do not find anything when I look at the overall employment status of the wife when it's decomposed into types of employment being self-employment salaried work and wage labor there is a significant take up of self-employment by 4.5 percentage point by these wife so now again this brings us back to the social norms right because we found that women are already engaged in self-employment and the husbands as well as the wife want to do work which can be performed from home and even when they are provided with this intervention the wife that are treated along with the network they end up taking self-employment so they are you know even with the intervention and having excess to other job opportunities which are not self-employment they end up conforming to the norm of doing self-employment the husbands on the other hand are benefiting from the portal and taking up other employment now coming to what is happening on the intensive margin so similar to the results on the extensive margin I'm finding that their days their work days are going up by 55% and note here again that the coefficient for treatment with the network is significantly higher compared to treatment without the network and the same results when I look at the number of hours which are going up by 58% so clearly men when I when they are treated along with the network are taking part on the extensive as well as the intensive margin more and this raises a natural question as to what happens to the household welfare right it doesn't matter whether the husband works the wife works as long as the household income is going up so as a natural follow-up we look at what is happening to the monthly earnings of the individual members so what happens to the earning of the husband and the wife and what we find here is that the husbands earnings when they're treated along with the network they are more than doubling and again note here that treatment to coefficients are again higher so this means that the income in treatment to where the treatment is given along with the network is having higher earnings for the husband as well as the wife compared to the treatment where this is given this treatment is given without the network compared to the control group where no information was given another interesting point is it's not just about work because if you talk about sustainable development it's about having decent work opportunities and that is where this intervention was really you know able to shed some light because when we look at the kind the type of earnings that these husbands who are you know taking part more in the labor market so what's really going on because already 96% of them were engaged in the labor market and what we find is that they move out in both the treatment arms from p-straight and delivery straight work which are more vulnerable and unsecured job types into the more secure and stable salaried work so clearly there is an improvement in terms of the their engagement in the labor market and that is happening for both the treatment arms but okay so I had so the thing is the idea was to look at the mechanisms right so why do we observe what we are observing so there are two key things to take away from here that one thing is that a treatment to has higher effect compared to treatment one the second thing is treatment to which is treatment with the network works only for the husband and it doesn't work for the wife so why is that happening so we have multiple mechanisms to explain that and most of them were you know they were not able to justify this because we find that the husbands right so when it comes to the norms they were improving for both with the intervention for both the treatment arms and we are finding greater adoption of the technology when you are treated along with the network but coming to the question of why only husbands benefit from it why not why because wife have lower job search cost so theoretically we'd expect greater benefit there and the answer lies in the demand side the gender preferences that they had because when we look at the registration data we find that women are throwing a very narrow net in search of work because they want they restrict to very few job types they want to travel very short distance compared to their husband also another very interesting point here was in terms of the mismatch because when we look at the expectation right they are expecting ten thousand Indian rupees of from working on the portal but if you look at women who are already in the labor market this is the last I said when you look at women who are already in the labor market they are earning only four thousand five hundred so they are expecting over fifty percent higher income of higher earnings on joining the portal but it's because of this mismatch that they were not effort as many of job offers from the portal as compared to the husbands who are throwing a wider net in terms of the job profiles the distance they were willing to travel and also the salary expectations were very much aligned with what is being offered in the labor market and the key learning from the study is that these platforms they can help in improving their employment opportunities the earnings as well as a shift towards more secure jobs but it is important to incorporate the network that they have when you're thinking about policies to introduce these kind of digital job platforms and with this I conclude my presentation thank you very much for your attention.