 Welcome everyone, good morning or a good afternoon, not entirely sure where you are in the globe. Welcome to these second sessions of parallel, sessions of the day. I'm Hurt Kairu, I'm gonna be the chair in this session. And we have four super interesting presentations on livelihoods and inequality. I think we're supposed to have a fifth one and we are just hoping that the presenter is going to show up at some point. If we cannot locate it, locate him, it just means more time for a Q&A. So please use the Q&A part of the chat to ask questions to the presenters and we can have a bit of an interesting discussion at the end. That being said, I think we can start with the first presentation by Jemima. Hi everyone, I am Jemima Bada. Thank you all so much for making the time to attend my presentation today. In this presentation, I highlight how COVID-19 responses by the government of Ghana or GOG have been marginalizing for rural and farming populations using the Upper West Region as a case study. In the interest of time, I will skip the background information and go right to the inequalities of COVID-19 policies in Ghana. The full paper is published in the Journal of Agrarian Change and so the more detailed discussions can be found there. The Upper West, Greater Accra and Aschante Regions were among the first to record COVID-19 cases leading to partial lockdowns. As part of the lockdowns, the GOG banned airline travel and large gatherings. It also shut down schools and imposed restrictions on movement. While these strategies were to prevent public health challenges, they nonetheless resulted in weeks of hardship and uncertainty for many in the country. The lockdown was lifted in April 2020 but alternative interventions by the GOG were still marginalizing for agrarian groups. Agrarian communities are characterized by their rurality, low socioeconomic status and heavy dependence on subsistence farming. The Upper West Region is a mostly agrarian society and one of the poorest regions in Ghana due to a myriad of historical, geopolitical, environmental and other factors. Although other regions in Ghana record significant proportions of rural dwellers and poverty, the Upper West Region is overrepresented in both counts. Now, I will discuss each of the response strategies by the GOG and how these are marginalizing for agrarian societies. The strategies are spelled out on the slides so I will only talk about why these strategies are discriminatory. Strategy one was to limit and stop the importation of cases. In the early stages of the viral spread, much of the attention was on borders in southern Ghana. Testing and contact tracing were also limited to travelers arriving to the international airport in the national capital. During that period, the Hamle border which connects Upper West Region to Burkina Faso witnessed an increase in migrant and smuggling activities. Also, most inhabitants of the Upper West Region engage in internal migration to mostly rural areas of Ghana. Therefore, focusing slowly on travelers arriving to the international airport neglects less privileged travelers who journey through other means. Strategy two was to detect and contain cases. As stated earlier, contact tracing in Ghana was limited to affluent travelers. Despite evidence of other routes of transmission. Later, testing prioritized community hotspots. Focusing on identified community hotspots and imposing a lockdown in two metropolitan areas. While doing little to address a recorded case in Upper West Region, we are treating policy neglect of the region dating back to the colonial era. Moreover, encouraging enhanced hygiene in mandating quarantines and the wearing of masks presupposes the availability of spaces and resources to do so. This is not the case for the average Upper West Region dweller. This measure also does not account for the extra burdens that quarantines and enhanced hygiene add to women's workloads. Lastly, a COVID-19 tracker app is important but exclusionary for people in the Upper West Region due to high rates of electricity, lack of electricity and the fact that many in the region do not own mobile phones, much less smartphones. Strategy three was care of the sick. This strategy presupposes the existence of health facilities and workers and it's exclusionary for largely raw context such as the Upper West Region. Since first, many health workers refuse postings to the area and the few training in the region prefer to immigrate for better opportunities. Second, the doctor-patient ratio in Upper West Region is over three times lower than that of the national capital. Third, the few health facilities in the Upper West Region are under-resourced. As of June 2021, all COVID-19 testing facilities were in Southern Ghana. Fourth, it was revealed that Ghana had just 200 ventilators to serve its population of over 30 million. Estimation suggests that there may be 10 or fewer ventilators for the entire Upper West Region. Later, 10 extra ventilators were donated to the GOG, all of which were distributed within Southern Ghana. Fifth and finally, the effects of climate change imply more complex infectious disease outcomes in the Upper West Region. For instance, earlier in 2020, there was an outbreak of cerebrospinal meningitis in the Upper West Region with an estimated 40% fatality rate. Despite this, governmental attention continued to focus on the national COVID-19 crisis with little attention to the double epidemic in the Upper West Region. Strategy 4 deals with the impact on social and economic life. This intervention assumes a homogeneous living standard for populations across the country. Given the historical trends of poverty alleviation strategies in Ghana and the lack of details about the coronavirus alleviation program or the CAP, it is likely that the northern sectors would receive minor portions of the CAP. Moreover, relief for health workers is effective only if health workers are available and have medical supplies, which is often not the case in the Upper West Region. Additionally, only a few people in Upper West Region operate businesses, hence an option better suited to certain such as the Upper West Region with the subsidies for smallholder farmers. Importantly, the absorption of water bills and rebates for electricity is perhaps the most discriminatory of all measures for rural dwellers in Ghana, considering that most of these communities lack both electricity and portable water. Strategy 5 was to increase domestic capability and deepen self-reliance. The Upper West Region has very few industries and studies show that most new businesses prefer to operate in southern Ghana due to the area socioeconomic and ecological advantages. It is therefore possible that the Upper West Region would not benefit from this intervention and residents may likely need to migrate to take advantage of this intervention in southern Ghana. Potentially increasing out migrations and further widening inequalities between the northern and southern sectors. Similar to historical development patterns, it is also likely that no infectious disease centers will be set up in the Upper West Region. In conclusion, COVID-19 has highlighted the stark local and global inequalities with the sustained attention on urban centers and co-occupations. It is important to examine how agrarian populations experience a pandemic. It is my hope that most studies will direct attention to these experiences to better center the needs of vulnerable rural dwellers. Thank you all so much. Thank you, Jemima. It was super interesting that this, it was printed a presentation on how the ways that government tried to combat COVID-19 don't really reach everyone in every place in the country. So let's move on to the next presentation. Katy, did you manage to find the other presenters online? No, she's unfortunately online currently, so. Okay, no worries. So in that case, we are going to reach your presentation, please. Good day, everyone. In this talk, I'll be presenting some results from a recent paper examining how adolescents have been affected during the COVID-19 pandemic in four low and middle income countries. The data that we use in the paper comes from the Young Lives Longitudinal Study. Since the turn of the millennium, the study has been following two cohorts, the younger and older cohorts, in four low and middle income countries. Ethiopia, Peru, Vietnam, and the state of Andhra Pradesh and Telangana in India. The participants have been visited in person five times, most recently in 2016. But in 2020, in light of the COVID-19 pandemic, Young Lives implemented a three-part phone survey. The second call, from which most of the data in this paper comes from, was implemented between early August and mid-October, a time at which our younger and older cohorts were aged roughly 19 and 26, respectively. It's the 19-year-old cohorts that we focus on in this paper. As of mid-October 2020, the four Young Lives study countries had had very diverse experiences during the pandemic. As you can see in the figure on the screen, by the time that our second call was complete, the number of COVID-19 cases per capita differed dramatically between the countries, with Vietnam having been exceptionally successful at limiting the spread of the virus and Peru being one of the worst affected countries in the world in terms of both cases and deaths per capita. The question is then, given these diverse experiences, how have adolescents fed in the four countries? Analyzing educational outcomes, we found the ability to continue learning remotely during lockdowns varied greatly by country. In Vietnam and Peru, roughly 80% of the 19-year-olds successfully engage with their schoolteacher through in-person or virtual classes or online assignments. However, in India, this proportion dropped to only roughly 40%, and in Ethiopia, roughly only one in every 10 participants managed to engage with their teacher during the stay-at-home requirements. We also found that in Peru, the country worst affected by COVID-19 cases, roughly 16% of the younger cohorts who were engaged in some sort of formal education before the pandemic had dropped out or chosen not to re-enroll by mid-October 2020. Analyzing time use, we found that participants reported spending more time on childcare and performing more domestic work than before the pandemic. We also have found that households have tended to resort to more discriminatory gender roles in times of stress, as the increase in household and caring responsibilities has fallen disproportionately on females in all four countries, while males have tended to work more in the family business. For example, looking at the bottom of figure one on the screen, you can see that in Ethiopia, roughly 70% of adolescent females reported spending more time on domestic work compared to just 36% of adolescent males. Looking at food insecurity, we found that around 16% of adolescents reported that their household had run out of food on one or more occasions since the beginning of the pandemic in Ethiopia and in India. Comparing this to figure two on the screen, you can see that this marked a significant movement away from the existing trend of food insecurity with the proportion of households without food increasing by over 200% compared to 2016. Peru and Vietnam appear to have been less affected in terms of food insecurity. But beyond just descriptive statistics, one of the unique strengths of the multi-cohored young lives panel data is that it provides opportunity to compare the outcomes of the different cohorts when they're the same age, but at different points in time. Previous Young Lives Research, comparing the two cohorts at the same age, has shown that in the lead-up to 2020, the young cohort had achieved improvements in critical aspects of human development, such as height for age, school enrollment, and cognitive learning outcomes. Continuing this inter-cohored comparison. In this paper, we're interested in comparing the outcomes of the 19-year-olds during the pandemic with that of the older cohorts when they were the same age or roughly seven years prior in 2013. Formerly to do so, we utilize a different and different approach, but we allow for differential linear age trends between the two cohorts. The equation we use is at the bottom of the screen here. One of our main outcomes of interest in this is subjective well-being. In all young lives around, subjective well-being has been measured using this tantral self-anchoring scale. This asks respondents to visualize a ladder of nine steps with the bottom step representing the worst possible life and the top step representing the best possible life. Respondents will then ask to identify which step they think they presently stand on. Figure three on the screen here is showing the changes in the average step of the subjective well-being ladder over time for each cohort. And as you can see, it depicts that there's been a striking fall in the relative well-being of the younger cohorts when compared to the older cohort at the same age. Before 2020, the younger cohort had consistently higher well-being at the same ages of 12 and 15 in all four countries. This is now no longer the case in Ethiopia, India, and in Peru. The exception is Vietnam, the country that has been exceptionally successful at limiting the spread of COVID-19. Formerly analyzing the results using our difference-to-difference framework, we indeed find that the younger cohort have experienced a significant drop in well-being relative to the older cohorts in Ethiopia, India, and Peru, but that there's been no significant change in Vietnam. When thinking about the channels as to why this may be, the descriptive results presented earlier already point to a number of areas of concern that could explain this fall in relative well-being, including worsening food insecurity and interruptions to education. And so just to sum up, in the paper, we find that in the pandemic year, the previous gains of the younger cohort in well-being has largely disappeared in all countries except Vietnam, the country that's been most successful at limiting the spread of the virus. Furthermore, losses in educational enrollment and a shift towards more discriminatory gender roles have been seen in all four young lives study countries. Thank you very much for listening. Thank you for this presentation. It was super interesting. It's very related with our keynote presentation yesterday from Mariana Bandiera and the big poll idea. So let's move on to the next one. The next one is by Munna Shefa. Virtual greetings to everyone and thanks for organizing for the opportunity to present our paper. My name is Munna Shefa. I'm a senior researcher from Sar Dhruv. The topic of my presentation is special in the process of the Prism of the Pandemic, COVID-19 in South Africa. This is a joint work by me and David from AFT and Professor Marie LeBron from Sar Dhruv. So the existing and growing number of research indicated that the adverse health impact of the current pandemic or other previous pandemics is not the same across population groups. In particular, there is evidence to suggest that poor people and people who is preexisting health conditions are disproportionately affected by such pandemics. What makes the current pandemic different from previous ones is that the measure introduced to contain the virus. Many countries implemented lockdown policies of different degrees and these policies are implemented both in developed and developing countries alike. As a result of this, in addition to analyzing the direct health impact of the pandemic, there are also research which try to look at what extent household living conditions allow them individuals to adhere to strict lockdown policies and to what extent household living conditions also allow them to follow WHO recommendation in order for them to prevent themselves from being infected by the virus. So the existing research indicated that individuals in poor countries have less capacity to deal with strict lockdown policies and also they have less capacity to follow WHO recommendations. These recommendations include, for example, the ability to wash hands with water and soap regularly, the ability to implement social distancing and the ability to get correct information about the pandemic. Our research is in line with this kind of research but we provide more detailed analysis, spatial analysis using South Africa as a case study. South Africa is an interesting case study in this regard because South Africa is one of the countries with high level of national equality and high level of spatial inequality. South Africa is also one of the countries which are highly affected by the current pandemic compared to other African countries. So in terms of data, we use the 2016 community survey. This is the latest data set available for us to estimate living conditions at lower geographic levels such as municipalities. Regarding vulnerability indicators, we follow Gordon et al's approach to select indicators. They provide a detailed justification why they have recommended these indicators. For example, sharing water sources and sharing a toilet with other households increases the likelihood of being affected from the neighborhood, person from the neighborhood. Lack of access to TV or radio limits access to information. Living in large households this makes it difficult to implement social distancing within the household if there is one person infected by the virus, for example. Older people have a high risk of dying by the virus. All the households can be adversely affected by the virus. Lack of fridge in the household might increase the likelihood of going to shops to buy food items more frequently. This might also increase the risk of being infected. So we use these six indicators and we map each indicator at different geographic levels and show the inequalities across different spatial units. We also try to calculate average vulnerability score using the six indicators. We weight them equally and calculate a vulnerability score for different aggregation levels. We also use a counting approach. We count the number of vulnerability indicators for each individual and aggregate that into spatial units, different spatial units. So the first result is showed that the average vulnerability score at the province level at the municipality level. So as we can see that there is inequality in terms of average vulnerability at the province level with provinces such as Eastern Cape, Quasulunatar, Fumulanga, and Northwest provinces have a high level of average vulnerability. Also within provinces we can see also significant inequalities or spatial inequalities at the municipality level. So we can see here in Eastern Cape there are areas which can be considered highly vulnerable while areas with less vulnerability. So this reflects also that areas with high level of vulnerability also are areas which are characterized by high level of poverty and deprivation. So in order to test this we try to analyze the relation between households, wells, and average vulnerability. So as you can see that in this graph we have on the exercises quantiles of household wealth and on the y-axis we have vulnerability index. So we can see clearly there is negatively range between the level of vulnerability and well the quantile. Individuals or households in the first quartile have a much higher vulnerability, average vulnerability compared to individuals at the lower, at the higher quantile, quantile five. We do the same kind of analysis at municipality level. So we calculate average wealth index at municipality level and we also calculate average vulnerability index at municipality and run simple regression and we find significant relationship between significant and negative relationship between the two majors. So municipalities with low level of average wealth have high level of vulnerability by far some. So in summary our analysis show that there is significant inequality, special inequality in terms of COVID vulnerability in South Africa because of households living condition. So this also indicate that there is positive relationship between household wealth status or income status and also vulnerability COVID-19. So the pre-existing socio-economic inequalities translate into vulnerability to COVID-19. So poor people have high vulnerability to COVID-19 because of their living conditions and they are also more likely to be export to the virus because of their working conditions. They are more likely to use public transport. They are more likely to to work in areas which requires contact with other people. So because of this, this has implication in terms of exacerbating existing health inequalities because poor people also have high number of risk factors under like risk factors for pandemic such as COVID-19. Thank you. Thank you, that was great. Okay, so I think we can move on to the last presentation by Vincenzo. Good morning everyone. My name is Vincenzo Salvucci and today I will present this paper on the impact of COVID-19 on consumption poverty in Mozambique. This is a paper that was written with Giulia Barletta, Fino Ricastigo, Eva Maria Egger, Michael Keller and Fintar. In this paper we do something relatively simple. We take some detailed estimates at micro-economic level obtained in another paper, translate them into impacts on consumption and poverty. And the way we do that is we take four sets of impact estimates taken from that paper by Beto et al. The impacts that we take are on wages, on sectoral GDP, on household income and unemployment. We combine these impacts and translate them into impacts on consumption and on poverty. We apply some consumption income, consumption wage elasticity and play with impacts on employment and in this way we get simulated consumption estimates and poverty rates. Our results turn out to be very close to those obtained by the World Bank in their Mozambique Economic Update from February 2021. This paper was originally designed as a policy paper. It was repressed by the Ministry of Economics and Finance of Mozambique. So it's not so advanced in terms of research methods but hopefully we hope it will be useful for policy purposes especially. Now we were lucky enough to have detailed macroeconomic estimates that were already available. They were obtained using a social accounting metrics approach and we heavily rely on these estimates to compute the impact on COVID-19 on consumption and on poverty. Now Mozambique is a country that after 1992 achieved fast growth rapid and strong poverty reduction from 70% at the end of the 90s to 46% in 2014-15 it still presents a strong rural urban divide and a regional divide. Now since 2015 several crises hit the country. And finally in 2020 and 2021 there was this health and economic crisis due to COVID-19. The data that we use are from the 2014-15 household budget survey that we call IOF 14. Now as introduced we assume that two main impact channels are at work a direct impact on income and wage and we use the estimated macro impacts on wage on sectoral GDP and on household income and two the second channel is the channel of employment losses. And for this we use the estimated aggregate impact on employment. The two are then combined to assess the final effect on consumption and poverty. The results are containing these table and table in the next slide. We have three different approaches and the average that we get is that consumption reduction was about 10% national but also at urban and rural level. Whereas the poverty rate increase is estimated to be about seven percentage points at national level slightly more at rural level and slightly less at urban level. This translates into about 2 million people at national level entering poverty in less than one year. About 1.5 million in rural areas and about half a million in urban areas. Now, even though the effect on consumption is similar for different education levels we see that the effect on poverty which is the blue bars is much higher for uneducated people. And also we observe some differences for what concerns the impact on consumption and on poverty for people working in different sectors working as traders, service, domestic workers, peasants, agriculture work. So concluding we see that consumption may have decreased by between 7 and 14%. Poverty may have increased by four 10 percentage points from baseline of 46% depending on the approach. And this corresponds to about 2 million people entering poverty in less than one year. Now, the poverty results reflect the higher probability of falling into poverty for households in rural areas, people in subsistence agriculture, individuals with low educational attainment, family workers, domestic workers. Whereas consumption decreased relatively homogeneously across provinces. More for small traders. And also more for people working in agriculture, mining, manufacturing, construction, utilities and transportation. We also find that inequality increased but only modestly. Now, COVID-19 certainly has produced a setback for poverty reduction in Mozambique. And many more households fell into poverty or experienced dropping consumption including household categories generally less vulnerable. Now the longer term structural drivers of poverty seem to be still at work but new drivers likely emerged. So for chronic poverty interventions they should address the structural drivers of poverty and they are still important. At the same time it is key to address these new drivers that were identified. Thank you. Thank you. This was really, really great. I think it really highlights how the effects of the COVID pandemic is going to, they are going to outlive the pandemic itself. So let's move on to the Q&A part of this session. We have a question for Jemima on the Q&A by Richard. Jemima, Richard is asking if you have to come up with policy examples in other countries that take great conditions into account. Honestly, I have, sorry, that wasn't equal. Yeah, I haven't. And at the time that I responded, so this paper was part of a symposium by the general of agrarian change titled COVID-19 and the conditions and struggles of agrarian classes of labor. So at the time that I was writing the paper for the symposium, I did not come across any. This was in early 2020. And when I was revising the paper around late 2020, there still wasn't much. There are some academic papers, but I have not come across any COVID-19 policy examples like that intentionally considered the needs of agrarian populations. Sorry, Cathy, do we have the other speaker already? So we have her online, but she needs to request to share the audio herself. And I don't see the request, so I'm not sure. Okay. Okay, in that case, let's continue with the Q&A for now. Yeah. Jemima, so I think you mentioned that there was an increase in smuggling and I think it was smuggling in something else. Yeah, smuggling and migrant activities. So it's a fairly small border compared to other borders. And so people are able to cross relatively easily. But given that there was a pandemic and there were stringent measures to control movement, you would think that some of those measures would apply to that border as well, but there was nothing. And so there was a news item about some smugglers had been arrested at that border. There was another news item about, I think, someone from Burkina Faso had crossed into the Upper West Region and they suspected that the person had COVID-19. And so, and among others, we just went to show that there was very little attention being paid to that border. See, so it was basically a diversion from the main official border to the. So Richard has another question for Mune this time. Did you look at indicators breakdowns within the vulnerability index? And if so, was there one or two that people were most commonly deprived of in Mune? Are you there? Yes. So yeah, we looked at, we also mapped each indicators at various special units. So the most depends also from whether you are looking at urban areas, rural areas about indicators. For example, in some of the urban areas, the indicators which are more important as sharing water, for example, is the most people are most deprived in terms of that indicator because most people share water, especially in rural areas. And the other one is also sharing toilet. Most people also in rural areas. So it depends where, which area you're looking at in terms of which indicator particularly is important in that particular area. So we provided every indicator and we mapped it at the province, municipality, the rural areas in the paper. Thanks. So I was wondering if you are planning or going to look at the special correlation between the vulnerability and actually COVID infections? And if you see something like that. I think that would be very interesting. Sorry. So I think that would be a very interesting future work given that there's data availability. We were trying to do though, so since we have this paper, but there is limitation in terms of data sets like we getting COVID cases, tests, because we have to control so forth test patterns because test patterns also vary across the space. So we're trying to get data sets, but so far we are not succeeded in terms of getting detailed cases at municipality level or then province level is not available. But if it's available in the future, we would like to do and push the paper. Thanks. Sounds good. Richard, I actually have a question for you. I was wondering how much of the differences across completely that you see can be explained by the COVID measures taken by the government or even the number of infections and how that is different from just the differences in the initial physical capital. Yeah, I was wondering if you could talk a bit more about that. Yeah, sure. I mean, that's a great question and something that I don't think we can fully ascertain the relative contributions of COVID specifically versus any kind of general initial deprivations or socioeconomic progress. Yeah, the data we have just doesn't allow us to do that kind of decomposition. But what I can say is that I didn't present here, but some of the other results we look at in the difference in difference framework are we formally look at educational enrolments. We look at the proportion of job loss or loss of income. And we also look at subjective wealth, kind of a liquid scale between one and six of your relative ranking of your wealth. And we find consistently for all of our outcomes that even in cases where all four countries are affected, unlike subjective well-being where Vietnam wasn't actually affected, the magnitude of the deprivation is largest in Peru and smallest in Vietnam by quite a long way of quite large magnitudes. And so, you know, if it was basing of kind of existing levels of deprivation, we wouldn't expect to find Peru being very heavily affected as they're a low middle income country saying comparisons to India and Ethiopia. But we consistently find this pattern that Peru, which has been worst affected in terms of the COVID cases and deaths per capita, seems to be consistently the worst affected in terms of the magnitude of the shocks. Also, just briefly, analyzing, we have a few indicators that we have recall periods on. So we asked, you know, what was the highest educational grade you achieved just before COVID-19? And what was your subjective wealth just before the pandemic broke out? And we consistently find that the countries that were doing well, such like Peru and Vietnam and even Ethiopia with its high economic growth, had continued to do well right up until the pandemic. But then they see these large declines during the pandemic itself. Okay. Thank you. Any more questions from the audience? Obviously. Otherwise, I have one for a bit. So which is, I thought it was really puzzling how the consumption, the reduction in consumption was so similar between rural and urban areas. Yeah. I was expecting that to be a much larger drop in urban than in rural, because I mean, usually the sort of measures are targeted more to urban areas and rural ones are a bit less affected. Like life goes on more or less as usual and people continue to go to their fields and plant their crops and whatnot. So I was thinking, can you expand a bit on that perhaps? Yeah. Actually, I was expecting, I was expecting the same and indeed when we use, well, there are also differences in the various approaches. For example, when we take sectors and we use detailed sectoral estimates, etc., we find that, for example, in urban areas, we have a bigger effect than in rural areas, but then on average, yeah, indeed we find a 10% decrease at rural and urban level. And, well, we think that, okay, even the urban areas were more affected than rural areas, especially in the beginning of the pandemic. Then the pandemics spread out also to other areas, even though urban areas are still more affected than rural ones, then, well, we think that, okay, it really needs as more stock in urban areas also to affect rural areas. That depends a lot also on the fact that the existence of markets for their products, the existence of, well, the possibility to have transportation or to get, for example, inputs and also food becoming available from neighboring countries, South Africa, etc. So even a small shock for rural areas can translate into a drop in consumption of a similar magnitude. Now, let's also take into account that the consumption levels in rural areas are much lower than in urban areas. So, for example, if average consumption in rural areas is 5 mtk per day, for example, a drop of 10% in that case is half a mtk, whereas, for example, in urban areas is about twice as much. And so the absolute reduction in consumption for urban areas was definitely higher. But, yeah, it is surprising for us as well. And that's why we are expecting now to have actual data, the actual data for the household budget survey that was done in 2019-20. And so have a look at actual household data if this is the case or if our simulations are not very correct. And so we have to revise the model. One possible explanation that I was just thinking about is also has to do with remittances and the drop-off remittances from the urban to the rural areas. So basically just the repercussion of this huge drop in urban that spills over to the rural ones. Absolutely. And you're definitely right. That can be a channel. And even though mobile phone transactions were increased in a way, so people from urban areas could send easily money to people in rural areas. Well, since they got a bigger shock, then the amount that was transferred to rural areas was definitely smaller. Perfect. Thank you. And I hope to continue to see you in the other sessions in the conference. And we can continue to talk. Thank you. Bye. Bye. Thank you.