Department of Agricultural, Food, and Resource Economics Innovation Lab for Food Security Policy, Research, Capacity and Influence

Good morning and good afternoon, everyone. I'm Dave surely, professor and egg Food and Resource Economics at Michigan State University and Director of PRC. I very, very pleased to welcome you here to on the set of 12 minute lightning talks from our new stars plus cohort. So will be welcoming Team from the Philippine Competition Commission. This is our second team from Asia, our first team from the Philippines. So very happy to have you in the program. Also whopping, welcoming a team from EPR, see the Economic Policy Research Center in Uganda. A PRC is a center for policy leadership supported by PRC I it, it's their second team under the stars plus program. So congratulations to you and welcome to this. We have a team from peel off from the University of Ibadan. Peel off is a Policy Research Center focused on food policy that was formed with the assistance of PRC ion based on the vision of everyone at the University of Ibadan, we're very pleased and proud that you all were able to compete in and enter into this program as well. And finally, a team from Esser at the University of Ghana. A SAR has been a key member of PRC I, a key new member of, of, of brain operate. So we're very pleased to see that you all have a team in this program as well. So high, I want to thank Chris Barrett, professor Barrett a Cornell for his fabulous leadership starting of stars, beginning of stars and then in stars plus as well. And and Kelsey as well, who you all know extremely well. She interacts with you constantly. And that really keeps this program on, on the rails. So we're looking forward to your presentations. And I think with that, I'll just I'll just turn it back to Kelsey and we'll get we'll get roll and listening to to what you have to tell us today. Thank you. Great. Thanks, Dave. I am not going to be Webber anything, but I'm just going to let the presenters know. I'm, we'll have you share your own screens. So DJ, whenever you are ready to share your spring, you can go ahead and you'll have control over your sides if we run into issues. I have a backup copy. So we always have that as a, as an option. But we'll start off with Wuji with each of you. I'm presenting your own slides. Um, but yeah, we have we have J. Tarantino here from the Philippines Competition Commission team presenting on behalf of their team. So I will pass it over to you, Jay. Thank you, Kelsey. I'm sharing my screen correctly. It would be great if you can put it in presenter mode. Okay. There. Is that in presenter mode already, sir? Okay. Okay. So good morning, everyone. I'm Pardis, JPL and P. Naught from the Philippines. I'm representing our BCC beam, the present research paper on market power in the, in the Philippine agricultural markets. So in terms of what the vision is, the competition authority, the Philippines, we monitor and analyze the practice of competition in markets that affect the economy. So there's particular interest in understanding the market structure and competition in agricultural markets. So for example, among others. The BCC has been looking at how farmers are affected by market power of buyers if they produce the rice traders in terms of pricing. So we're interested in that, among others of the things we're looking at in the agricultural markets. So the goal of our research, likely to update the Philippine market power literature in agricultural markets. To detect and measure market power using methods that have not yet been applied. In the Philippine context. The current set of PHP person in literature or market power in the Philippine agree markets. There are relatively few and generally can be classified into price tests or Structure-Conduct-Performance style papers. And these papers indirectly show existence of market power via price tests, which rely mostly on symmetric breath adjustments or lack of price integrations. For example, in rice markets across supply chain, the test or on the price asymmetry in Farm Gate pass-through price and the whole theory, the levels, adjustments using AIDL or for differences. Million Duffy 9092 example, the Intel CPU, Yes, yes, In 2012 are using for different cell type models are either 2000 Wolfram hook and error correction methods, baby garden 2011 example in horizontal integration or across geographic markets. There has been price tests between provinces, orange, a country in 1994 be silver boolean, Syria. And there's also been tests to if there's integration between national prizes and international price of rice. Yao ship masters in 2007, also using EIDL first differences and more recently released in 2018, tested the horizontal market integration. We cinco integration tests and finding a relationship as evidence of no market power. On the other set of literature on the Structure-Conduct-Performance framework. Due to the constraints, these papers are relatively rare, but the mostly followed them. Mythology of computing market shares and concentration in the system such as the Herfindahl index or concentration ratios. And examples would be the supermarkets began in 2015, or a more recent paper, one more recent paper in a PCC from medulla came better. So you on refactoring sleeper. So we use this in using the Herfindahl index S in and in our analysis whenever possible. However, these concentration measures are not always reliable in measuring market power. So for example, increases in market share, the Jew to decreasing costs are in movement. Inefficiency would not necessarily be related to increase in market power. Other papers on panel data econometrics file with structural interpretation, where price or price cost margins are directly measured through cost of goods sold from publicly available income statements. Well, dab in 2008. Or these are sometimes called the accounting approach. No marginal cost are estimated. But the economic assumption is the average cost will be equal to marginal costs. However, conduct parameters are not directly estimated. But an example would be an exception of branding in 2009, power generation. Another paper or publication, there were attempts to directly measure profits alone, supply chain, and actual geographic lows of rice using a survey and key informant interviews by the Philippine Rice Research Institute in 2018. Okay, So since our goal in this research to measure market power, which we will be proposed to use this via production function method to estimate market level, market power, its magnitude and existence. We can use the production function in cost minimization as Hall in 2018 or 1988. With its generalized end, hall, generalize the decomposition of total factor productivity to include a markup term. And in this equation, the mark up there mu is mu and a technological progress is presented by delta z. And hold use instrumental variables to instrument for technical, technological progress to consistently estimate markup. Our Roadrunner and 295 extended this to get the dual relationship to get the difference between a duel in primus solar residual, which is also a function of the markup. And most requires the assumption of constant returns to scale. But subsequent work allows introduction of the scale them. So we recently gained access to survey data, our census date of establishments to, to conduct this research to estimate the different function. So this establishment survey, we have variables pertaining to employment, income, expenditure, capital data. So for agricultural data, you have ten years of survey data on establishments. And in the most recent survey in 2018, we have 1620 observations. So just to discuss a few of the general variables, we'll be using decimation, employment to the employment is a 100 percent available in the dataset can be broken down by men and women are paid and unpaid employment. But as you go as a CDS aggregate by that, the sum of the observations would be missing. So we also have data on income, total revenues available. And we just indicating here the number of missing observations in the inside of parentheses. So total revenue is derived from revenue from sale of products and sales to domestic market. We also have data on costs, not able to indicate in the slide the town labor expenses until the expenditure on labor, part of cost of goods sold, capital, stock and expenses. But a lot of these variables have missing observations. Lot of missing observations. So you have depreciation expense is the capital stock book value and which can be split into land machines could debated assets. Okay. In terms of materials expand, we also have data on with your expenses, okay? So using this data preliminarily, we're going to estimate the version of Hall, which would extend the Solow residual framework. And this will be estimated three digit industry classification. To estimate this equation. So in the setup, we have inputs, capital and labor, but can be expanded to include capital types and materials input. So here, where mu is greater than one is the MSM is a markup, and v equals one is constant returns to scale. So both can be estimated. And when u equals to one. And if this perfect competition and in constant returns to scale, the measure of total factor productivity is theta will be equal to the left-hand side. And one key problem were key indigeneity problem is the correlation between it and productivity. So we will use OLS and the gym and gentlemen, the moments IV to deal with indigeneity. So, so that's our research. We've recently looked into the data and we're crazy and progress to do the estimations. Thank you. Thank you very much. Thank you very much, Jay. So now we're just going to open it up for comments or questions for the PECC team from the group, feel free to raise your hand or unmute, or you can add a question or comment into the chat or the group. Chris, go ahead. Yeah, really nice presentation day. And the more I hear about the project, more excited that I get by, by what you and your colleague you doing. I was struck by how many missing observations your reporting. And I'm wondering whether you've looked into how many of those are truly missing versus they appear to be 0 that just weren't recorded as such. But there were lots of small businesses, for example, that don't have land and people who were are just selling their services there an accountant. They work from their home, they go to other people to take care of their account. Though, Pleiku commonly won't record any land at that. And I wonder whether, because I've seen that in other data that doesn't get recorded. But it's actually true, Nero, that would matter a bit for how you do your analysis with data cleaning. Have you have you considered have you look through the data carefully enough to have it? How many of those are truly missing versus they're just that they're a valued observation. So personally, I haven't seen the data yet, just Gabby and dash of seeing data. But we understand that how the data is the route, the variables into which variables without complete information and wrap them to two which has the most observations. I understand it the missing data is actually missing in the spreadsheet. But I think you're correct. I think we have to consider whether these are truly missing data or maybe there are zeros there have been recorded, will definitely check with our Philippines. It's called authority. With that observation. Yes, Chris will be checking with them if the missing data is in B, I'm missing. That's one of our questions. And thank you for the comment and the pattern things transgene. You also have a question in the chat. He wanted to go ahead and get a quick response to that. Gabby issuing this question to answer this, Abby. Yes, From my understanding, using we can use lagged values of our independent variables on the start as potential candidates for instruments. For my reading of the literature, that's a feast. Women start when we start with a GMM IV estimation. But then love to hear some of my ideas as well. Just for those who weren't able to see the chat. The question was about which possible IV's the team will consider and for what variables. Any other questions or comments? Advice for the team from the Philippines. I will also note and give thanks to their mentors. So we have two mentors per team for each of our stars plus teams. So the PCC team has mentored by RNa Basu in Mali in gram, and I believe Molly is here today, so thanks, Molly. All right. We have a question from Maggie. Thanks case. The action is just a small comment. If you can go to the slide with showing the Guo. Sorry, which slide? Again? Maggie, Go where you were presenting the goal of your honesty is okay. Okay. Hey, there in that slide. Yes, I was just thinking at the end of the goal you are saying you are going to update the three prime agricultural economics literature to detect the major market power using methods that have not been uprights. Out. I was thinking maybe from your presentation you're saying I'm going to use the production function. So as I was thinking, maybe it's better just to say used in the proposal, the production function. Thank you. Sorry, could you repeat that? I didn't quite get the question. Is just a moment. I was seven instead of the end of the synthase saying using methods that I've not yet been upright in the and context. You could just say using the products AND function because that's the method that you're going to use to actually update the market paella wherever in your study instead of remove, right? Yes. Thank you. Okay. Thank you very much. I guess the context here is that we were hoping to do some more or some other methods later on after we've accomplish this particular one. In the original proposal, there were these two things if we wanted to do, and that's why this sounds are reads like it's more open-ended than it is, but correct in this particular context, we should've been more specific in that. That's what we wanted to do, the production function approach. Thank you very much. We have time for at least one more question. If there's anything lingering in a group, I will, I'll so let's say we have Vmax and dabble in if you are ready with your signs, I would say get prepared because we will be coming to you shortly. Okay. Well, I think that was a great start to the to the session. Thank you so much, Jay, for a great presentation. If you have questions or comments as the group and you want to directly contact the team, feel free to go ahead and reach out to them. But with that, I think we will move it to everyone. You. Thank you everyone. I think we'll move on to our EPSRC teams. So dabble in if you want to go ahead and share your screen and your slides so we can go make sure we see everything. And I think you'll also want to go into presenter mode. Are present. All right, So I guess that we have Dublin here with the EPR see team that is going to give their presentation. I will just go ahead and shout out their mentors at the side of the presentation this time. So they are mentored by Bouton and where we've been pKa Dublin over to you. Thank you, Kelsey. Hello everyone. My name is. I'm from the Economic Policy Research Center. It presenting guy, the wonderful team of my query such as in the test, I'll be presenting our study titled linking crop productivity, market performance and technology use among farmers in Uganda. And this is my presentation outline. Sorry about that. In terms of background. And we tested a big deal in Uganda, looking at the number out. So it's that derive their house, their livelihood from agriculture. Many sick days don't without chatting, is there a couple of sharing is added to have I seen the sector. The importance of agriculture has been reaffirmed by carrying COVID-19 pandemic because it came to became a fallback position whereby different businesses that were that twist shaped k-means in urban areas and descent as modulatory Ugandans been back to the villages, to rural areas to engage in agriculture. So we thought that it's prudent that aim. You know, that for policymakers and researchers to look into issues that can prohibit Ugandans and most especially rural, realized from poverty. The challenges that time are facing in agriculture which are within the sector, should be addressed during. So here we are looking at a two-pronged agricultural programming problem. The low productivity, low market growth, market participation, and performance. Looking at table 1 and we are looking at the disparity between the average yields that is prevailing if I'm vamps and amaz. And the potential that would have been achieved. A single out maze yields a look at 2.3 times by a tear. In terms of what diplomats producing. And the potential is up to eight tons. Exhale. And we see that this displays a potential to increase this to around four times. So this shows the, the picture and Maxwell out slow pro, crop productivity is concerned. In terms of the local market participation and performance. We see that you're going to seal has a high, big subsistence economy where most households produce four on consumption. And of the 3.5 million households in the subsistence economy, 62 I steered in the subsistence agriculture. So to sing out, when cash crop and one food group looking at it, we see that production is mainly driven by the increase in area pasted don't necessarily increase in productivity. So looking at this blue line, which is area vested in cats, there has been increasing steadily, that's for coffee. And the production has also been increasing. 2005 it was around 15 to 8.18.1 tons, up to around 300.600 thousand IN 1000 kilograms, which is 3.312. Going to extend. What we get from here is that these orange paths, I guess this the show they read in terms of kilogram per acre and it is declining or Fe2, this production that we see at is increasing, is mainly driven by area arrested. Now the picture is relatively similar but slightly different from that of coffee. We see that the idea of a state is also increasing here. And then production has been increasing. But in some areas, productivity, represented by the orange buzz is stagnating. So what do we seek to answer? Number 1, we seek to find out what could be the effect of technology use on our productivity in Uganda and taking lords they produce and market performance. But most importantly, the mechanism. How does technology sort of the two-pronged problem that we have identified? Problem of declining productivity and low chromatid performance. So in terms of R0 to each other. And there is remarkably clear evidence that shows that Pacific in crop yields and technology use. And several studies have also documented compliment type of technology or use that different technologies shouldn't be used in isolation. One example is that most improved seeds are obliged to walk, will flourish and grow Beta when they're supplementing these inorganic fertilizers. It has been documented, but we see that recent evidence shows that this is not the norm and mostly if yes, consider them as substitutes. So you also intend to investigate this. There's also evidence that has been documented as far as i is strong and positive relationship between our crop productivity and market performance in terms of the intensity of sales and the new and emerging technologies, for example, smartphones, I tend to diversify market options. So in terms of the energy gap, we see that these different studies look at productivity, market performance and technology in Piazza separately. And you think this badly solve the problem. So what we intend to do, HEDIS to look at a holistic approach. You're looking at all the productivity, performance any technology and we sell. You know, you can mitigate or soluble this two-pronged problem, prevailing agriculture. Now, terms of conceptualization, we look at a typical household that ALU produces foreign consumption, but also has pair production as some production that is going to be sold. So the household is supposed to produce phone consumption, but also participate in the market. It places the different challenges that they are in the sector. And dream may mean propose position is that a household that adopts technology, course, experience enhanced, repealed. But also it will perform better well in the market through sales and all that. Of course, other factors that can mean being a member Farmers Association or the struct as three, re-feeding he and affect the outcomes of increase crop yields or stagnant will reduce crop yields. So in terms of the estimation strategy, at the start, we had thought that we are going to disaggregate a look at groups that we had a cookie and maize, looking at the conventional way of measuring productivity as in your program. But then when we looked at the context of Uganda, It's very common for a Ugandan promised, if I'm at, will have within the same protein they have coffee and then the plant made it so you have black flip-flop. So it can be quite problematic to add tides are saying inputs to a space probe. So now we add a region measures of total factor productivity as put forward in this recent study by a lagoon. So we shall taboo, we juggle these two and see how it pans out. And in getting that, take the data points as far as what we have seen. So find the data. They single crops. Be forthcoming. So in terms of market performance, we are following a framework by being FECA, we will shall we do a crop sales index and vegetable crop market performance. Once you telling that participation performances when edition on participation, if a family or a household is faced with a discrete choice, so participating on participating. So this introduces the sample selection bias. So here we are looking at the Heckman two-step technique. We shall do in the last stage, analyze the district, the discrete choice or participating on participating in stage 2, where we will look at the continuous outcome. So the data we're using, it is for web. So you can imagine a pannus habitats 13, 14, 15, 16, and 18, 19 plus they listened to and which is a modulated with a 920. And the pia, we single out just the most commonly grown crops, that's means coffee and bananas. And we see that, or otherwise 70 percent of households and at least see these crops contribute to this. Most households grow these crops, but they are similar that are grown. In contrast, you see here in this kind of yes, was there so many? In terms of preliminary findings, what we get here quick one is that I like I'll say whatever we have seen so far in our delving into the data. There is low usage of technology, the different technologies that passenger you, for example, if you have any fertilizers, pesticides, and improve seed as well, very slow market performance. But what we derived, that is, we look at now complimentary it as we have seen in the literature. What's the picture? And we ask ourselves, do Fama's view these technologies as substitutes or complements? So we see that usage of fertilizers, that around 6.9% of households using vinegar, an organic fertilizers, that is very long. Then I, what's the percentage of households that use organic and inorganic fertilizers. Now the drops drastically. This is the same picture for improved seeds, which is like 77.9. But then when you combine improved seed with inorganic fertilizers, it drops to 0.5. Now this is contrary to what agronomist recommend, that the fact that some, most, if not all improve the seeds blade will ablate to work well for data when they are supplemented with a good inorganic fertilizers, It's, it's learned what is happening in the country descend into chaos also with the parallelism and pesticides. So there's a possibility that they compliment ISIL. Technology is not happening. This is for further investigation. Now, here we'll look at briefly, the decision-makers. Does the complimentary of technology use its kings on the 6th by the sex of the decision-maker. Yeah, we'll use household heads for example. We see in general, if you look at the four columns, this and is whether female or male 8 user Yup technologies. It is high in town in male hits compared to female hits. Now, dish, which has shown that females or women take care of their families. Beta compared to men, now would be expected him to know or maybe use a higher complimentary to technology use compared to mean. But we see that throughout, irrespective of the decision-maker, complimentarity still low, as you see 1.101, but still it's relatively better when you compare a male to female heads. Heads, as in terms of decision-makers. So this is what we have. So if I times or bad, I descriptive analyses. Thank you. Thanks so much. Dabble in. So you already have a couple of questions in chat that popped up during the presentation. So I guess we can start with those. And if anyone else has questions, feel free to raise your hand and we will get to you. Martin asked, this was all the way up. I want to say around slide for more and also feel free to unmute if you're able and if you'd like to just chime in with your thoughts. Yeah. It's Martin Fowler here from USA, Kampala, Uganda. Just to say, I see figures like this being used a lot and I'm always worried when I see FAO stats, I have nothing against FAO and statistics. I've been at the other end working in planning units watching how this data is, is prepared each and sent to FAO. So all I can say is I can give you some examples which would make your hair stand on end. But really, I wouldn't use this information. I would use the UNH SUN ps data that you've used later to show any trends in yields over time. But I wouldn't use. On this information that at all, most of it is guesswork done by how that plan is in ministries of agriculture, not just in Uganda. I also witnessed the beginning of my career in the suit to wear the permanent secretary when I asked him what I should do with the latest FAO data request for data, he said add 3%, which I did. And those figures are now cast in stone back in Rome for 970, seven, 78, and 79 data for crop yields in, INLA C2. So be very wary about using this data. I will look at the sources properly. Sure. Thank you mentally. Thank you for that. I'm going to be looked into. Thank you. And then David had a question about slide on slide 10. Just to clarify what your measure of market performance is. Good. Go ahead. Yes, they are where it came out Bureau referring to market performance there and it wasn't the went kind of fast and it wasn't clear to me how you're measuring market performance. Shahid, if I can take I might take that quick. Oh, yeah, Absolutely. This is your time to answer questions. I'll okay. For now. What we do as the major megabytes fish. And of course, we're discussing this yesterday with our minutes. We have, they, they showed us another major market performance where we're going to do bite production and they end up to the production and how much is sold. For now, we're using a direct question. I'll of course, with discretion where we see the percentage of the harvest that he sold. But then that can come with some problems and issues. But we can explore different modes of participation. Because here we have the labor market participation and we have the passengers sold, then the performance is the value of cells. Yeah, So if for example, if you have maze and that's the liver, but then the question, the discreet question of whether you participating on notice what gives S phase. So we're just looking at the percentage of harvest in this case for descriptives. The facility obviously have is the likelihood. Okay, So likelihood of participation and then amount of sales, value of sales conditional on participation. Okay, that's, that's at least clear now. Thanks. And let's turn it over to Chris down with his hand up. Thank you, Kristen. Yes. Thank you very much for this opportunity. I wasn't clear on how they're going to make up for the TGA on into products. So I mentioned of using the total factor productivity, but I wasn't sure how that would be useful. Plot switch into a number of crops. I'm another question is, what types of my kids I'm considering in your analysis because as saying, one is low market but special. So I was just wondering whether you can see the NDB, that I can be the local farmers markets where maybe the majority then the editor will also be Lipa uses and therefore, hello My kids by expansion. But I'm not spelling that recitation. So maybe if you can find that. Whereas up into the Teams terrorist time. Yeah. I can take on the first 10, maybe another team member can come through. I guess they are all here. So thank you Christian, for those pertinent questions. The first question of the intercropping the mat cookie. Now I going to measure it. Like I said, at first we had, so we are going to of course, be specific and explicit to look at using the conventional way of measuring for acuity, that is yields kingdom, milgram producing less output. But then we realized that it's with the context of Uganda, It's really easy. And if that has within the same plot as planted coffee, as planted bananas, but then they also growing pins. So looking at the inputs we are having, it's, it becomes problematic to assign and put an input to a specific crop and then advise that, okay, now start looking into a productive images. Now here we are looking at from now the unit of observation analysis changes there. We shall look at diff-amp, look at the total agricultural Drew a picture from the output in that regard. But we're not abandoning the specific crops, for example, crop productivity, where we say, okay, what discovery productivities of coffee specifically about myself, this the output of coffee per acre that is produced. That is the conventional way. So we're just looking at how we can use different measures to avoid a shortcomings that and then connect the context of Uganda goes into Brookings. A big thing. Then in terms of markets, we have not experts, they looked at this specific markets. For now it's an aggregate where we looking at sales and the nave, the aim is for the household to participate in the market. So we anoint a, currently we don't entirely looking at the final destination or where is it going, but we can look at those and it's I'm correlations the idea. Yeah, it's a, it's a very good question. C, and this, my equity inequality has something to add. That's right. Thank you. Thank you. Thank you for that response is what you have to put Poitier that we are still trying different measures on how to come up with a productivity measure we haven't yet completed. And one good, I guess, preliminary findings. Then I see there's a question that is coming up again from Dr. Medina. But they useful when instrumental variable. And this one still has also fast looking at the post-war that we have from the data. But the first thing would be to look at the different lags of what is causing the endogeneity issue. But still, we are still just starting. Let's hit possess we are going to be at so we kind of give you a definite no. Thank you. Thanks, guys. So we have about two more minutes. Molly has a couple questions in the chat. Her first one in terms of assessing complementarities, how worried should I be that results are driven by financial constraints versus knowledge? To maybe take two minutes to answer that and then I can copy down the rest of your questions and you can follow up either in the chat outside or or via email. Yeah. Sure. Thank you. And you take that on I can go invest and any supplement refers to I know works. The first thing you may just see shortly. Yes. So the first question of whether they get differences are driven by groups. We just have to check for us. We cannot give a definite answer for now. We're still studying the data before we come up with a conclusive results. Then maybe you and a complimentarity. I am she said financial constraints and a v naught. I would say personally, both can be a possibility. So that's our task now to find out. Because now this is descriptive. You cannot claim any causal inferences. Yeah, we cannot remain equals inferences now. So we can maybe see in terms of mechanism, how that we can attribute nepa financial constraints or knowledge. But I think both are possibilities and confined. Problem is excess submitted. Yeah, that's what I can save money, but that's a very good question. Great, Perfect timing. If you have any other questions for the EPR, see Jim, excuse me. Please follow up with them either in the chat or via email. But big round of applause for the team on an another great presentation. Thank you. Thank you. Kill something. I'll stop sharing. Perfect. Alright, Benjamin, we're going to turn to you. So if you are ready to share your screen with your slides, you can go ahead and do that at any point. So now we are going to turn to the peel off team. And Benjamin is going to give their presentation this morning, this afternoon. They are mentored by Justin, George and Martina or jelly. And they have a great team of researchers working on the team. Benjamin, I am going to turn it over to you. Everything looks good on my end. Thank you. Kelsey. I want to say good morning, good afternoon, and good evening. To all of us. Use permit me to stand on the existing protocol. My name is Benjamin or Lucia wheel, I'm from Nigeria. And I will be presenting our proposal on resilience to COVID-19 and insecurity shocks. Evidence from Nigeria. Actual, Okay. As a matter of introduction, COVID-19 started December 2019. It came as a global shock, and soon it was declared to be a global health emergency by WHO in January 2020. And few weeks down the line. In March 11th, 2020, it was officially declared as a pandemic of a global scale. And in Nigeria, the health sector was worst heats, such that by March 14th, 2020, 2022, rather 254,945 confirmed cases. I've already with 3000, 142 deaths. And the, However, this figure me look relatively low countries. But this is reflective of the fact that in IgD idea is actually a low level of testing, taking a population of about 200 million Nigerians. As of now, around 4.5 million had been tested. So this could be the reason why the cases appeal to other countries. However, before the COVID-19 stroke in Nigeria, there has been a lingering challenge of insecurities. How these US actually common that different sheets, such as Foucault around binary tree from either clashes, militancy, sensation is agitations. And all of these are been affecting household food security status, as well as livelihoods, the Dasani or well-being. So in this situation, COVID-19 came. I'm there by the observation of a number of those on the slide. It hasn't been confirmed. That could be 19. As we're seeing, there are increased food insecurity. It increased unemployment, increase inflation, as well as different sheets of insecurity cases. So the green coenzyme in the mind of the statutes is our Nigerian households copy. And the outcome we described earlier will progressively on the boat shocks of COVID-19 and insecurity that I've been on ground before COVID-19 came. And resilience as a terminology, has been described as the ability of the household to prepare for, to cope with, and adapt to shots in such a way that their well-being is still preserved and they do not fall below the poverty line. So the motivates Chernobyl study comes from the fact that we wish to look at the combined effect of the shocks from COVID-19 and insecurity. So previous studies have looked at the effect of 19 and insecurity differently on food access, food security, Parvati, financial inclusion and coping strategies there. While studies also look at our schools, children in school still in to COVID-19 economic shock. However, we noticed that their expositor of study on the resilience of our schools to combined effect, to the combined effect of insecurity and community shops. Also, a few studies that have made use of the high-frequency COVID-19 Soviet we tend to use in the study of not actually deploy the total runs. Our main level rounds found in this dataset and this study hints to do that. Also, this study aims to look at the agenda or been segregation across our schools deserve this, the effect on the combined effect of insecurity and fulfillment. So we made such into literature. And they also fund some gaps for that confounder, horrible dead by the authors on this slide. So it actually, and discover that it's unitless. A team that COVID-19 deed caused higher food insecurity. Even though we know it up on demand measures across different countries. Presumably who'll bring about increased job losses, economic slowdown. And all of these can affect the poverty level of households, which in turn will affect their food security status. Get trich. So it's yet to be very clear on the port that Linux between COVID-19 and food insecurity. And this study seems wishes to field have that. Also we found out that there are limited research on COVID-19. Our COVID-19 affected food insecurity is the tuition of already vulnerable groups. The only study we found that did that was in Uganda, which looked at the effect of whom in 19 on give it food insecurity of refugees. So in Nigeria, we are looking at it, that this study will fill the gap. Sorry, that we were looking at vulnerable group, I mean, insecurity, insecurity region vulnerable group are now COVID-19 has affected them. And lastly, they are limited studies on the COVID-19 prevention policies as it affects different intrapulmonary regions. So many of the studies on effect of COVID-19 has been on a national scale. Solely study wishes to step it down. That we see the variations across different zones in the country and local governments in the country. And our COVID-19 prevention of policies has affected households alone could do. So. This gives rise to our research question. The first research question says, How was food security, which we want to actually capture as our outcome resilience indicator out Isaac been affected differently pre pandemic and post pandemic periods in conflict affected zones of Nigeria. To what is the prevalence level of a containment measures to COVID-19 implemented by various states in conflict affected zones of Nigeria. And the third question has to do with how as rest billions capacity of hours would fade in conflict affected zooms, pre and post COVID-19 pandemic periods. So the data methods, this still a work in progress. We are proposing to make use of LS MOSAiC data sets using the four waves under this dataset started In 2011 to 2019. And we will deliberately be isolating insecurity reading areas, which we will be using the accolade, does it actually gets handy. The applied acclaimed dataset is the pamphlet location and event data, project data. Okay. It is this georeferenced and a Nigerian data on this particular dataset is very robust and active. So we'll be using this to support the ALS MS, ISA data set to be able to capture the insecurity reading our portion of the country. And a, in addition to that, we were using the COVID-19 National Longitudinal Survey, which assay living rooms and was actually collected between 2019 and 2020. So in addition to that, and we hope we still plan to embark on some more detailed information about COVID-19 prevalence and containment measures deployed up to the local government level. In Nigeria. Actually, in the public domain, the available data is not beyond the state level, but we wish to step it down or by contact at the NCDs see office in Nigeria and also trying to gather information via different websites to be able to buy some data using the indicators presented like on Washington. Use of sanitizer, no handshake or physical routine, use of mass, use of gloves, and so on. All these indicators are such that they are already used in the eye frequency from Soviet data. And we're trying to actually use theory to gather the data. And in that way, we can expand the available data in the public domain for a deeper analysis in this study. So after we've been able to standardize this, we can still make it available in the public domain for the Arctic suggest you use. So one beauty of data blend in this study is the combination of ALS, MS, IC dataset, I played dataset will be solving additional detail. The acetyl group in 19, all about prevail and somebody may measure that we'll be exploring. All of these will be used to exploit a combined effect of COVID-19 on conflict shop. Among our schools in Nigeria, Sudan, design be using is, there was the experiment that is Zach. So our proposed empirical analysis include descriptive statistics. To use fixed effects. Regression with us will have to double difference or analytic approach. Thank you very much for your attention. Thank you so much Benjamin. I already see some activity in the chat, although I think Nathaniel and assisting and helping out answer some of these questions. Nathaniel, if you feel like there are things you want to speak to, to the, to the audience. Feel free to unmute and address the group. Otherwise, we might just move to some of the live questions up to you. Now that you've seen this question on the coffee type and that mutation on the present, our present that to say that they have different types of conflict, which hours of labor we add a random quantity. So we are trying to buy Nike grandma. It's possible. Benjamin, Can you mute your mic? I think we're just great. Great. Great. Okay. Okay, go ahead. Okay. So I will see that we also, we're trying to sample across the Tyrannosaurus characterized by conflict. So it's possible also to conduct the test of difference across different zones by resilience, capacity angles, outcome. So that I'm trying to see the second question that I responded to. Okay. Yes. Post pandemic. Was pandemic pre-pandemic. I said by post pandemic actually mean shortly obstacle was declared a pandemic. Yeah. I think that that question. Yeah. Bang. Great. Thanks. And then there's a new question from, from Nika in your plan to use the four waves of the Nigeria Alice to mess I say how do you intend to handle the case of sample refresh of the wave for data which makes it quite different from the samples in the preceding waves. That was his sample fresh. Look into that and if you have any advice or a knee, knee ideas as to how we may want to go to 10. You can just go ahead. But I'm just getting to know. Despite the guitar sex. Yeah. Although you sincerely thank you. A couple more minutes. Any other questions or comments or advice for the team? Okay. Medina. Thank you. Yeah. Okay. Thank you. Kelsey and the team. And for sure the topic really caught my attention because it's tropical diseases, the COVID-19. And it's about issues that arise really interested or say knowing within this leg sets of coordinates and insecurity. I think when you look at the topic of security to shops and cause them, I think when you went to the doctor presentation focuses on food insecurity at the end of the day. So why would it be more coordinated talk to the foods, food insecurity in thermodynamics as COVID-19 and food insecurity. Because I didn't the day you left it to y. And then we do not see that context-dependent within the preamble. The presentation, which I think we'd go to a little bit more on the issues around the health aspects of it. Within the introduction, that would be not tightening the gaps within the actual COVID-19 and security aspects directly. And given that shocks usually in the literature, shocks valid and they are very sharp so they idiosyncratic, understand. And I think the LMS be stemmed in that sense, that Ada some covariates or idiosyncratic. And then we'll be able to see that division in terms of which of these shops are driving a little bit more issues between your post or pre pandemic period of COVID-19. So then probably to be, to be more clearer in terms of what should we be capturing, indexing. So the shops that have beyond covariates, shops that they didn't have paper strength focus on where the idiosyncratic side of the shops more than the covariate shops. So there is that bit of that aspect that when you see into the presentation which is on now pillow thoughts, I said, Oh, where we're going with that, it'll help a little bit for bit. And then in essence, the ways that economic, I told the team time the aspects beyond the actual textiles that COVID-19 Kemeny tumbled economics or beats in distance. So I think that that aspect that I lost you kind of tightening datasets to a preamble to the Wahhabi. That's I think-I Edwards was the graph I saw meet Kelsey. Thank you for that comment. Okay. Well, if there are no other questions or comments, I will leave you to follow up with the presenters if you'd need. But we are going to move on to our final presentation. So if you are ready, you can go ahead and share your screen mode. And I will let you know if it looks good. So we have made it to our last presentation and this is by the University of Ghana Team. I'm Kwame is going to be our presenter for the morning. And this team has mentored by Kip Rome, by and justice Mensa. So it looks good. On my end. I will turn it over to you. All right. I like to test my microphone. Can you hear me? Sounds great. Okay. All right. Thank you very much. To everyone. Would absolutely muddy. Thanks for coming foot notation. Our topic is and rural transformation in Ghana. And we want to find out which constraints bank. Okay? Okay, so basically I'll introduce what we're trying to do and then emphasize a problem statement. Discuss their conceptual framework. And the hypothesis is that we have beat for this steady and then talk about there our cues and the objectives, that methodology we hope to adopt. And then I end with a brief review of the literature. So in many African countries, the agricultural sector isn't that just is in Ghana for instance, that in place I'm 61 plant focus anthropogenic baffles in rural areas. And these are areas that are characterized by limited opportunities for non-farm wage employment. January, I'm good. Yeah. I mean, therefore remains the main source of livelihood, particularly for these those areas. And it's kind of indicative of the nature of the countries for our transformation tragic tree. Now, similar to report et al 2017, reconceptualize transformation as a process that begins with increased agricultural productivity. Characterized by a gradual movement away from farm to non-farm activities. Increase that logical adoption scale economies and then shifts to higher value commodities. Is that even the most policymakers have an agreement on the need for transformation, there's insufficient understanding of their dry, there's some constraints of transformation, particularly in framing. That's an answer of transformation in Africa. Lost 2016 feel my et al in 2015 noted that the agriculture sector is critical because it has the potential to provide employment, sustained growth. You know, stuff like that. Despite this critical nature of that Greek societies still constrained by myriad of challenges that it faces. So we conceptualize that the traditional structure of rural communities is that it is her coach, our dominant, agreed domination of our communities. And so we stats transformation conceptualization from the point that their economies have this ad Greek dominate, abstract our economic activity. And then we hypothesize that over time, the dominance of outbreak is going to be keen. And we could probably see this in day. Decreasing I coach Alibaba shares communities of our time. Again, we hypothesize that this could be named by this weekend. Dominance of the sector could be explained by diversity, diversification of the economy, which is itself motivated by any factors including access to general purpose technologies such as electricity and Isaac Newton. So this study we seek to achieve the following objectives. The first one is I want to explore that Nietzsche in the Hudson of transformation in Ghana. A pattern of transformation been like over time. And then we would like to identify the nib less constrained to a transformation where special interests would focus on the rule of general-purpose technologies, such as electrification and ICT have reached on agriculture and the buses. So our objectives are linked to that. What is anytime personal transformation in Ghana saw in Namu limit? What is a set of general-purpose technology. And as I said, the focus here is on electrification and then ICT coverage or an agricultural nearby locations. All right, So I'll talk about the data that we are going to use to the fs are q, which is to describe the Nietzschean Python for our transformation in Ghana, we are going to use the Ghana live hashtag. That's a city. So this is entity repeated cross-sectional data sets. You're going to rely on waves three to seven weeks span more than two decades to try to paint the picture of what the raw transformation looks like in Ghana over that span of close to 30 years. And then with the second accurate, which is to examine the drivers and the constraints, inhibits in promoting transformation in the trajectory that we identified from the first research question, you are going to combine the GNSS data with administrative data on electrification and low bandage a coverage in Ghana. To arrive at the solution to that research question, of course, you are going to have 12 for a host of other factors that we'll find is on our literature search and so on. Okay? Alright. So talk about a key indicators of transformation is actually use our main variable of interest. So we are going to construct transformation. First as a share of households engage in agriculture on a set of individual workers employed in agriculture. We also looked at the ratio of how this worked in agriculture to the tata Alice wetland individual. And then we'll also consider the mean Sarah famine household members who work full-time in agriculture. And then the means of family household swaying be non-farm activities. And then lastly, we'll also like to look at household agriculture income versus household reach n. So what proportion of the total household income is resulting from income? What proportion of total household income is household rich endings? Onto a brief review of the literature. So Dan Coleman, 10, 11, for instance, found that electricity can act as a catalyst in reallocating labor away from nearby intensive activities. And indeed in many countries in Africa, agriculture osteichthyes labor intensive. So there's some pointers to that direction. Again, fish avant then our courts and our courts have found that improved access to electricity not only increase the Australian I coach I put activity, but then it also facilitates the entry and exit of firms. Google found that in the US, lithification significantly impacted the structure of occupations. So clearly we see from Dell each other at Leptis they might be playing a role in allocation, labor allocations. And we would like to focus on that in the case of Ghana as well. There's also been studies that have found that the emergence and growth of ICT is an important enabler of structural transformation because it gives you provide new opportunities for employment and ends up in Yathrib and stuff like that. Kayla and tab find the raw Vietnam that I mean the arrival of paint and it's increase. I will coach, I put activity and we see that ICT hours also, yeah, come up first. Phosphate in terms of weights access to rural communities in Ghana and want to re-examine how that may be influencing or whatever we find. As the RAU transformation tried a tree or pattern in Ghana. So our current fucose, So what we are focusing on as a first draft is we are working on the data. We are trying to construct various measures of transformation from there five runs. This is to tell the trends and patterns story of Fourier transformation in Ghana. Again, we update knowledge I review to include more studies of transformation and see what we can learn from the literature. And then also, at the moment, we are exploring what is the best empirical strategy to adopt or what is the most appropriate strategy to adopt? To ASA secondary research question, which is to examine the drivers are the constraints to rural transformation in Ghana. Thank you. Thank you so much Kwame. All right, We are going to open it up to the group machines comments. Not seeing anything in the chat just yet. Chris, go ahead. Yeah. Thank you so much for me to really an exciting project. I'm, I'm interested to know a little bit more about the electrification and mobile coverage data that you plan to integrate with TLR that you already have the data. And if so, have you look at the spatial spread and how that matches up with the GLM survey community. I'm unaware of that. Over such a long sweep of time are able to correlate the minified necessarily going to cobble that the correlation between differential rate of structural transformation and differential arrival of electricity and don't serve. That strikes me as really promising line of research. Can you say a little bit more about the electrification and mobile phone data point? All right, thank you very much and please focused. And so this electrification in them, Luba coffee States is, is, is it the Saudis? You're right about that. It's also switch areas had electricity, at which point in time. So about a 30 year period we are going to, for each area, each communists, your district, we will see at what point that's community was electrified when electricity was first introduced to that community and then similar to the mobile coverage, at what points that community had access to mobile, mobile, mobile service. And since they're GNSS, also has kinda jealous as we will be able to identify which communities and district the respondent is coming from. So that is where we are looking into Max data are of the household level data on the to the point where that community in which they respond to entry sites received electricity and then we see it mobilizes and then we try to see the relationships about it. Right. Any other questions? Comments, Monica? Thank you. Yeah. Thank you, Kelsey, I just wanted to add to what Kwame said am Chris was asking if we had access to the data, if we had actually seen the data, and I just wanted to respond to that. So fortunately for, for us, one of our mentors actually has the data. Justice has the data. He's worked with it for some time. So he's agreed to share the data with us when we start working on that second part. And he's going to guide us in terms of how to actually merge it to the GNSS data that we have at the moment. So yes, we will have access to the data, but no, we haven't actually looked at the data ourselves yet, but we plan to do so soon after we finished dealing with the first parts of the research questions. Thank you. Thanks for our Monica. Thanks, Monica. I can just follow quickly. It strikes me that that both physical different from and temporal different from how long have you had access to electricity or telephone service? And especially in the case of telephone service, how far are you from community that have telephone, chairman. And how long has it been that you hadn't easily approximate service could all be very important features in helping to explain patterns of structural transformation. So I think you'll want to really explore the data. And this is very prominent. Thank you, Chris. Thank you. So you so thank you. Okay. Other questions or comments or advice for the team? Okay. I'm not seeing anything else. Uh, so this is going to wrap up our presentations. Thank you to all of our presenters for very well presented topics, for keeping on time, for your good handling of questions. I really appreciate all of your efforts and for everyone that attended, thank you for taking the time to to get a sneak peek at what our teams are doing. Hopefully, we will have the same session at the end of our program. And alumnis presentations will be complete projects and you can come and check back in with the teams. But with that, if there are any final comments or questions, I will open up. Chris, you have anything that you'd like to say? No, I just wanted to thank the long bow at the Michigan State for setting that up and kill people organizing everything that Well as always. And thank you, especially to the stars team than their mentors. Barricading the two that were coming along nicely. That great project. Very, very eager to see how these programs to watching your, your development of outstanding and highly irrelevant project. Everyone in the company working that much, everyone. Thanks everyone. Have a great rest of your days. I expect that we'll be able to post to the recording of this, Steve, I'll follow up with you, but yes, thanks for for coordinating and we will talk to you hopefully soon. Take care, everyone. Thank you. Thank you. Bye. Thank you. Bye-bye. Thank you. Thank you. Bye. Bye.