 Welcome everyone. Welcome everyone to the Center on Finance Law and Policy's November Blue Bag Lunch Talk. I am thrilled today to welcome Professor Eduardo Montero from the Ford School of Public Policy. My name is Christy Bear and I am the Assistant Executive Director of the Center on Finance Law and Policy. And as usual, before I can let you have what you came for, I have to do a little promo for an upcoming event, which is the Center on Finance Law and Policy is in less than two weeks hosting our second Central Bank of the Future Conference. This year will be co-hosting with the Federal Reserve Bank of San Francisco. If you are interested in financial inclusion and monetary policy in Central Bank digital currencies, then you definitely will want to stop in the conference is free. We have a impressive group of a few dozen speakers for this, including Mary Daly from the San Francisco Fed. Sorry, blanked for a second. Assistant Governor Shae Sarri from the Central Bank of Cambodia. Futurist Dr. Claire Nelson and former CFTC chair, Timothy Massott so it should be a good event and I hope that you will attend. I put the link in the chat so that you can see how to register. So without further delay, I want to tell you a little bit about Eduardo Montero, who has agreed to present to us today. He joined the faculty of the Ford School of Public Policy in 2018 as an assistant professor of public policy. He's originally from San Jose, Costa Rica. And he's an economist whose interests include development economics, political economy, economic history, and specifically he studies how variation in institutions, such as property rights and cultural norms, like mistrust, affect development and development policy in Central America and Central Africa. That is a mouthful he's going to make it sound so much clearer than I just did, which is why he is the professor and I am not. At the Ford School Professor Montero teaches microeconomics, applied methods for development, and a class that's called the economics of developing countries. Today he will be discussing one of his papers, cooperative property rights and development, evidence from land reform in El Salvador. And you should have seen a link to that paper in the email that you received advertising this talk. So if you have not had a chance to download it and read it, I hope you will. But if you didn't, good news, he's right here, and he is going to tell you all about it. And so without further delay, Professor Montero thank you very much for being here today. Thank you so much for the invitation and for the really nice introduction. Really excited to present this work to this audience. Just because it's so interdisciplinary it's a topic that I think is quite important for Central America but there's not much work but I think this audience will have a lot of interesting perspectives. So really come forward to that. The project is called cooperative rights and development evidence from land reform in El Salvador. And I'm an economist, and to a lot of economists, when they see this title, so they see cooperatives, they see land reform. The first thought that comes to their mind is an image like this, an image of Soviet collective farm propaganda and a really negative reaction, they're like why are we studying this we know these were a failure, like why are you wasting our time. The definition of a cooperative actually is, this is a firm where workers jointly own and manage production, and they make decisions on a one member one vote basis so they make decisions. Democratically, we see many examples of cooperatives around the world today. So even just in the US. For example, law firms in the US are cooperatives where the partners on the firm, and they share profits. It's a very common type of ownership structure in the US even though we don't think about it as cooperatives. Also in the US in the Pacific Northwest timber production is historically been done through these worker cooperatives as well so there's examples of them in the US. And across the world there's like many other examples we can think about. This is the famous kibbutz system in Israel, and agriculture cooperatives in Latin America. These cooperatives are particularly prevalent in agriculture in Latin America, because of land reforms that happened between 1920 and 1990, that really tried to create these cooperatives. So here's a map of Latin America and countries in dark blue are countries that experienced a land reform that actually created agriculture cooperatives managed by workers. As you can see it affected about half of the countries, even though today we'll be focusing on the country in red. So, just taking a step back, these cooperatives are very common form of organization. And there's been arguments about the pros and cons of this type of organization. So on the pro side, some people have argued that giving workers ownership stakes and decision making rights may have beneficial incentive effects, and also have beneficial equity effects. But there's been a lot of work arguing that this is actually bad for efficiency. This is because profit sharing may also lead to free writing problems. If we're all sharing profits I could potentially slack off and get some of your profits. So while cooperatives might increase equity potentially they decrease efficiency potentially. So, there's been a lot of this theoretical work in economics in particular about how cooperative ownership may differ from outside ownership. So by outside ownership I mean where there's an owner who contracts workers but isn't working himself or herself for equity and efficiency. But there's very little causal empirical evidence. And by that I, what I mean is that it's been very hard to study what are the cause like what how does being a cooperative affect equity and efficiency. The empirical challenge here is that property rights are not exogenously determined. Basically they're not randomly determined. The type of ownership structure that's chosen in different places is affected by a number of other characteristics that we think might affect our outcomes of interest, which is efficiency and equity. So for example, cultural values, Latin America has a long history of cooperative ownership, but not so much perhaps in other places. Capital requirements might also affect whether a firm becomes a cooperative or not, but also things like equity and efficiency. So it means that comparing all cooperatives so non cooperatives empirically won't provide causal evidence will give you correlational evidence but that's not causal, because this comparison could capture the effects of the other differences on our outcomes of interest, rather than ownership differences. So this meant we kind of lack this empirical evidence. For outcomes such as, you know how to benefit workers benefit what are the impacts for development. So what I tried to do in this paper is I asked this question. So what are the impacts of cooperative ownership relative to outside ownership for equity and efficiency. And to try to provide some empirical evidence on this, I examined the land reform program from a salvo in 1980 and take advantage of a particular feature of the land reform, which was that all land belonging to individuals with cumulative land holdings over 500 hectares was reorganized into individuals with cooperatives, but this didn't happen to properties below this threshold. So what I'll do is I'll use our aggression discontinued design, which just the intuition is very simple be comparing properties that are just above and just below this ownership threshold, where the properties above this threshold were reorganized into cooperatives managed by the But the properties right below that are probably quite similar remained in the outside own system. So specifically for the case of a salvo, these are also known as assiendas. And so I'll use this design to examine the impacts of cooperative ownership on outcomes such as the efficiency of production specialization so which crops they decide to produce an equity so how our worker incomes affected. And I think this contributes to three literatures, but I'm open to thinking about other literatures outside of especially like legal, for example, literatures. First, again, I think the main contribution here is providing some of these empirical evidence about equity and efficiency what are the implications of cooperative ownership, we don't have a lot of literature on that. Second, I have a model in the paper that tries to think about the effects of this. These types of ownership structures in settings within complete contracts when I mean by that words it's very hard to write very detailed contracts and that's kind of the case in agriculture and I'll talk about that later. And then more broadly just thinking about the development of Latin America. So historically there's been a lot of these land reforms. As I mentioned earlier, but we don't know what are their lasting consequences. It's also important for development policy today in Latin America, there's a lot of these cooperatives. And it's quite a contentious topic. So I think just providing any improve evidence is hopefully a contribution that might inform policy and I'll talk about that a bit more towards the end. Let me first start off by talking about El Salvador and the 1980 land reform. I'll then go into some of the data I collected. Some of the empirical results, specifically for agriculture. Then I'll go into some focus group evidence to try to understand the patterns I see in the data. And then I'll go over some of the results on incomes for workers. So first on El Salvador, El Salvador is actually the most densely populated nation in Latin America. It's about the size of Massachusetts in terms of area and it has a population about 6.5 million people. And historically it's had very extreme land inequality, even compared to other Latin American countries. So for example, in 1971, just 1.5% of landowners held half of all the agricultural land in El Salvador, it's pretty staggering amount. El Salvador experienced a very brutal civil war from 1979 to 1992 has a history of being run by military sponsored leaders. So since 1932, and a lot of elections with a lot of fraud, where the military would basically commit a lot of fraud. In 1937, there was a fraudulent election, violence broke out. And a coup was undertaken in 1979. There was a government that comprised both civilian and military, and so there was a lot of fighting until 1992. And land inequality was actually called the epicenter of this crisis. For the start of the civil war, there was a large demand for access to land that people are upset about how much land the elites controlled in the country. And to sum up this, on March 5, 1980, the military junta government passed decree 153 on land reform, and this reform focused on the reorganization of these asiendas into cooperatives, and it's supposed to be carried out in two phases. So phase one called for the immediate reorganization of all land belonging to individuals with cumulative land holdings over 500 hectares. Phase two called for the organization of all land belonging to individuals with cumulative land holdings over 100 hectares. And it said a few years after phase one but it didn't specify when that was supposed to happen. And in fact, phase two would never carry it out actually due to organized opposition following phase one. So only phase one happened phase two never happened. Why did the government decide to do this land reform, officially, according to the law, the land reform had just three motives. The first was to try to diminish this land inequality that I talked about and try to increase agricultural productivity. They thought that the asiendas were unproductive and if workers got ownership rights, the agricultural sector would be more productive. Second, it was to try to increase incomes and reduce poverty in the rural areas of the country where poverty is very widespread. And third, it was to try to reduce the privileges for this economic elite that had basically ruled the country for many years. Unofficially though the land reform was also implemented to try to reduce conflict and keep people from joining the insurgency. So that was an important part of it as well but that isn't in the official, you know, law. So what was this land reform plan. So it was actually prepared under immense secrecy and execute a very full velocity is prompted by the unexpected addition to the junta leadership on March 3 of a pro land reform Colonel, Colonel Mahana. And between March 3 and March 6 so really fast in the span of three days. Basically the entire planning and execution of the slander form happened. So on March 4 they called for this inter agency coordination seminar was actually fake. And everyone from the Ministry of Agriculture was important there. And from other institutes these green key cards that barred them from leaving the hotel so they're basically locked in there for two days, and they were given national police escorts everywhere they went. There they drafted the law, they published it on March 5, and then the military gave them overnight transport to the SDS to notify these SDS owners of the reform. I mean, this 500 hectare threshold was just chosen to be a temporary threshold for implementation capacity reasons, they didn't feel like the military didn't feel like they had enough personnel to do phase one and phase two at the same time. So it decided to break it up and they just chose 500. So this is executed. So phase one was carried out immediately after the form was announced. So it was actually published on March 5, and then March 6 was when people were notified. The way this work is that intervention teams made up of agronomists technicians and military personnel were sent to notify the owners. These owners were compensated with these 30 or bonds based on their tax declarations from 1975. They didn't get compensated, but they're pretty upset about the compensation. It's a pretty contentious issue, because a lot of them had been, you know, claiming that their property wasn't very valuable to avoid taxes. So they kind of got a bit upset when these, these bonds were tied to these tax returns, but they did get compensation. And so by the end of 1986. So this is a map of El Salvador, where I'm showing you the cantons of El Salvador, which is the lowest administrative unit of El Salvador, there's about 1400 cantons, and I'm coloring in light blue, the cantons that had at least one asienda become a cooperative. And so as you can see here, the reform is pretty widespread across the country, it wasn't just concentrated in one area of the country. It was pretty widespread so the military went throughout the country to try to implement this, this law. A lot of the properties, the biggest properties happen to be on the coast because that's where the most fertile land is. But they did expropriate land in a lot of other places. So what is, what are the aftermaths of these, this land reform. So for landholders, obviously they weren't happy, it was called an economic, political and social earthquake in the countryside. Landholders saw before their eyes something that they never imagined could possibly happen on lands that they had always governed absolutely. And it's coming from a person in the military who helped draft the plan, but who was very tied with the elite. But what about for workers. So this is from focus group evidence that I did in El Salvador just talking to cooperative members who happen to be alive at the time of the reform just to get their memories of that time. And I heard this like throughout this phrase, they would say to me on March fifth, who ends to sleep as poor colonos or sharecroppers, and on March six we woke up rich as landholders to them it was a shock a complete shock, completely changed our lives. And in fact, since 2013, March six is known as the land reform commemoration day in El Salvador disperse importance in the country's history. So let me tell you a little bit about the data I gathered the empirical strategy I use. So I collected pre land reform, land holding data from the property registry of El Salvador from 1980 right before the reform was implemented. And this includes property size, location in the country at the cancel level, and the name of the owner for every property over 100 hectares so I wasn't able to find records for properties below that only properties above 100 hectares. Then I gather the land reform records archival records from the Ministry of Agriculture, and then this Easter which is the Institute for varying transformation which is created for these land reforms. And they continue lists of the properties that were reorganized into cooperatives with the name of the cooperative, and the name of the old ascenda. And the former owner for each of these. That was reorganized into cooperatives. So I'm able to match these two records. And that allows me to use this regression discontinuity design that I talked about earlier, where the intuition super simple I think economists can often do this they like give fancy names to something that's very simple. So let's be comparing properties just below and above the reforms ownership threshold so these properties in 1980 were probably very similar just one happened to be owned by an owner that had over 500 hectares. Whereas the other one was owned by someone who had slightly below 500 hectares. So here's the empirical specification but I won't go into the details too much but two things that are pretty important to assumptions for this type of design. Are that first owners weren't able to selectively sort by that they weren't able to sell off land and avoid the reform. If that was the case then people right above the 500 hectare threshold could have avoided the reform transferred land etc. This was difficult to do for two reasons one there was a freeze on all land transactions since 1977. So the blind reform like I said happened very quickly. So we can look at this empirically and see is there evidence that there's less, there's fewer properties right above 500 hectares, we don't see that so here I'm showing you the number of properties at each owner land holding them out and we don't see a discontinuity right at 500, suggesting that owners weren't able to selectively sort out to avoid the reform. If they did we might be worried that these owners who sort are different than the ones that don't. So the other assumption is that properties that happen to become cooperatives are very similar to those right below the threshold that remained as us and us. So we can look at this by looking at differences in geography. These are things we think might matter for productivity worker incomes etc. We can compare these things and we see that actually these properties right above and right below are quite similar consistent with the way the reform was planned and this arbitrary threshold being chosen. So on things like land suitability precipitation rainfall elevation, we see that these properties are very similar. We also can look at the suitability geographic suitability for a number of different crops. So coffee suitability shirking suitability and cotton suitability these are the three main cash crops in El Salvador historically cotton is no longer produced very much in El Salvador. But coffee and shirking are in there, big part of the agricultural sector, we don't see differences in suitability there, and then we can look at staple crops and two main ones in El Salvador are maize and beans, and we don't see differences there so what this suggests is that these properties are quite similar. They were quite similar before the reform, except that some became cooperatives and some didn't. Another important thing here though is because this reforms a bit unique is, did they actually enforce this threshold, you can imagine they might not have done that. But we actually do see that they enforce this threshold, pretty well, so not perfectly for sure. But we see here is that holdings over 500 mostly became cooperatives. So what I'm showing you here is on the x axis is the cumulative land holding of the owners, and then the y axis is whether or not the property became a cooperative. And as you can see there's this discontinuity right at 500. So, right as you pass 500 hectares and land holdings, you're more likely to have your property become a cooperative. So this is the variation I'm going to exploit. So some properties were able to somehow avoid the expropriation, for a number of reasons. And I'm happy to talk more about that, but about, you can think about like 80% of the properties that should have become cooperatives became cooperatives. Okay, so let me go into some of the empirical results, specifically for what choice crop choices are they making and their productivity in these crops. So I gathered data from the census of agriculture in El Salvador from 2007. This was the most recent census that they've done since 1971. They just because of the war and a number of other reasons they hadn't conducted. It's conducted by the Ministry of Agriculture of El Salvador, MAG, and it's pretty detailed it contains information on the types of crop produce the area cultivated by crop, and then the amount produced for each crop. One shortcoming is that it doesn't have information on worker hours or worker incomes do something we do really care about. So I compliment this data using household survey data from El Salvador. El Salvador is the proposal multiple in many ways of it. This is a smaller sample size is not a census but it contains very detailed information on the worker incomes hourly wages and their consumption. So I use these two datasets to look at efficiency and equity. So the first result that I found is that these cooperative properties so the ones right to the right of the threshold to vote less land to these cash crops so coffee and sugar cane, but to vote more land to staple crops, maize and beans. So here I'm showing you what are known as the regression discontinuity plots, and you can see there's a big discontinuity and the type of crop produced. When I look at the yields so a measure of the productivity. So how much is produced per area unit of land. I find that cooperatives are less productive at these crash crops as well. So their sugar cane yields are lower and their coffee yields are lower. But they're more productive at staple crops actually. So their maize yields are much higher and their bean yields is plausibly higher. And so what does this tell us just to summarize what we found so far. This is relative to Sien does these outside own systems cooperatives owned by workers are more likely to specialize in staple crops over these cash crops. They devote more land to them, and they're actually more productive with them compared to cash crops. I also really understand what was going on here it's an interesting pattern. And so I decided to go to a sovereign 2017 and interview a lot of these cooperative workers and some of the workers to try to understand these patterns and see why are they choosing these crop choices. I think a lot of economists think that cash crops are which everyone should be producing because they're higher returns, but I want to understand from their perspective why they weren't doing that. So let me just talk about a little bit of that and the conceptual framework for the paper. By interviewing a few cooperative members. They really highlighted that this, these stark differences between cash crops and staple crops in terms of contracting and sharing profits. So cash crops, like coffee and sugarcane basically require centralized processing to be valuable. Otherwise they rot really quickly they're not that valuable so think about coffee or sugarcane coffee beans themselves aren't really valuable you need to process them take off the pulp and dry them. Once they're dried, they have to be roasted. And once they're roasted that's when they're valuable. This happens to this only happens in a centralized manner same as sugarcane, sugarcane actually rots really quickly. So you need to process it very quickly. What they told me this meant is that it's actually really easy for them to pay workers based on how much each worker produces, and it's very easy for them to keep track of in this cooperative how much the person contribute to the cash crop production, and then decide to share profits based on that. So they share profits for cash crops. In contrast, for stable crops, these crops are quite different. They're not processed centrally, and workers actually can consume their own production very easily so if I produce maze, I could just set of, you know, bringing it to the central cooperative center. I can just consume it myself. What they told me this means is that it ends up being quite difficult to pay workers based on how much they produce. And it's very hard to share profits here. So imagine we're sharing profits for our maze production. Well, I could just consume some of my maze production and still get the benefits of other people's maze profits. So that means that cooperators are like it's not really worth it to us to try to share profits for stable crops, just because of these differences and this is something that I, they told me a lot about it and it's quite quite interesting to hear them talk about it. So just to summarize what I found here is that cooperatives really do share profits for cash crops, but not so much for stable crops. And this partially drives some of their decisions that they're making. I also learned that cooperatives have to agree to this cooperative constitution, which I think is super interesting too. So I think there's four things. So first is democratic decision making is is a rule. Most decisions have to be made by a one member one vote basis. Things like how much land to devote to different crops. That's often democratic. And a lot of these decisions and how much to pay workers how much profits to share. That's democratic and most of these things require just a majority a simple majority. So there are things that require super majority. So selling land requires super majority so say I'm a cooperative member and I want to sell some of the land. It's really hard to do that I need to get 67% of people to agree. And then there has to be a public government managed auction. This is partly done actually so that the former owners didn't go back and buy the land. So that's why the government manages these options that they make sure the land isn't going back to the former owners. So selling a cooperative actually pretty hard as well. So, because profits are shared across members they want to be really careful about who they let in. So they often require a super majority. In the case of a family member joining after the death of someone, they're only allowed one member for family for sure. The other members can then apply to join. So selling a cooperative is also hard. So if I'm a part of a cooperative and I want to leave. This also requires a super majority. And I lose all value of my membership otherwise so I lose the value of my land. Anything else that's provided like schools or other things like that that are provided by the cooperatives. And so, if we think about this from an economics perspective it's actually interesting design. There's been some work in economics about potential incentive issues in cooperatives because of this profit sharing. There's been three things so the first is adverse selection. What that means is that people who are not going to contribute the good of the cooperative might join just to benefit from the profit sharing. So it's actually kind of addressed by two and three it's pretty hard for people to sell or join. There's also the sphere of brain drain. What that means is that the most productive members might decide to leave the cooperative because they don't want to share profits you know they're they're producing a lot of the, you know, the total amount in the cooperative. But they have to share some of it, they might be better off producing on their own. That's called brain drain, but that's kind of addressed by 24. Then there's this other thing that's known as moral hazard and effort, which is the focus of the conceptual framework which is this idea that profit sharing might lead to me working less hard because I know I'll get some profits from you. So how likely is that to apply in this case, especially compared to outside owned asian thus that also suffer from this problem. So, to try to structure the interpretation. I provide a principal agent models to structure the analysis. So, in the model, again, just cooperatives make decisions via voting workers are the ones who decide. And that's really important. And the outside owner system. There's just an owner. Maximize their own, which is very different. So the model explains what are the implications of these two types of decision making rules and things like choices to two main features. So first is owners can't perfectly observe and contract on effort. They can only contract on output. They can't write contracts that say you have to work this hard. I can only be like, I'll pay you if you produce this much coffee if you produce, you know, a little bit more I'll pay you a little bit more etc. The small assumptions that workers are heterogeneous. And then crops different their contract ability matching focus group evidence so cooperatives can share profits for cash crops but not stable crops. So outside ownership systems, what this means is that it's really hard for the owner to pay people share cropping contracts for cash for stable crops but not cash crops. So model has two main sets of predictions I'll test in the subsequent analysis. So interestingly, if you're able to specialize in different types of crops, neither ownership structure necessarily reaches what we think about as the efficient outcome. There are two different reasons. In asian does the, there's this motivation rent extraction trade off. The owner wants to increase his own profits. To do so he doesn't pay workers as much as he ultimately should, giving them less than ideal incentives. And so he's trying to balance these two forces I could pay them more they work harder but then I get less profits. So there's this inefficient trade off that happens there in properties. It's the standard profit sharing problem. If people can vote, then you might have too much profit sharing has another a number of other predictions for crop choices productivity and equity. So equity I haven't talked about that. And that's what I'll talk about next, but it matches some of the other results I've shown you so far. So cooperatives are less likely to specialize in cash crops. So cooperatives will specialize much more in staple crops and are really productive at them is because workers really care about them and they benefit from producing staple crops. It also predicts that cooperatives will have higher equity. So there'll be less income spread. So let me talk about these two predictions. So first what are the implications for just aggregate efficiency before going into implications for equity. So when I look at aggregate profits, very little evidence that they're that they're starkly different looking at figure 16 this is showing you the standard measure for the productivity of a, you know, an agricultural firm just profits per Hector in this case but any land unit. We see that there aren't huge differences in their profits, which is really interesting. There's some differences in their revenues per Hector. But what we really care about is profits. This is because cash crops have very high input costs as well. So they might give you higher revenues but also higher costs. And that's why we don't see much differences in profits. And these differences are a bit imprecise. But I think this kind of goes against the stark idea that some economists have which is that these cooperatives are for sure going to be much, much less efficient. It doesn't seem like a strong evidence for that but it's imprecise for sure. And so just to summarize the agricultural results before I turn to worker incomes. I find that cooperatives. If they're able to they specialize in staple crops over cash crops these are workers that these are crops that really benefit workers they can consume their production and workers don't have to share as much of their income for this. And I find inconclusive evidence for aggregate productivity differences, we can think about that as like overall efficiency. And I think the model helps us think that it's probably due to the ability to specialize in different crops. If they just focused in cash crops, probably wouldn't be the case. They just focus in staple crops. Maybe we'd have a different pattern, but here because they can allocate work to different tasks. We don't see these differences which I think is quite interesting. You could apply to a number of other cooperative firms where there's this ability to do different tasks. The labor has a number of robustness checks and extensions that I won't really talk about right now just in the interest of time, but I'm happy to talk about them much more in the Q&A. But let me go to the income results. So income and distributions of cooperative workers. So here we're going to look at how does cooperative ownership effect worker incomes in El Salvador. So to examine this, I use the household survey data. And the survey includes a number of questions on where people live, whether they're a cooperative member or not. And then the number of employees for that property. So I can use these questions to match individuals to cooperatives and as in this. And I check the matching using some household surveys or actually do have the property and cooperative name where they work. The caveat here is that it's a smaller sample properties in the census the census includes, you know, like 98% of properties in El Salvador but it doesn't the household survey doesn't include that. So that's just the caveat, but turn to the results. In column one, I'm showing you differences between cooperative workers and Asienda workers today on their earnings per capita in the previous month in dollars. So here, we're finding that cooperative workers on average seem to have make $50 more per month than Asienda workers, which is actually a really large amount because in these areas, individuals make about $100 a month in agriculture. There's a pretty stark income effects they're a bit imprecise, but they're pretty large. And turn to the distribution here measuring the spread of income so let's compare all workers in a cooperative and find out how spread out their incomes are same thing for Asienda workers. You find that cooperatives have less spread out income so more equitable income so not only do they have higher incomes, they also have more equitable earnings, which is really interesting. This doesn't take into account the earnings of the owner for the Asienda us, but it does show you that the workers so the people actually doing the production have higher and more equitable earnings. And we can try to dig into this a bit more and see who benefits the most from cooperatives. Low income workers in particular or is it like high income workers in particular. And so here we're looking at quanta estimates. What that is is split all the workers by their income earnings and look at cooperative versus Asienda workers for low income workers and go off the quanta. And as you can see by this estimate, it's very positive for low quanta and then zero for high quanta and perhaps somewhat slightly negative. What it suggests is that cooperative ownership really benefits the low income workers, the most. This is very consistent with this idea that cooperative share income across workers. So if I'm a low income worker, and I happen to be in a cooperative with a high income worker. That's really good for me because we're helping each other out we're sharing earnings, which doesn't happen on Asiendas are outside of own firms this type of ownership structures particularly beneficial for low income workers, like the workers in all Salvador, I think that's very important. So, let me just conclude right now. What I do in this paper is I examine the 1980 land reform, I use it as a setting to try to understand what are the empirical differences, causal differences between cooperatives and non cooperatives, and I find three main things. So first is that cooperative ownership induces different patterns of specialization, particularly specializing things that workers care more about or benefit the most from. So in this particular setting what that means is that cooperatives are less likely to specialize in cash crops and more likely to specialize in staple crops where workers are the residual claimants. I don't find strong evidence that cooperatives are less productive. So not strong evidence on efficiency, but do find large impacts for equity so cooperatives tend to have higher and more equitable incomes. And I think these these results are potentially informative for cooperative policies more broadly. We don't know very much about cooperatives I think people have very strong gut reactions about cooperatives. But it's very important for policy to the point that the United Nations for example deemed 2012 the international year of cooperatives for their potential to help low income workers, but also did note that there isn't a lot of evidence on this. So just providing something that helps helps us understand when our cooperatives good or bad where their impacts on workers can potentially be informative. Thank you. That's all I have in terms of the main presentation, but I'm really looking forward to any questions and and comments as well. Thanks. Terrific. Do you want to stop sharing screen and then we'll kick it off, Professor Yang, do you want to go first. You're muted. Sorry, Professor Yang I was trying to call on you. Go ahead. Great paper. Yeah, so this is, it's striking that there's not not an income difference that you're finding. And I guess, you know, I interested in what you think about, you know, just the broader impacts on life satisfaction. When you think about non pecuniary being in a cooperative versus being a Hacienda worker. Certainly, I think I would presume that people are happier. All things equal. Yeah, being a cooperative member, you know, part of the decision making structure of the enterprise versus just being an employee or a tenant farmer. You know, you're, you know, so I think, you know, we think about, you know, more broadly, you know, utility as opposed to just income. What your thoughts are on that and whether there's any way you can measure, you know, sort of broader life satisfaction like just the phpm survey have a question along those lines. Yeah, no, that's an awesome question. Thank you. The HPM doesn't have a very good question on that they have questions on consumption. I also find positive effects on consumption, which you would expect just given the income levels. But I can tell you from just some like focus group evidence, I think you're right like, I don't speak to this but speaking to corporate workers there they there's a lot of pride and owning and making your own decisions which I don't think that's captured and being an Asyanda worker. And I think that's critical. And that's one thing that interested me interested me a lot originally in this project was thinking about the differences and how they see themselves, like sense of self and agency just to get very good data on that. I think that's definitely the case. You could potentially imagine that very productive Asyanda workers are happier like if they're, if they don't have to share income, but you're very productive in an Asyanda you might be better off potentially because then I don't have to share income, but that doesn't capture these non-procurenery things that you're talking about the ability to make your own decisions to feel part of a community I think that's something that they talk about a lot that they're in this together that they can help each other out during shocks. And that's something I don't look at in this favor too much but it would be very interesting to see like workers in both types of properties were after facing shocks. My sense is that there's so much more resilient in cooperatives because they come and help each other out, share income which you wouldn't also have in an Asyanda and that's hard to value gives you utility if you're risk averse, which we think people are, but that's hard to capture, you know, in surveys or something but yeah I think there's a lot there that would be super interesting like these other outcomes. So if you want to ask questions, otherwise just raise your hand and otherwise you know I have questions. So I thought it was so interesting when you were talking about how people went like overnight boom they became landlords. And then when you were spelling out the differences between staple crops and cash crops and talked about the processing now I don't know what's involved in making coffee, or in making sugarcane but my immediate thought was, well, there's really some equipment that would need to that you would need and so you didn't go where I thought you were going to go, which is like oh well now you need the capital to be able to buy the equipment to process these crops and so I'm curious if you if capital factors into it or if you looked at that. Absolutely. I think that's an important part of potential difference between just broadly cooperatives and outside owners, like, it's hard for workers, especially in the setting to gather capital you would think. I think one thing that are yes two things are important in this setting is like you said the, the reform happened very quickly. So kind of like overnight, which meant that workers were owners weren't able to like take their processing capacity with them or destroy it and a lot of other land reforms, like if the reform was announced people are like okay I'm going to take the important parts of production. So I meant that mostly the cooperatives already had some infrastructure so they didn't need to raise capital for that infrastructure. Historically like the big infrastructure now there could be differences in credit access today, which could drive the effects. But the important thing in El Salvador is, so in this in the data, there are no differences in credit access. And the reason is that the cooperatives in El Salvador actually can use their land as collateral as as like the cooperative as a whole. So I think that's different than some settings where you can sometimes can't use a firm as collateral here because it's land, they can, you know, access loans, actually very good loans from banks and stuff, using their like pledging their land as collateral and there are because prices of agricultural commodities go down, but if that's a default on loans, this also happens in Aciendas, but they do have access to credit here through that mechanism which might not be the case for cooperatives and other settings where you can't you know pledge land as collateral. I think that's an important thing. I do think credit access is huge if you're thinking of like expanding a policy like this to other types of industries. Workers might not be able to raise credit. They might be constrained and that might be another reason for differences. So I think that's critical question I think the setting of El Salvador is interesting for that reason land in particular but I think you're right like that's specific to this setting. Interesting. Amy and then Triana. Hi, thanks for a really good presentation. Thanks. I had a question actually very similar with respect to infrastructure. The reason that I ask is because I was recently in El Salvador and Guatemala also do reviews and I spoke with just a couple of cooperatives and certainly not about these topics. But something that consistently came up is that they often were receiving external assistance for these kind of hard infrastructure inputs. So those processing pieces farm to market roads. And those were allowing them to start to make some of that transition from kind of more stable crops to more cash crops. And I suppose I was interested if you see that that's kind of, you know, maybe just this kind of random noise or if you think that there's any sort of role that might be played that cooperatives are going to qualify for you know this type of assistance to potentially increase productivity, whether it be you know nationally some sort of regional assistance or even kind of international assistance or whether his cooperatives and qualify as he had is obviously not. And I'm interested in whether you see that you know that potentially plan you will and give it totally similar. That's a really interesting question. Oh, thank you. Thank you on about it for a while but it is the case that like international organizations are quite especially in El Salvador interested in these cooperatives a lot, and they do get access to that so in the census data there's a question on, do you get help from NGOs and I mean there's a there's a jump there. So cooperatives are more likely to help from NGOs today, compared to us in this. So yeah I think that's definitely the case. I'm not sure how. So I think one thing I've done is look at the cooperatives that don't get NGO assistance, and I find similar patterns there. And that to me suggests that maybe this these NGOs are helping but they're not drastically changing their decisions. Because you're right it's a lot of it's kind in some ways I was I thought about a little bit like getting credit access another way to get credit really cheap credit, you know, because it's like input assistance or things like that. Something that they did talk about a lot and do you know this much more than I do is they're, they're able to like get access to inputs and things like that that they need up for the further farms. And sometimes through NGO help but also through banks, but what they complain a lot about is like roads and road maintenance. So it's things outside of the cooperative I think they're, they struggle with that I mean historically there's reasons for that so the government. That's it go over and after 9092 after the Civil War was very right wing. And they were really against these cooperatives. They are actually, they didn't invest in road maintenance and reasons with more cooperative and stuff like that so I think I think there are these political forces at play as well. In 2013, though, another government doesn't want any other government to go over. That's more amenable to these cooperatives and has helped them out a lot so they say things about a much better but I think there are these ideology forces at play here, for sure for things like road building and things like that. There's a think about a bit more how to test that yeah. Brianna. Hi can you hear me. Yeah, so I was thinking about the heterogeneity of like geography no salad and how that might affect like the proportion of lands like below and above the thresholds and how that might affect like the concentration of asiendas and cooperatives in different areas and whether that had an effect in terms of like, maybe if there's a higher proportion of cooperatives in some areas maybe they'll like they'll be like increased production or because of like, like network effects and maybe the sharing knowledge or whatever or not and whether you saw something like with that or whether you studied it or whether you had enough information to analyze it. Yeah, no that's a great question I think I mean I do think that there's a lot of these network effects for agricultural production like you really learn from what people around you are doing. In a lot in general, the coastal areas I mentioned this really quickly. Those are the sugar cane areas, except land is super suitable for sugar cane. There was a lot of cooperatives there, but they choose not to produce as much sugar cane. So, so they happen to just the case of sugar cane specifically. What I would say for a bit is, you would think that would go against some of the findings, they happen to be in areas with a lot of sugar cane production around these networks of sugar cane producers, but they choose not to produce as much they definitely produce it but not as much. So I think that was interesting to me that that kind of suggested it's always for sure. It's not as driven by by geographic, like geographic location, a little bit more by these workers making these decisions like we'll produce but we also want to produce some other crops for ourselves, which I think it's kind of what's going on. Coffee is a bit trickier because coffee. It really depends on the altitude so like even in a like a small region, like, like really tiny region, there's so much variation in coffee production quality because it depends on the altitude but also the shade and like a number of other factors. I have to think about how to look at that much more but I have tested whether there's differences in like the altitude of these places so elevation, which is really important for coffee. I don't see differences but again that's coffee is very tricky like it you know you could have like a property that's at this level but one that's like that and that one's better for coffee because of the shade and it's really tough. I have to think about that more but yeah just for sure again I think it would it goes in the opposite direction of what I find. I have a question from the chat. Can you say a little bit about how are the original co-op members chosen and relatedly it sounds like they can never leave. Good Lord. Yeah it's not that bad. So just on the first question the original workers or the co-op members originally so the way they did it is that the government got there or the military got there and then just told workers that were on the property so these are Asyandas that are really big so workers often reside on the properties so whoever was residing there that morning and then they kind of just trusted the people so they were like hey you guys are the workers here right now everyone who happened to work on the Asyanda before they're a co-op remember give me all their names but they limited the family members so like then you might cheat and be like yeah yeah my family also works here but they only allowed one family member at a time and then they allowed them to have this voting and let people in so that's kind of how they did it but yeah actually the planners talk about how that wasn't perfect necessarily you could imagine some fraud happening and some people claiming to work on Asyanda that didn't but in theory so it was supposed to be the people who worked on the property before which is nice for the empirics because that means that like the workers on these two types of properties are probably similar so some happened to then own it but there might have been a little bit of fraud I've looked for those records of what the names of or at least I haven't been able to find it in terms of leaving you can leave but you forfeit a lot so yeah I think you're right like that's that's one thing that I think is in the paper I'm not able to study but I think it's really interesting to study is how it might hold back you know migration to the city center if there are opportunities into the capital because I can leave but I forfeit a lot like I forfeit land I forfeit I mean the community I forfeit the profits that were sharing there's a lot of things that might tie I guess some historians have called us that cooperatives tie you to the land. It keeps you in the countryside and that might be bad for the economy moving to something like manufacturing and stuff so I'm not able to study that so much here but I think that is definitely an important mechanism to explore is how it might impede you know migration to city centers or things like that. So it sounds like they're making a lot more money to yeah being a co-op so maybe they don't care like why go to the city if you're making so much more money. Yeah, so something I think is important too is that it's very good for intergenerational ability. So I might, if I'm a cooperative member, it might be hard for me to move to the city because of all these benefits, but talking to a lot of cooperative members there, their kids were often in the cities because they were able to pay for their school there or for the university degree. So they have I have higher income therefore I can pay for better education for my kid. Therefore the next generation might be better off and isn't in the cooperative. So that's a potential benefit so it might restrict migration for the current generation, but actually help migration intergenerational migration and mobility for the next generation. So that's also quite important. So the next question you get the last question unless it's a short one. And then somebody gets to show them no pressure. I hope. Thank you so much for professor. I guess my question was, my question was with one of the transition periods that you were talking about was basically that if a family, a family, if the family of the person and the person passes away, they elect one of the family members to take their place. Has that always been the case. And if somebody doesn't it's even an occasion where they haven't replaced that person with a family member or do they just, and if that doesn't happen to the deep that's he'd open or do they like, is there a particular way to do they select someone to take that particular spot. I don't know. I don't know. That's a great question. And so that varies so much by cooperatives. So, you know, in these settings, it's like you have these formal rules, but people have their own interpretations of them. So there definitely are cases when I heard about a few examples of this where someone passes away and their family members actually don't want to join because they're in sunset about or you know working somewhere else. And that happens like someone passes away and then there's one fewer member. There's no like mechanism there, like a spot doesn't open up or something. If people want to join them they can go through the standard process but there's no exception made there. So there are cases where someone passes away, either they didn't have kids or their kids and or wife doesn't like they don't want to join or husband don't want to join the cooperative. So that can happen for sure. The thing that's also there's a lot of variation is to say someone passes away. Some cooperatives are like no we're really strict we only let one person in the other people have to go through a very formal process. Other people other cooperatives are like someone passes away we kind of let all their kids in pretty easily you know like there's not this formality of like who's the chosen person you know we kind of just let everyone in but that depends a lot by cooperative and the rules that they have in place. Yeah, some people are like we don't want to do that because like what if they let in their laziest kids you know like there's a whole there's a whole like internal dynamics of what the rules should be but I think there's a lot of various. I think it also actually depends on how tight or close knit the cooperative is like if the cooperative is small and close knit. They often like, like yeah we let everyone's family member in but I noticed in the very big ones. They were a bit more like no no no we really follow this rule they have supply it's very strict. So, yeah, that's a good question. Thank you. Great. Well from here, we are going to wrap up. If we could all take a moment to either push the button or show your video and acknowledge Professor Montero. Thank you very much for coming. And for all the questions is awesome really good questions. I hope you'll come back. We have one more lunch talk of this academic semester, Luke Schaefer will be with us on December 3 from the Ford School of Public Policy. And he is of course, one of the co faculty directors of poverty solutions. So, please come back and thank you again. Thank you. Okay.