 So thank you, Christine, for organizing this conference and for inviting me. And thank you, C4, for hosting this. So the title of my talk now is Migration and Forest Transitions at Multiple Scales, Evidence from a Globalizing Latin America. And what I've done here is basically tried to cobble together examples from my own research over the last 15 or so years in Latin America, where there's some evidence directly or indirectly linking migration with forest transitions. My own research has to do largely with this topic, but also with other population and health transitions as they relate to environmental change. But so again, what I'm attempting to do here is really to focus in on the topic of this workshop, how are migration forest transitions linked? Giving due respect to what we don't know, some cases, to some degree, even hand-waving about potential linkages, given some of the things that we do know. There's no easy way to organize these rather disparate empirical studies that I've conducted around this theme. So I've kind of gone on a geographer. So I'm starting with the macro scale, more or less, and then going down to small scale to look at some of these links. There's three framing concepts that I'd like to introduce, integrating migration forest transitions. The first is that, despite all of the world's net population growth projected to occur in the world's poorest cities from here until 2100, the pace, magnitude, and geography of rural demographic transitions will have a disproportionate effect on global population size and distribution. So Cecilia showed the other day in her talk how she mentioned also that U.N. projections indicate that most population growth or all population growth will happen in urban areas. It's more than just that. It's that all of the net growth will occur in the world's poorest cities. But this is going to happen despite, all right, we have contact. This is going to unfold despite rapid urbanization. This will be largely, as Cecilia pointed out, to the majority of this change, of the source of this population growth will remain rural areas. So that is why rural demographic transitions are going to have such an outsized impact on future global population size and distribution. And then the second point I think that's important integrating migration forest transitions is this. Just to put it simply sort of in one phrase, how many people, eating what, produced where, will describe the vast majority of future forest change on the face of the earth? And point three is that point one will have a huge influence on point two. So what do I mean by this? This is, I think, a very illustrative histogram showing population growth by decade, starting with the, more or less, the beginning of the industrial revolution, and then projected out to 2100. And what we observe here is that where it's population growth in the world's wealthier nations, Europe, North America, peaked by the middle of last century, now it's declining. And that the bulk of world's total population and its growth is happening in developing regions. And one thing that I can't emphasize enough is that although population projections have basically been revised downward over the last several years as developing regions have progressed through the demographic transition more rapidly than demographers anticipated, the huge error bar around how many people we will be is precisely in the remote regions of developing countries, precisely where there's, guess what, forests. We don't really have data on these people. So one thing to take home from this is that yes, the projections have been revised downwards. Should they be? Should they be? Probably they should, but as much as they are, I don't know. And then the other thing is that if you understand how population growth works, kind of like the way stocks grow, geometrically rather than arithmetically, that an anomalous area in the demographic transition in terms of lagging in the demographic transition can have a huge outside impact over time in terms of the world's total population. And this is not just something that's just trivial and that certainly remote rural places, many of them do have remaining high fertility, and in some cases, whole countries still do, such as Niger. So why research tropical deforestation? I don't need to preach to the choir here, but approximately one-fifth of the world's large intact or old growth forests remain from approximately several thousand years ago. Why do I start with approximately 8,000 years ago with this graph, which would probably make many geographers blench? But it's just a rough, it's just to give you an idea. Why 8,000 years ago? What's happening then? It's a thing called the agricultural revolution. And if you're concerned about forests, then it's important to understand that with due respect to many different kinds of forest systems and human uses of them, by far, the hugest imprint of human occupation on the Earth's surface has been the conversion of forests to agriculture. So if we, the estimates here are very rough, but if we're just going to throw out some general numbers in terms of the Earth's surface, which is very much the domain of geographers, what's going on the Earth's surface. Maybe more or less 30% is in some sort of forest. Maybe 30% or so is producing food. And maybe somewhere between 35% to 40% of the remaining land is frozen, or desert, or mountainous, or wetlands, or somehow not easily or preferably used for agricultural production. And well under 1% is in human infrastructure and urban areas. So let's take a step back here for a second and say, well, then around a third more or less is in production for food stuffs. Well, over three quarters of that is in livestock or in crops to feed livestock. And so that is really the big human imprint on Earth's surface and certainly on forests. So why do we care about demographic transition? How does this relate? One thing that I can't emphasize enough is that both in the popular media and knowledge, but also among demographers, what's the big deal in demography right now? It's urbanization. This is what you hear about. It's all about urbanization. And in fact, and we are, as I mentioned earlier, revising downwards our population estimates because the pace of urbanization has exceeded what we anticipated. That's the developing world. And the developed world was the big story, aging. So we're shrinking our population. So that's the big deal. Yeah, except these transitions are happening and unfolding really at different rates in different places. This is some work I did with WWF, looking at population change in and around their core conservation areas. And I made this index of relative progression along demographic transition. Red is basically scarcely begun the demographic transition. They very few of the variables related to demographic development have occurred yet. And then so you have places such as the vast Congo Basin, second largest tropical forest in the world, where demographic transition has scarcely begun, if at all. And then nearby in South Africa, if you've been to Joburg or Cape Town, the demographic transition is similar to what it is in North America or Europe. So this is quite interesting, I think, that we're right now living in a time that is unique, demographically. Never happened before. And very likely, it will never happen again. And we're in it right now. And that is that we have these two worlds. We have places where the demographic transition, as we define it, has largely been completed, other places where it's scarcely begun. We have that huge transition across space happening right now. And so what took a couple hundred years for North America and Europe to go through in this transition of high mortality and high fertility to eventually low mortality and low fertility is happening in developing countries in some cases in just decades. Where it's happening, where it's not, is going to be a huge predictor of future forest transitions, but increasingly in indirect ways. And I'll get to that in a second. So Latin America, where I've done most of my work, and which is going to be the topic of some of the empirical research here, is a place that I think also is surprising to some people that know Latin America, that it's largely urban. It's 75%, 80% urban, like North America and Europe. You talk about the Southern Cone, Chile, Uruguay, Argentina. They're even, in some ways, a bit further ahead, in some cases, the demographic transition in the United States. But there's still remain pockets where it's very early in the transition, high fertility, high, particularly infant mortality. And guess what, these are places, not surprisingly, it's not coincidental, near high conservation priority for us. So what I did, excuse me, with this talk, is to basically ask questions for probing questions. These are not wealth that I did this last night, but sort of linking to some of the research I've done to these questions of migration and forest transitions. What do we know, what don't we know? Just a first attempt to link, again, some of these questions with some of my own research. So my question one is, how and where do dynamic migration landscapes, both rural destination and origin areas, relate to demographic transitions? And so in a sense, we're talking about, because these ultimately will affect also forest change. So here I'm talking about this link of demographic change to another kind of demographic change to ultimately forest change. And so I'm going to report just briefly on some work I've been doing with colleagues at the Carolina Population Center since I was a graduate school there in the 90s on forest change and land use change in the Ecuadorian Amazon. And just for the purpose of this question, I'm focusing on fertility change. And during the 1990s, we found that the total fertility rate, which is the average number of children per woman during her reproductive years, her life cycle, fell from around 7 to 4 during the 1990s. Now again, think about in the US, how long did it take for this decline to happen? Well, the fertility rate was around 6. Then it went up to 7, particularly with Western expansion. Land availability increased fertility in the US, just like we see in these areas now. And then over about 150 years, 150 years it got to 4, and now it's around 2. So this is a change that happened again in North America, 100 and 150 years here, 10 years. In some ways, this is sort of a classic story of the oil discovery and then infrastructure development, transportation, infrastructure, and then spontaneous settlement, which we've seen in other places throughout Latin America. And we've published quite a bit of research on this topic. But again, here I'm focusing on fertility transition. So this transition, which took so long to happen in North America and Europe, and was more rather homogenous across space in the developing world, as I was trying to show you with some of the maps earlier from Africa and North America, can happen very quickly in a very close proximity. So here we have La Guardia, which is one of these rainforest cities, which has sprouted up over the last few decades. In the Amazon, where the total fertility rate now is well below 4. If you literally cross the river, the fertility rate is over 7. So this never happened before in North America and in Europe, where you had places right next door to each other where the fertility rate was so different. And what's the connection with forest transitions? Well, the causal arrow in some ways goes both ways. And I just want you to be aware of this. There's not a lot of research on this, but I think it's fairly compelling what is out there that availability of land for households that for whom land is the primary resource does have an impact on how many children you have. So amount of land with land increase, it was very positively associated with increased number of children born, and whether a child was born or not, which is a regression. So moving on to a similar topic of origin and destination regions and how these processes can affect fertility. This is some work I did with Jason Davis, who's now at the Carolina Population Center, on remittances and the effect of remittances in Guatemala and some Highland communities where we researched this on land use and fertility and consumption. And so we asked, is there a relation between a rise in remittances among households and a reduction in fertility? We also wanted to know if there was a change in land use. We saw no notable change in land use, just a couple of case studies around a sample of 85, randomly sampled households in two communities. And we didn't see a notable reduction in fertility either yet. But interestingly, something we did find that emerged from this research that surprised us was that one of the first things that households invested in was education for their children. And a lot of the research we had seen in South America showed that, for example, Brad Jokers' work in Ecuador, that people were investing, kind of putting their capital almost in their household construction. They don't, at least until recently, don't have bank accounts. So they're kind of investing in their household and in their land. We didn't see that. Here we saw people investing in their children. So here, of course, education is the strongest predictor, at least underlying predictor, fertility decline. So over time, we do think that if this is true in other cases, investing in children's education will, over time, reduce fertility, which can or might not. But in a direct way, all other things being equal might reduce pressures on the forest. Second question, how do coupled urban and nutrition transitions in one place affect forest transitions elsewhere? And why there? So we have these two worlds that I talked about demographically, the urban world and then the world that's more rural. And remember, I think Cecilia also mentioned that since 2008 or 2009, the world became 30% urban. And but that belies, again, this real dichotomy where we have places like North America and Europe that's 80% urban and places like rural Africa and still many parts of Latin America also that are well below 50% urban. And these two worlds are associated with two kinds of pathways when we're trying to link migration to forest change. Pathway one, remaining high population growth, more likely to be rural, rural moves, low technology, low yields, poverty, subsistence, and people dying of infectious disease. And pathway two, declining population growth or already stable urbanization or already urban. With that incremental increase in disposal income, increased meat consumption and animal protein consumption and produced not necessarily where they are anymore, usually not, but through high yield corporate agriculture with high technology. And with this increased diet with sugar fat, animal protein and people dying of degenerative diseases. So this is work that Mitch Aidled on our CNH project that he presented some work on. So you basically saw another version of this map before, significant areas of deforestation and significant areas of reforestation in Latin America during the first decade of 2000. And Mitch already talked a little bit of this trend, so I'll discuss how this may or may not be related to migration. As Mitch and others have shown, we have rural migration or rather population growth in rural areas in Latin America by the early 1990s, late 1980s, actually start to decline, whereas urban population has sort of continued in its linear ascent. And some places even lost rural, lost population growth during the 1990s. Much of Latin America now is a declining rural population. And this is something that many of you working certainly in Brazil understand. So we, this is a conference paper that we did in Arizona, I think, for the land change meeting. We looked at a relationship between population change and then we saw a level of 16,000 of them. For Latin America, a woody vegetation change, we found another, R squared of 0.000. This is about as low a correlation as you can find. So then we, do we conclude then that there's no relationship between population change and forest transitions? Well, if we just look at that data, that scale, we would. But things are more complex. We have in Southern, or Central, South America, Southern Brazil and Northern Argentina, Paraguay, the growth of soybeans and export agriculture. We have in Northern Mexico and also here in Northeast Brazil increased precipitation and some forestry growth based on that. And then we still have some, oops, excuse me, some remaining areas such as in Mesoamerica and also along the slopes of the Andes and the Amazon where there's still this sort of pathway one largely poverty related and still high population growth where rural migration subsistence based forest change. And so one of the things that we found was that pig and poultry production globally and soybean production are both soaring. And we wondered if these things are related to forest change in South America. So we did a regression where we looked at various variables related to forest change in these 16,000 municipalities. And we also looked at the second level effects potentially of state and biome. And we found that beef production was the largest predictor beef production at the national level of deforestation and mitigating reforestation. The biggest predictor in land use was soy production. So if we break this down by the, where these products are being shipped, we see that Russia is purchasing a lot of this US and then followed by these countries and the production is very large in Brazil followed by Argentina, Mexico, Colombia, and Uruguay. In terms of beef, the majority of the beef is consumed within Latin America. So the rising affluence largely in urban areas in Latin America within that region is related to, apparently, seem to be related to increasing meat and animal protein consumption. In terms of soy, much of it is going to China and the Netherlands, which is probably a proxy for larger EU distribution in other European nations, produced largely in Argentina, Brazil, followed by Paraguay, Bolivia, and Uruguay. We've had a lot of soy here while I've been visiting Indonesia. What percent of soy do you think is eaten directly by humans? That's produced? One percent, one percent. Now, and this is not scientific research. This is the industry reporting this, actually, one percent. And so the rest of it is largely for animal field, also for oils. Then the question is, is the land change within Latin America, that we see internally, is unrelated, at least at the municipal level, with population change, is it really unrelated to population change? Well, this data is showing us that it might be related to another kind of population change, something that's going on in another place, another time, not just sheer numbers, but what that population's doing. So, virtually all of the soy that's shipped to China is fed to pigs in poultry. And we know that China has added over 100 million people to urban areas over the last couple of decades. And with that has been a sharp uptick in the consumption of meat products, particularly pigs and chicken, not so much beef, although increasingly, yes, beef is becoming important as well. So, it's a migration process, rural to urban, which facilitates an increase in disposable income vis-a-vis urban-based employment, which changes diets and demand for foodstuffs. And in a sense, what we see is that an exporting of the agricultural frontier to Latin America from China. So, population, including migration trends, are very much related to forest transitions, but increasingly, with greater distance in space and time between those population and related drivers and land change outcomes. So, scale is really important, as an analytical lens, clearly. This is a classic example of the ecological fallacy, where we ask, is population related to change? Seems like we have a pretty good study to examine that question. Came out with nothing, but if we look at the larger scale, both smaller scale and larger scale, we see how population does relate to the land change. So, another way to look at this is through the individual or the household, as a consumer, producer, and as a migrant. So, this leads into a third question. How do household decisions simultaneously and or sequentially, approximately, and distally in space and time, how might they affect forest transitions? Why, and when, and why then? So, one way to look at these issues is that a household or an individual in any given place is living within a context of macroscale demographic, political, economic, social, and ecological dynamics, and they make choices based on these pressures in these contexts, in these opportunities. And there'll be variation in this. A response for a rural household can be to seek off farm labor, can be to decide to have more or fewer children, it could be to manage land differently, or it could be to migrate. And these different options are related to different outcomes, and different, particularly different outcomes related to forest. So, if a choice is to migrate, one can migrate to different areas, internationally, urban destination, when a household decides to migrate to a rural destination, or turn to the top of the chart, there's a greater likelihood that you're gonna have this sort of pathway one deforestation process. This leads me to question four. So, how do different migration patterns and stages affect forest frontiers, both internal and external? So, I think it's important to distinguish between these two different kinds of frontiers. This pattern of the poverty, high population, subsistence agriculture, is often, the early stages of this is often at the external frontier. As these frontiers develop, then the remaining forest is actually on people's farms. Or, if it's a corporation, similarly the remaining forest within those farms. And so, these are really different contexts and very different questions you have to ask to understand forest change. If we're talking about the internal frontier, where people already are farming, where people already are impacting forests, then as you've seen many, many, many studies, for example, the Amazon researching what predicts deforestation, where people, researchers will go in and survey hundreds of households and they'll do a regression analysis of the household characteristics related to the farming, right? And how much forest they have left. And that's fine, and you can get a high predictive value for that for what predicts how much forest is left remaining on farms, the internal frontier, but you still don't get at the vast unoccupied forest still. To get at that question, you have to understand who's not there. Why are they not who could be there in the future, whether it's people or corporations? And that's a very different research question and research method to adequately address that. So a fifth question, how where and under what conditions do remittances affect origin forest scapes? This is work that Jason Davis and I did using the LAMP data, Latin American Migration Project by, started by Doug Massey, a demographer at Princeton, where we have several hundred households in most countries for Latin America, Central America rather, and there's very detailed information including on land use. And so we wanted to ask using this data how might remittances be related to land use and other outcomes. And we found that the international migration and the sending of remittances correlated with an increase in land and pasture purchases, but did not relate to changes in agricultural intensification methods. This is a small data set, but we're unaware of many other studies doing this. And if this sort of result is replicated, I think it has potentially profound implications for forest transitions. Of course, pasture has a much greater impact on forest than other kinds of land uses given the return in yields. Sixth question, where and why does scale, space and place matter? So this is an image from several years ago of Guatemala at the municipal level. And this is land in forest. And some of you may be familiar with the Mayarain forest. There's lots of forest there, very little forest along these rich volcanic slopes. And southern Guatemala. And so prima facie, you look at the situation of Guatemala and you say, well, this is a situation of land for people, for people without land, there's a map of population density. So you see a lot of migration from high population denseries to low population denseries, low forest cover to high forest cover. And we looked at this topic both using multi-level regression with the second level being the departamento, first level being the municipality. We also looked, we did globally weighted regressions, spatial regressions. And we basically found that we had both the pathway one and the pathway two process where we'd expect it happening. But do local dynamics then a forest change reflect these national continental patterns. So this is work I've been doing for the last 15-ish years in the Maya Biosphere Reserve, particularly in this core conservation area known as the Sierra de la Candon National Park. And the La Candon, both the richest biodiversity in the Maya Biosphere Reserve, receives quite a bit of rainfall and also has quite a bit of relief. But one thing that I wanna point out here, this relate to this seventh research question is, where are roads built? You know, it's something that we sort of, there is some research, there's particularly economists have done some research on this. But where roads are built are not random. So it's an important question to ask because without roads, it's very difficult that there's penetration in the forest in a meaningful way for real agricultural driven forest conversion. So this is the road, the Narajo Road in 1987. This is an aerial photograph. You can see in an 87 a French oil company built this road in actually the mid to early 1980s. You can see some farms starting to spread up along the road. And if you fix your gaze on this bend in the road down here, that's here. And there's a composite image of deforestation during the up through the 90s and then after 1990 up until 2000. And you can see that a lot of the forest was cleared during this time. About 10% of the park was deforested. And you can see there's sort of a standard pattern of the maize based corn, switten farming and sort of each one of these blotches kind of represents a maize plot farmed by a household. There's anomalous blotch here, which is also one household, but it's a different land use. What's going on here? This is cattle. So scale matters and what a household is doing with the land matters. We wondered what would happen. So we have this pattern of deforestation. This is supposedly a corn conservation area in the Maya Bias Reserve. And then you have this, but much of it is now deforested or compromised. And so we really have this area here that remains in large intact tracts of forest. We wondered, well, you know, are people going to still keep moving further and further into the forest where now it's quite a long walk to the roads? Is it worth it to people? What's going to happen? So this is 2005 to 2007. And sure enough, we did see people moving even further into the forest. So roads, in a sense, collapse space. I'm at UCSB Geographer and we have the famous geographer Waldo Tobler, who has his first love geography that places closer to others are more similar than those that are farther apart. It seems kind of obvious. You know, in a real sort of obvious way, of course, this makes sense. But any of us who study these sort of dynamics understand that place is important too. And roads, the impact of a road really changes our dimensionality of what space means. I mean, all of a sudden it opens up these frontiers. So to understand, basically, the pace and magnitude location of deforestation in this park for my dissertation, I started field work here in the mid-90s, interviewed several hundred households and household heads and wives. And then a doctoral student of mine, Laurel Suter, followed up in 2009 with her dissertation and followed up in the same households. And we're doing a third survey now next summer. And we did the surveys in Spanish and about a quarter of the households speak Maya, almost all of them Kekchi Maya and we saw we did those in Kekchi. And this is the average farm in 1998. And you can see we're tackling internal frontier, lots of forest in the internal frontier. So there's still sort of, from a forestry perspective, these plots, these farm plots are still of interest. By 2009, it's only 10 years, you see that most of the forest has been eliminated and you have a very extensive swidden. Which again, is indicative of what sort of land use will have very potentially different impacts on the forest. Eighth question, how, when and where do demographic, political, economic, and ecological factors help contextualize or explain some of these forest transitions? So yes, demographic factors are important on the frontier. The fertility rate actually up through the nineties was about eight, which is as high as any place you can find on the planet, even historically. And larger households here were associated with more forest clearing. And this has been also replicated in other similar places in the Amazon. The transition here as an aside in when the fertility rate is so high and you wanna look at fertility transitions, you look at an eight of a TFR and you say, gosh, well, it hasn't begun, right? This is another way of getting at this. That if there's high desired family size, if the desired family size remains also, for example, eight, that's really at the earliest stage of the demographic transition, which we have found in places not here, but like Madagascar, Niger, rural places. Yes, the desired fertility rate is still seven or eight. But in the nineties, in the Lakando National Park in Guatemala, the desired fertility rate for a number of children was three among women and about four to five among men. So this has real implications for what, A, more equitable decision-making among men and women in a household, the impact that can have on family size. And also the potential impact of basically providing education and health access to people in remote areas. Political economic factors were also important, proximity to markets and capital matter and has different impacts on the forest, as you might expect. One perverse example of political economic factors was the issue of land titling, which from the Ag Econ literature, many, one might expect that it would lead to more conservative land use. Instead, it led to agricultural expansion because with land title farmers, this is on the edge of the park where they're given land title in some instances, they often use that land title as leverage for loans and with the loans they purchase cattle. Ecological factors are important. Low, high population growth, but low population density leads to expensive farming. There's cultural differences as well and here temporal scale is important as well because if you had just gone here in the 90s and you would ask farmers about their land use, you'd see that the Maya farmers were clearing more forest because they had a more expensive swidden. But if you go back into the 2000s, you see that now the Ladino, the non-indigenous farmers, have cleared more forest because they're more likely to adopt cattle. This leads me to a ninth question. Are frontier migrants more likely to move to another frontier? You know, are these already sort of, in a sense, pre-selected for? So we asked people in the follow-up in 2009, in the household, how many people have moved out and where have they gone? And amazingly, just in that 10-year period, only about 63% remain in the same town. This is a very dynamic frontier. And this gets back to my household diagram, you know, in which, sort of borrowing from Kingsley Davis and the multi-phasic responses, these are very dynamic people, very dynamic households. They might be forest once and then several years later, they might be clearing forest in a new frontier. And in fact, nearly 10%, we're doing precisely that. You think, well, at 10%, that's not very many and maybe not but it's actually many times more people percentage-wise that are doing a rural-rural move than we would see globally or elsewhere in Guatemala. So how important are rural migrants and rural originaries to forest transitions? And who are these migrants? Where do they come from? Why and why do they leave? So this is sort of a follow-up to this research. You know what? I could describe deforestation in the park based on several hundred household surveys. Come up with an R-square in terms of predicting forest cover, in terms of predicting different land use outcomes, explain anywhere between two-thirds to three-quarters of the variation among households in land use, okay, and thus in deforestation, forest cover. But of course, there's a hundred percent correlation between people being there at all in the first place and forest clearing. So roads are a, in a sense, in this case, a necessary but insufficient condition, right? You know, why do people leave originaries to go to this remote frontier that's malaria-infested? There's no drinking water, there's no infrastructure, the kids can't go to school, they don't go to school. There's no healthcare, you know, why would you do that? And so this is an important point, most don't. So, you know, this is sort of a needle in haystack effect. You have a tiny, tiny fraction of rural migrants that have a really, really outsized impact per capita on the forest. And so, you know, from a political economic institutional standpoint, you can say, well, it's obvious why people are leaving, right? I mean, you have, you know, the most skewed land distribution in Latin America, Brazil's right up there. For example, you have about 2% of the population controlling about three-quarters of the land in a country about the size of Ohio. But unlike Ohio, which has Cincinnati and Cleveland, right, and all these cities, Guatemala remains about 50% rural. Almost all of those rural inhabitants are indigenous, and certainly almost all the rural inhabitants are farmers. Well, you know, that's, in a sense, a real population-slash-land pressure that's building up. And on top of that, you had an extended civil war that displaced over a million people during the, it just ended officially in the mid-1990s. I mean, so you have these sort of institutional macro-structural conditions for our migration. But nonetheless, most people don't migrate. They don't, despite this, in a tiny, I mean, we're talking just several thousand households that moved to the Maya Bazaar Reserve and have that impact. So who are these people? And where do they come from? And so, using the same sort of household model, we're asking this question based on all these different factors, why, and under what conditions do people decide to migrate, and when they do so, to another rural destination, a remote rural destination. So to ask this question, triangulating three-duty services, including my own, went to the 30-ish, I think it was 28, actually, municipalities of highest outmigration to the Maya Bazaar Reserve and found that about 10% of men and women during the 1990s had outmigrated at all. And now, again, I'm selecting for areas of highest outmigration to the Maya Bazaar Reserve, and even in those places, again, a highest outmigration, only about 10% outmigrated during that decade, and of those 10%, only about a third outmigrated in search of other land. So we can surmise from this that, again, in the 28 municipalities of highest outmigration to the Maya Bazaar Reserve, less than 5% moved to the Maya Bazaar Reserve. I mean, this is very, very few people. So I think that in terms of the importance of, for forest policy, to the extent that they're, yes, the world is shifting toward the sort of pathway two, and we'll get to that in a minute, but to the extent that there's still a remaining pathway one, who these people are, where they come from, their individual characteristics, and their sort of macro communities, structural conditions are, I think, quite important. You can have a lot of potential bang for your buck in investing, in targeting those locations. So why did some people migrate to the Maya Bazaar Reserve? There's some common denominators, as you might imagine. These are in many ways rural subsistence farmers that are eking out a living in under very difficult conditions, ecological conditions. They are more likely to be located on steep slopes, areas prone to erosion or flooding. They tend to be in areas where there's skewed land distribution. They tend to be larger families, more likely indigenous. Quite a few came from places where there has been expansion in corporate landholdings, and this is an interesting slide that says, any of you speak Spanish, have another one, and you have this sign that says, but don't park here. These manicured lawns, and as far as I can see in the Pacific Slope of Guatemala, these really rich, dark, basaltic volcanic soils, it's just in sugarcane, pretty much to the horizon. So there's some structural reasons for this, but are origin places of these migrants different? And why, and what are policy implications of that? So we've talked about some characteristics of the people and the places, but can these still unfold differently in different places? Now I'm putting on my place geography hat. So lots of different case studies. I'll just give you two, because I have no idea how much time I have. Yeah, okay. So here are two municipalities of high out migration to the Maya Biosphere Reserve during the 80s and 90s, no longer. We have Morales, and we have Nueva Concepción. And Morales, this municipality grew more quickly than any other during the 1950s in Guatemala. The population doubled. Why? Because United Fruit had expanded land holdings there. We're talking about post World War II consumption, baby boom and consumption boom in Europe and North America. People want to eat bananas, and you have now technology that enables that with refrigerated big shipping containers. And so United Fruit gobbled up a lot of great land in Morales, and then you have a doubling of the population because there's a great demand for labor. People flood in. Quite a few, by the way, from this area in and near Nueva Concepción, the Pacific Coast. But within a decade, all of a sudden the migration pattern reversed. You had this area of highest immigration in the whole country to very rapid out migration, many to pretend, initially to Southern pretend, and then to the Maya Biosphere Reserve. What happened? Why is that? Well, as many of you know who work in rural areas, these kinds of jobs on plantations are often seasonal and low-paying. And so, farmers who relied on these jobs still needed to be subsistence farmers to maintain their families. And so with United Fruit gobbling up the best land, there just simply wasn't enough land for survival. The jobs were insufficient. So when a policy implication of this are jobs the answer, you provide apparently sufficient jobs. But there's a quality to this too. It's not just quantity, what kind of jobs? And so even if we double the number of jobs, you'd still have out migration here because you wouldn't have those kinds of jobs. People wouldn't have land to compensate to at least grow some subsistence corn. So another popular policy prescription for the developing world to alleviate inequalities and particularly among rural farmers is to redistribute land. And Nuevo Concepción was one of two municipalities in Guatemala that during the 1950s, it did exactly that, carrying out the dream of Jacob Waterbentz, the president who was assassinated. And so Nuevo Concepción in 1952, some large land holdings were redistributed among several thousand households. Each household was given 25 hectares. And in the intervening period from 1952 through the 1990s, there was basically this land redistribution and the rules governing it were quite strictly adhered to and there was no further land consolidation as is very common in other places. And yet from 1952 to the mid-90s, the average farm size splintered from around mid-20s, 25 hectares or so to only around three hectares. So how did that be? What happened? It's no land consolidation. It's a really rare case where you can isolate that. It's because the fertility rate remained high and people are married. Remember also in this demographic transition, one of the transitions that people don't think about often which is really critical is that marriage age gets older as this transition occurs. So not only do you go from having seven children to two children, but you go from starting to have your family at age 30, 35 instead of age 14, 15, 16. So you literally have even with the same fertility rate, double the number of people living at any given time when the age of marriage is 15, 16, 17. And so within just a few decades, you have basically a splintering of land from around 25 hectares per household to three, you know, really isolating, high fertility as the cause. So was land distribution the answer? It was insufficient. Land distribution without taking into account these demographic processes was insufficient. And these are just two reasons why place matters. So is this a sunrise or sunset over Latin America's forests? As I came to this stop sign Southern Guatemala, which says here, stop, here says two ways this way and then one way this way, I'm unsure. But I'm sure of one thing and that is that migration transitions will continue to be intimately related to forest transitions, that there's still a pathway one that we need to be aware of. It'll be increasingly, it'll be very important in terms of understanding future transitions both directly how many people are still living this pathway one, still in that world and having a direct impact in space and time. And what will be the timing, pace, magnitude of that population becoming urban? Is it inexorable? Demographers seem to think so. Is it really, is there some spatial variation in that and what is the implication of that for forests? You know, how soon these or if at all populations live in urban areas and start having impacts through the pathway two. That's it. I look forward to going to the garden so anyone who wants to go.