 Well, thanks very much to the organizers for including this paper on the program. This is joint work with Marcelo Melendez at UNDP, Ignacio Flores, at the City University of New York, and Nicolás Urdaneta, who used to be here and is now a student at Duke University. So, in the interest of time, let me jump right into what we do and start by saying that this is simply a facts paper. We're aiming here at laying out basic facts about what is the relationship in Latin America between inequality and the organization of the productive sector, which is the sector where the income of many people is generated. This is a question that is actually very prominent in the current global debate. The context is one where inequality has grabbed more attention recently because of the increase of inequality in many advanced economies, most notably the Anglo-economies. And in that debate, the discussion of the role that firm size and especially super star firms, very large firms, and very large market concentration plays for the increase of inequality in those countries. So the basic question we want to lay out is what is that relationship here in a region where one inequality not only is high now, but has been very high for a very long time. This is one of the two more unequal regions in the world, where market concentration is also high, but where market concentration has a bit of a different flavor than in those economies. And in particular, it has also the feature that not only we have very large firms, but also we have a low, a bottom part of the firm size distribution with a plethora of tiny firms. And in a region where the huge fortunes perhaps of very large firms or owners also combined with a very high prevalence of poverty and very deep poverty aggravating the income inequality problem. This is just an illustration of those last two facts. Here to different, sorry, here to different ways to divide the income distribution in Latin America according to two different sources. The world inequality database being one that corrects the income distribution with different sources and basically both of those versions show that as compared to the European Union or the United States, we have a very thicker, lower tail and one that is longer is also reflected in these two other pictures where what we see is that income of the lowest and the top deciles or any part of the distribution are farther from the median income as compared to both the U.S. and the European Union and that that also that there is a more marked asymmetry reflected in the fact that in the bottom decile that problem is exacerbated. So what we do is very simple. We basically want to lay out two basic facts. We want to characterize first the firm size distribution, which is a classical question in the macro literature, but one that has been traditionally addressed in the context of firms that are registered and that can be followed from business statistics. But that is not a useful a useful way to characterize the firm size distribution when the question is one of inequality because the incomes of many people in Latin America are generated outside that traditional firm sector. So the first thing that we do in this paper is this characterization, I'm sorry, of the full firm size distribution trying to include all businesses, all productive units, which implies a mythological challenge that I will talk about next. We not only do that for Latin America, but compare to a wide, the widest arrange of benchmarks that we have been able to find to date and continue into that effort. And then the second thing that we do is characterize in a very broad manner what is the cross-correlation between that distribution of businesses and the distribution of income, of the incomes of workers, rather the earnings of workers that come from those, from that productive sector. As I said, there is a crucial mythological challenge and that is that we will not find many of the relevant productive units in the traditional business statistics. And that creates a problem because the bias in those traditional statistics is precisely against the smallest productive units, against people who are, who have what I will call unipersonal business units, people who are self-employed. And that is a very important part of the occupied population in Latin America and on the one hand, and of the business sector in Latin America. So what we do is we actually resort to a different type of information, one that is not a likely suspect when you're going to talk about businesses, but that it turns out to be useful. And that is information on individuals, either the labor force surveys, the household surveys, or other types of individual level surveys that give us a full picture of all the people who are occupied in the productive sector of the economy. It turns out that many of those statistics have information not only on the worker itself, himself or herself, but also on the productive unit where that worker works. And in particular, it has information on the sector and it has information on the size of the employer, measured in terms of other employees. In some of the countries that we have information for, we mix that with information from administrative records on firms. Of course, this also brings limitations. This is individual level data, and that also implies that we do not have much information, much more information on the employer itself. We don't have productivity, we don't have labor intensity, so what we can do with this is limited. And the other problem that you're going to see is the size categories tend to be very broad, and they become even broader when we try to compare different countries, given that the definitions of the size categories are different. And the third problem is that many of this individual level surveys take the labor force survey of any country are indeed representative in general of the population, of the labor force, and of the occupied population. They're also representative at sort of wide beams of those distributions, but they're not representative when you focus on a very thin bin of distribution. And that generates the problem that some of the people that worry us when we think about inequality, and in particular the very rich, the richest of the rich, that's a very selected population that is not necessarily captured by these types of statistics. And so what we try to do is we test the robustness of our results to the attempts that people have run in different countries to correct for that. Okay, so with all that introduction, I'll give you our two basic findings. So there's two things I want you to take away from this. One is, if we just think of the first question, the firm size distribution of countries in Latin America, the basic conclusion is that now that we have incorporated in what had been measured before from business statistics, the remnant of the productive sector, that is the people who were not employed in the firms captured in those statistics, the concentration in small businesses that has been known for a long time for all underdeveloped and developing economies becomes even more prominent. And so what we have is a truly massive concentration of the productive activity in very tiny businesses in Latin America, which has no parallel in any advanced economy. And when I say tiny businesses, again, I mean all productive activities, including people who run their own business in the form of self-employment. So let me illustrate that with a few numbers. So what I'll try to do here is present numbers from Latin America and then some comparators, again, with the very wide categories that we can look at. Here you have the percentage of total workers that fall in three different categories. One is they work for firms with ten or more employees. Notice that I'm here, this large firm category is not really large firms. We're talking about small businesses, those between, say, 10 and 15 employees, medium businesses, those above 50, and depending on the country, the limit may be 100 or 250, and then we do have also the very largest of businesses. Then we have a category of people working at businesses, firms of less than 10 employees, and then we have a third category of those who are self-employed. Let me just show you or underline what I find most striking. If you take any advanced economy you have here in the US, you have here countries in the European Union who are above the median income per capita, these are countries in the European Union who are below the median income per capita of that group. This is rich economies in Asia. In any of those, the vast majority of workers, the vast majority works at a firm with ten or more employees. In the US that is close to 80%, in the rich countries of Europe that is close to 70, as is in the rich countries of Asia, and even the poorer countries in the European Union have the characteristic that most of their workers work at something that would jump to people's mind when they think about a firm or an enterprise or a corporation. Then on the other extreme, those comparator countries, rich or richer economies, have very few people who are self-employed. In the richer groups of these comparators, that is less than 10%, but even in the others, it is not more than 15%. By contrast, on average in Latin America, that is above 30%. The other category of people working at what I'm going to call microenterprises here is also one where Latin America has much more presence than these other countries. That's the basic first fact. I think white striking in terms of its magnitude, we have always known that the firm's distribution is shifted to the left in less than about economies, but the numbers are really striking. Let me also say that within those microenterprises, it is the smallest of categories that's predominant in Latin America out of roughly a third of workers that work in less than 10 employee businesses. Most are in businesses of one to four employees. And then let me also point to now trying to go a bit finer into the categories of size for the countries where we are able to do that. In this case, we can only divide here Colombia, and I'll show a few more numbers in a second for other Latin American economies. One feature that we're also finding in this particular economy in Colombia is what people call the missing middle. So it's not only that you have a lot of people in the tiniest businesses, but also that in the rest, those with 10 or more employees, most people are concentrated at the top. Here in particular, at businesses of 100 and more workers, whereas only 12% are in businesses in the very wide range between 10 and 100 employees. Let me also point that there's some dispersion across countries in the region, but still the high predominance of self-employment and work at microenterprises is a consistent feature. The missing middle is less robust. I'm sorry, this had been corrected. These numbers for Mexico are incorrect and also for Peru, but the basic point is that the missing middle is less robust, but still in many of the countries where we do see, say, less than 15% of workers in that very wide range, again, of between 10 and 100 employees. That is to say, not only we have lots of people in microenterprises, but what we are most missing is people who can move or businesses that can grow to being small and medium-sized, and we do have a concentration at the very top. Let me also say that if we don't take these other comparators, but we, for instance, we think, well, what's going on, it's simply that these countries are sort of at an earlier stage of development, and they will catch up eventually. So what if the comparator is not any of these countries, but say the US, whenever it had the same level of income that Latin American countries have today. Let me just say that what that was the case, sorry, I'm going the wrong direction here. When that was the case, say, in 1940, the US had a level of GDP per capita equivalent to that of Peru today. It still had very little self-employment compared to that, sorry, to that particular comparator, 48%. In 1950, the US was similar to Costa Rica in terms of today's GDP, and it had lower self-employment, et cetera. So that's a different way of looking at the same problem. So that's fact number one. Then let me move to fact number two. Is there a correspondence between that and the low levels of income of, sorry, the high levels of inequality? And you'll see that my glitch right now has to do with the fact that there is a very high correlation and in particular, there is a very high correlation between having a low personal income, a low level of personal earnings, and being at a particular type of business. In particular, being either self-employed or working at a micro-enterprise. So this table here is showing you, again, the distribution of workers across different types of businesses. Each column adds up to 100, but it shows it at different points of the income distribution, and in particular for each of the five quintiles plus the top 5%. The basic fact I want to point at is perhaps best captured by the fact that if you take one or the people at the bottom quintile of the income distribution, you will find almost no one, 5%. Almost no one who works at a firm that has 10 or more employees. Almost everybody here is either self-employed, which is the most prevalent category in that part of the income distribution, or working at a micro-enterprise, or an entrepreneur that owns a micro-enterprise but has very low income. That is to say, we're talking here about, sorry, transformational entrepreneurs, sorry, not transformational entrepreneurs, sorry, five entrepreneurs, sorry, again. Public sector is not in this picture. It doesn't change that much, that picture. Not surprisingly, the public sector employees are more towards the top end, and they work at a very large, quote unquote, firm. Okay, thank you, Mariana, I had forgotten to say that. The other thing to notice is that as you move across the income distribution, that changes substantially. So once we are with people at the top, we actually have sort of the opposite characteristic, and this looks much more like, say, the United States where most people work at businesses of 10 or more employees, or own one of such businesses. The other thing I wanna point out is that in the US, there is much less of a gradient between firm size or firm business type, and the income distribution. Though there is some gradient, it is still the case that even for the poorest of workers or the lowest income workers, the vast majority here, almost 70% of them, work at firms that have 10 or more employees. I'm short on time, so the only other thing I wanna say is basically that this differential distribution is not simply driven by the fact that the productivity distribution is shifted to the left to start with. So it's not simply that we have less productive businesses, and thus those businesses end up being smaller because they don't have the ability to grow, but also that we have more disproportionate allocation of resources towards precisely those businesses where the productivity is lowest and where incomes are lowest. Let me just show you, rather than this overall, the composition, let me just signal that with some numbers, what you have here is for the first decile of the earnings distribution. These is the gap between the median earnings and the earnings at the first decile, and that gap can be as large as 12, so one person in this category here, the average person, has 12 times less income or earnings than the median person, and the fact that I was pointing at in my previous words is the fact that it is precisely the categories where this gap is largest, even within the first decile, even within the lowest productivity decile, it is this largest gap categories that grab the most concentration of workers. Just to leave a number in your heads, sorry, this here is say the composition of how much can we attribute the gap between the 50th percentile and the first decile, and the difference in that gap between land America and the US, how much can we attribute that to a larger gap in every business category that I showed you for the first decile in land America versus the US, and how much is this quote unquote misallocation in the sense of the region allocating precisely the most resources where the largest gap is, and the number is quite striking, 60% of that gap can be attributed to this disproportionate allocation of workers to those categories. So I'm not going to summarize, I basically gave you the two basic facts, let me just say that this is robust to using the world inequality database method for correcting incomes with the tax statements, especially to capture whatever happens at the top. The basic facts survive, even though the distributions are not identical, we basically this correlation, yeah, this is perhaps a good way to show it, that the basic correlation is sustained, the basic facts are sustained, and the basic correlation between the type of business and the income that people receive is sustained. Just a few words of discussion for closure. As I said at the beginning, this relationship between inequality and market concentration is a focus globally. For Latin America, this is certainly one story, there is much more information in the paper, for instance, showing that indeed market concentration is higher in any of these countries within the entrepreneurial sector, but the facts that we have just highlighted basically demand that in the discussion in Latin America, we do not forget about the one, the importance of poverty, to the, sorry, this is telling me that we're done. One, I was saying the importance of poverty for inequality in Latin America, so we can not simply focus on how we take the huge fortunes of these very rich people and translate it to other people because to start with, we have a lower level of income and much more concentration in poverty and the most vulnerable incomes, so that's fact number one. In fact number two, we may have high concentration within the traditional firms sector, the traditional, the other sector that we traditionally understand as the business sector, and that is something certainly that we need to pay attention to. Competition is crucial, we need, as I said, to populate that small and medium-sized enterprise sector, but we cannot forget the fact that those people in poverty and vulnerability are actually being left out of any sort of business sector association. So to start with anything we do to try to reduce inequality, of course the theme of these conference, by attending the business sector needs to take into account the fact that we actually, to start with, need that business sector to grow and to be able to embrace more people and produce incomes for more people rather than letting them out. Thank you.