 Is it better? Yes. Vietnam is home in many ways. Also very pleased to be here to talk about the well-developed report, 2013 on jobs and some of the implications of the thinking in that report. When so many people also around the room make it feel like being home, Ravi Campbell was one of our advisors. We had extensive discussions with Gary, Haroon Barat was one of our contributors, David Malone IDRC was one of our supporters, this was wider, so really feels like continuing a conversation that has been held in many different ways. But perhaps what I want to do is to take advantage of the fact that this is a more academic audience than the usual audience to which I've been presenting findings from the well-developed report to do a bit more on some conceptual issues that come about this report. You know, this report is very unusual for the World Bank to have a well-developed report on jobs. Over 36 years of existence of the series, this is the dean of the series in annual reports from big developmental organizations, only twice there was a report about jobs. Once was when the Uruguay round of the WTO was being concluded and there was a feeling that globalization was there, the issues of trade and labor needed to be addressed. This report, the second one on jobs came after the global crisis, which gave a sense of fragility or the importance of jobs, came after the Arab Spring, which perhaps was a wake-up call for many economists that jobs matter way more than just because of the earnings or the productivity, the entire social fabric hinges on them. Economists should have figured that out long before. Anyway, that's the way it happened. And so I want, these two events, the global crisis, the Arab Spring, pushed us to rethink jobs. Okay, if we think development through a job's lens, in which way will it be different? And I want to talk about the conceptual and methodological implications of doing that. We started by this simple realization, is that jobs, the word jobs has very different meanings for different people. It's not by chance that in the cover we put all these many words, because even in the words you see the nuances of what people think about when they think about jobs. Here you have people that work in different places, the young woman on the left is in a factory in Vietnam, these are people working in Poland, and perhaps when we think about jobs we tend to think about this young woman in a factory or these white-collar employees working in some back office processing. The reality is more complex. When you take the standard definitions of the ILO will use, the convention statisticians will use for what is being at work, you realize that overall in the world less than half of the people or roughly half of the people who are working are working for an employer. The employer employee is only for half of the people, much less so in developing countries, even less so in the poorest countries, and I'm just talking wage employment, I'm not talking formal wage employment. So when you look for instance at East Asia and the Pacific you see roughly thirds of people working as farmers, self-employed and wage employed. And so one of the first things that methodologically we thought is well we don't have a theory of jobs. We have very good labor economics, and labor economics is to a large extent organized about the notion of a labor market. Labor economics was developed in countries where most of the people at work were wage employees, and our way of thinking about jobs is very much shaped by that. So we think about labor supply and policies for labor supply, the skills, labor demand, and that's a business environment, and the matching of the two, and that's fine. And we want more wage employment, all that. But it still doesn't change the fact that for many of the people at work in developing countries wage employment is still a remote possibility. For many of the transitions out of poverty that we care about they don't happen by people moving from farmers to formal wage employment. They happen through other intermediate transitions, and we needed to understand that better because the policy implications could be very different. We also realize that even our thinking about the value of jobs is very much shaped by that. For instance, we tend to think that there is a problem about jobs when labor supply and demand don't match, and there is unemployment. The reality is that in many developing countries unemployment is a relatively irrelevant concept. And that's a point that I'm sure Gary will emphasize. Most of poor people in the work, and the work very hard, is just that their jobs are not productive enough to sustain their livelihoods. In many places you cannot have the luxury of being unemployed. It's not the case in middle income countries and other distortions can create huge unemployment especially among youth. But overall we tend to think that a problem of jobs is jobless, gross, unemployment, and it tends to be more complex than that. Another way in which we are tainted by the labor policy or labor market framework is when we think about other aspects of jobs like voice. We tend to think about voice as a problem of representation of workers and employers and collective bargaining. When you think about the ILO, the International Labor Organization as an organization, it's structured around this tripartite structure of employers, workers, and governments. But it turns out that for many people in developing countries at work the voice issues are not with an employer. Even the narrative of the Arab Spring is in a way tainted by this mental process because today Koshik was talking about how the mental frame for matters for institutions. The narrative we started with for the Arab Spring was one in which in Tunisia a young man sacrificed his life to protest the fact that he likes a job. The true story was different. This young man was a street vendor and he sacrificed his life to protest abuse by police and local authorities. That's a kind of problem of voice that we see in many developing countries. As Ravi, one of our advisors was Elabad, the head of the self-employed women's association. So we tried to think time and again to have the discipline to say labor economics is a wonderful tool, but it's one tool in our toolkit. How do we think about jobs in a way that is relevant for developing countries as a whole? The way we think also about jobs as economies and one of the reasons why we didn't have many world development reports on jobs is that growth is the thing. If we get growth, then jobs will follow in one way or another. And there is a lot of truth to that assessment. Here in these two figures you have countries, the little dots are countries. On the left hand side you have average wages in manufacturing in the vertical axis and income per capita in the horizontal axis. On the right hand side you have the proportion of the people at work who are covered by social security, formal social security. And you see that there is a good relationship. Interestingly it's not the same relationship. The shape is very clearly different. Interestingly also there's a lot of dispersion. So we're not trying to challenge the idea that growth is probably one of the most important things you can do for jobs. But what we saw is in which ways jobs and which jobs are there and are being created affects the pattern of growth. And what we did to put it in terms that work with an academic audience, they don't work with a broader policy audience. So you will not find this in the report itself. What we did is to think let's take jobs out of labor economics or out of labor economics exclusively into public economics. Let's think about the social value of a job. Before our report, before the World Development Report 2013, the previous report was on gender. And that report found that one of the biggest avenues for the transformation of the place of women in society was paid employment. The report before was about conflict. It found that one of the avenues for peace was the availability of jobs. So when you think about that, there is a social value or a spillover or an externality, if you put it in the jargon of modern public economics, that we may not be tapping if we only look at supply and demand and equalization. And so that pushed us to think that we should fully capture this multi-dimensional nature of jobs. The fact that they do many things at the same time. And to see if at any point in time in the development of a society, we are missing out on one of these big development payoffs. And therefore, what are the policies that need to be in place so that the players, the actors involved fully internalize the benefit these jobs have and take action accordingly. Now jobs do many things, but one way to caricature about jobs is to say that jobs are what we do, jobs are what we earn, and jobs are who we are. And these are the three pillars you see here. One is productivity, what we do. One is living standards, what we earn. One is social cohesion, who we are and how we interact in society, at the household level, at the community, at the political level. And we try to get expertise in each of these three areas. Again, Ravi was our main advisor on issues related to living standards. John Halty-Wonger on issues related to productivity. George Akerlof on issues related to social cohesion. And the idea was are there points in time in which some of these things are missing? And we did with these three columns because we could link data to each of three. On living standards, we have household surveys, the kind of household surveys we use for poverty analysis that can be used to look at jobs and living standards. On productivity, we have enterprise surveys, planned level surveys of different sorts. And on social cohesion, we have value surveys. So basically, there was an empirical foundation to work on each of these three. So let me just give you insights in each of the three areas on the payoffs that jobs can have that we may or may not be fully internalizing. And a good policymaker who acts in a kind of simple Pygovian environment will have to take into account. Let me start with living standards. First point made by this figure is most of the main source of earnings of houses in developing countries is labor. No matter how you put it in place, that's what you get. Of course, you have a few on the left that were transition economies at the time when transfers and pensions could matter. But in general, the more you move to the right, the more you move to developing countries. The main source of earnings at the household level is labor. And the main driver of transitions of the poverty is labor again. We looked at houses, places where you could have a panel and follow households over time to see what happens, what is different between two points in time when I look at a household that was poor in the first period and is not poor in the second period. The light blue is the labor income related effects. More people working in the household, people changing jobs, people earning more in the jobs. You have other events, marriage, separation of families, inheritance, lotteries and things like that. They do not count for much. Second thought I want to leave with you. So the implication of this thought is if you care about poverty, if in a way you say a job for a poor person for me matters more because of my social objective than a job for a non-poor person, then there is a value to the jobs in particular that are behind these transitions. Second point is, and this is an important methodological point, is that we were at a loss when it came to productivity. Because on the surface there is a wealth of studies by now on firm dynamics that more or less have replicated the kind of line of research that John Halty-Wonger and others have been pushing for in terms of job creation and job distraction and seems to be yielding very similar results to, in many ways, to the ones we find in industrial countries. But then we realized that the surveys on which we are doing this account for a tiny fraction of total employment in developing countries. We did a simple exercise. We took countries where we could identify from household surveys, we could reconstruct the size distribution of firms. People say, oh, I work for a small firm, I work on my own, I work with my cousin, I work in a company that is big. And you do that, you reconstruct the distribution of firms. And then you look at the distribution of firms from your planned level service. You do that for Germany, you do fine. You do it even for a country like Chile, which is an OECD country. And you completely miss a picture. So we reconstructed employment dynamics, trying to reweight what we knew about the true distribution of economic units by size using the information we had about observed transitions. And we came to the conclusion that even simple questions, like who creates more jobs, we don't know the answer. Everybody would say it's more than medium enterprises. No, it's not. Clearly, most jobs, even in middle income countries, are with micro enterprises. But we also came to the conclusion that there is something fundamentally different in the dynamics of economic units in developing countries. In industrial countries, you have small firms, which are mice, lots of them, but you have gazelles. The gazelles are the mice who grow and become something quick and big. Most employment creation in industrial countries come from these young firms, not necessarily small, young, essentially young. There are very few gazelles in developing countries. There are many mice. That's where you see a lot of employment. You see a lot of creation by these mice, a lot of distraction. It's just churning. It goes nowhere. Very few gazelles, many elephants, many big firms that do not go anywhere either, that don't have any. They are more productive, but they are not dynamic. So we see an entire agenda, which is to really reconstruct the dynamic of enterprises. Here you have the share of micro-enterprises in employment by country. And you see that in places like Ethiopia is 87%. Even in places like Chile, it's 39%. And we did some tentative work on social cohesion. And of course, there is more difficult because one can always say, well, you find correlations. You don't find causation. We find correlations that people who lack of a job tend to disengage from society. We find correlations that people who have a motivating job tend to engage more in society. I will not go into how we define and operationalize the social cohesion. But one could say this correlation, the causation. But we went for studies that really followed people over time. Here is engagement in the community by people depending on whether they were employed in two points in time, 2007. And you see that people who move from no job to job engage much more than people who move from job to no job. So we got all these ideas. And then we thought, OK, how do we put these all together? And I will try to put it all together in five minutes, which may be challenging. But please, please try to bear with me. And I will skip a very interesting survey we conducted with support from Norway and several countries trying to capture the social and private value of jobs. So how people see the value of a teacher or the value of a civil servant for themselves and for society. If the value was the same, all these dots, which are the value of different jobs, will be aligned. Clearly, they are not. Everybody would like to be a civil servant. Very few people think that civil servants are a great contributor to development. Everybody values teachers more for what they do for society than for what you earn as a teacher. But what we find is that a lot of the things we had been talking about in development were, in fact, about these other aspects of jobs on poverty reduction, on gender, on public finance and distortions from labor markets, on urbanization. Urbanization is people becoming more productive because they work together. So in each of the areas, you can always put names to the key authors in these different areas, Glazer, Melitz, whatever. And you can write the basic equations on many of these fields of development almost as a spillover from a job that can be quantified. And so we ask it ourselves. So can we, as a social planner, could do understanding in a country ways that there are these billovers? How much is a job in a city worse to society above what is worse to the person, and so on and so forth? And we find that it's possible, but it's probably impractical. There are all sorts of empirical studies, but it will be controversial. We'll take time. So what we did was just to say, OK, can we tell what the main challenges, the main potential and top benefits from jobs are in different kinds of countries? And we came with a typology of eight jobs challenges. Doesn't mean that each country has one job challenge. What we found, we did a classification with inputting in the reports that there will not be people discussing, oh, my country's A or rather B. But we found that basically every country has two or three big challenges. You have a gradient economy. You have urbanizing economies. You have aging societies. You have resource rich countries. And in each case, you see what is ways that you have a problem. Is it in the productivity pillar, in the living standards pillar, in the social cohesion pillar? And that brings the nature of the policy challenge to address. If you think, again, as in the environment, you say there is something I'm missing. I'm missing the productivity of cities in a place, and that's what I need to work for. Or I'm missing the fact that all people should remain productive in aging societies, and that's where my effort should go to. And we find two minutes. I will make it. And then we found that for each of the challenges, we had champions. We could identify countries that had done the right thing in terms of addressing those policies. Vietnam was one of our champions, by the way, as an agrarian country in the 1990s, what it did with its policies to redistribute land, take the state-owned enterprises out of the commercialization of agricultural products, put agricultural extension in place, was the right way to make jobs more productive, job agricultural jobs more productive. But for each of the challenges, we find a country that had done something remarkably good. For instance, Chile is our champion on resource-rich countries, on how to manage a bounty in a way that preserves incentives to work and produce and export and avoids the fluctuations that come with being a resource-rich country. And so we came to this idea that when it comes to policies for jobs in this framework, there are three layers. These are layer of fundamentals. That's what we always emphasize. They have to be in place, micro-economics stability, business environment, the rule of law, respectful rights, some human capital. But those are necessary conditions. They are not sufficient conditions. If you don't have that, you don't go anywhere. But if you have that, you may or may not go somewhere. Tunisia, again, had all that in place. So that was not the problem. Second layer, which we tend to emphasize when we look from labor economics, is labor policies. What we find is that the debate was much more heated than what the evidence could bring. We came with this idea that I will try to summarize in 20 seconds, because now I've reached my time, which is that it's not like more or less regulation is good or bad. We found that there is a range, we call the plateau, where you have some efficiency gains, some efficiency losses. You move a bit up the minimum wage. You get, I mean, more job security. It's very difficult to tell whether that makes a big difference. And then you have two big cliffs. You have the big cliff of, say, I'm working in South Asia now, Bangladesh, the lack of health and safety, collapse of Rana Plaza, that's a big cliff. And you have the big cliff of India, the industrial dispute site, the regulations that prevent firms from restructuring and adjustment their employment. Those are the cliffs. And so the message was stay on the cliff. But the big message was look at the top. Look at the priorities. Look at what Vietnam did in the 1990s, what Chile did with its copper, what Slovenia did by entering the European Union. The key policy for jobs may not be in the labor market. If you just use labor economics, we end up in this heated debate about whether the minimum wage should be 5% higher or 5% lower, that's probably a waste of time. Let's try to focus on the most important things. Finally, just to open up the appetite. And through, I finished 10 seconds. Each chapter finishes with a question where we use this framework to say, OK, so what is different if we apply this framework? And we try to go for questions like skills or jobs, which comes first, or growth strategies or job strategies. These are like four pages. They can be downloaded separately. They can be read as an economist article. If you are curious, please go to the website. Also, the entire database of the report can be downloaded from there. Thank you very much.