 Thank you for being here. I am Isaac Moussland from University of Maryland, Central, and this is a paper with a fellow at UNOWEDER in Malo, Kelly. We're looking at where the export promotion agencies can stem the industrialization process in Sub-Saharan Africa. And what motivates us is that we have some papers like this one that says that limited export opportunities causes manufacturing plants to scale down their production and employment. And Stein also says that if you need to raise living standards in Africa, in Africa, especially, you need some form of industrialization. But we really agree that investment in the industry sector needs a lot of cash or a lot of financing. That means we need to find some way or some form of financial development to see whether we can finance the development of industry. And there is one thing that has the potential to do that and countries with weak financial markets. It looks like it is EPA, which is export a promotion agency. And that kind of agency, which is EPA, it implements a wide range of policies to sustain both manufacturing and export. Everywhere around the world. And Britain, for instance, has implemented a search agency about 100 years ago. But still now, there are plenty of countries that have not implemented such policies that can help them to boost manufacturing and at the same time boost the export. And what did we do? In Africa, actually, there are 28 countries, or NSSA. There are 28 countries where they have implemented an EPA. And there are 20 countries where they have not implemented an EPA. And EPA, it deals with lack of capital. That means they allow firms to get access to external market. There is market tensions, especially when lenders give out shorter maturity loan, a smaller loan amount with widespread. When you have an EPA, that EPA serves as an intermediary between the traditional capital market and the firms so that they help the firms to get access to the private credit market. And they also give insurance to exporters. Imagine that you have something you send to another country, and you don't know the buyer. And that buyer can default, may not pay you. Then if you have insurance, that insurance will help you to get paid back. Then that means you deal with country risk, with credit risk, and a lot of counterparty risk. Just having the insurance. And because you have insurance also, the banker would now relax some bonds standards. They become more lenient because you have insurance and you have someone somewhere that would buy and pay back. And also, let's say you send something to a foreign country. And you are in a competitive market. You cannot force your buyer to pay you in advance. What if the buyer says that I don't want it? It's at the dock. You have a trouble. Then if you have an EPA in your country, that EPA will help you deal with the insurance provided to you. They help you deal with a lot of uncertainties. Also, if there is a problem between the buyer and the seller, then you don't know the legal system of your importer. But once you have the EPA, they have a wide range of services. Legal, marketing, finance, a wide range of services that they provide that sustain both manufacturing and export. And also, they allow firm to export directly and indirectly. They give some brochures, some markets. They get around some information asymmetries. Because if we are private competitors, why would I tell you about business opportunities in another country if I want to sell a product that you want to sell to? And once we have an EPA that will make the information available to everyone or every potential and actual exporter or manufacturer, that we get around the most problem of lack of information and market failures. And also, they allow firm to have access to finance their operation, like working capital and trade finance and trade credit. And what is the punchline of our study is that we see that three years after you have an EPA, the industry value added was by about 3.24%. And seven years, it was by 14.64%, which I believe is very substantial. And the trade to GDP ratio was by 70% within three years after having an EPA. And when you have an EPA that works with a country that has an export processing zone and the country has an EPA that reduces cost terms and trade regulatory for exporters and manufacturers. And EPA also reduces the amount of collateral that you need to access financial markets. And some countries firms need as high as 80, 90% of the value of the loan in terms of collateral. That's really an inhibitor to a credit market. And they also arrange a facilitated credit for firms, especially for manufacturing firms. And what do we want to answer is whether EPA has a policy, whether it has a real impact or whether the policy has been successful in Africa. And we also look whether that agency allows firms to have access to the private credit market. And we have, as we said, we have 28 countries in Africa with EPA. And Gabon, for instance, has the promo Gabon, the EPA since 1964. And the latest one might be the TISA in South Africa as of November 2009. That means we have a lot of variations in terms of date when EPAs have been implemented. And there are some early trade reformers like Ghana, Bukina, Faso, and Nigeria. These countries have implemented the EPAs since the 1970s, but we have 20 laggards in the region. And what do we do? We collect some data on EPA or TPO. That's how they call them at the database of the International Trade Center. There they give you information about who is heading the agency. You can have the email, the phone number, et cetera. And we contact them. Some do reply, some don't. But we have some specific information when needed. And also there is a paper by Leleman. That paper gives a lot of information also about that structure. And we take our main manufacturing variables from the WDI from the 1960 to 2011. And we take some other variables from the enterprise service database. And there is one thing. Spence has a growth report that has been written for the World Bank in 2008. They talk a little bit about this kind of agency, EPA. And the take is that these agencies, their impact cannot be really evaluated because of lack of counterfactual. And so we pick up on this, we apply a difference and differences technique. What do we assume is that manufacturing output or manufacturing outcomes would have been the same, would travel the same path through time between EPA and non-EPA country. And what makes the whole difference is the EPA event that has occurred in one country and hasn't occurred in other one. And because of the assumption of parallel trend, now we can apply a difference and differences methodology to generate the counterfactual so that we can tell what it would have been. And those countries where there is an EPA, hadn't these countries implemented that agency. And what we do, we do a matching, but the matching is not necessary, but it gives us more confidence. We look at initial industry value added, trade to GDP and manufacturing value added between a country where an EPA has been implemented and we compare it with the average, with that of another country where EPA has not been implemented and if it is not significant at the outset, we match these two countries. And we do that and there are some, you may agree with me that as we go far back in time, 1960s, then the data is not that good, that means that allows us to drop eight countries and to have a perfect match with 20 countries where we have EPA and 20 countries where we don't. And then after we do the matching based on our T-test, we are now able to do our evaluation from two years, five years, seven and ten years before and after that agency has been implemented. But we have that gentleman who said that it's not really important to have a perfect match or that the two sets of control and treatment groups, the two groups don't need to have a similar characteristics or in terms of outcomes from the pre-intervention and from the pre-intervention. And we collect some data also from the enterprise service, as I said. And after we did our, on this data, we do not do the difference and difference, or difference and differences. What we did was some fixed effect regressions only. But after we performed our difference and differences, we do a placebo test to see whether the results stand. What did we do there? We take some countries now, the randomness is more guaranteed. We take some countries around the world where they have not implemented EPA at all. And we compare the outcomes with those of the countries in SSA where EPA has never been implemented. And there is no significant difference in our placebo. But there are a lot of significant differences between our treatment and a control group. That gives us a better confidence and that this is basically one way to test the assumption of parallel trends and outcomes between treatment and control groups. And we don't go over the literature, no time. And you can see we gave this in our punchline already, but we have more results than this, that we do it over three years, five, seven and ten years. And also, even when it's not significant, but we have a design, the signs of the estimates are correct. And I may add also, in addition to our placebo test, we do an average treatment effect test, like before and after, and the results still stand. And we do them also in some financial outcomes, like you can see a trade from the private sector and is significantly, that means if those countries didn't have EPA, trade would have been 11% lower. Not trade, I mean to say credit, private credit would have been 11% lower. And we talked about this in the punch. Here are the policy lessons. And we said that those countries that have not implemented EPA probably need to do that. And because, and if they do it, they need to subsidize a credit form. Like you don't give an EPA usually, an EPA usually can give some subsidies to firms or also subsidize the borrowing cost and provide a matching grant or you do some equity infusion. But in countries with weak institutions where corruption might be high, equity infusion might be much better for the greater good than providing a matching grant. And also we would recommend that there are some financial incentives in terms of tax breaks, like reducing some barriers and offer insurance and do some intermediation between the traditional financial sector and the manufacturing firms. And also don't do an EPA just because you want to have another state agency. You have to mainstream it, like resource it, have competent staff and finance it. In conclusion, we said that EPA or EPAs have strong effects on manufacturing as early as three years following the implementation. And EPA may assume a certain task that are left out by a traditional financial intermediation. And SSE countries without EPA probably have missed out an opportunity to have better outcomes in terms of manufacturing and export. And our counterfactuals suggest that decline in industrialization in the region marked by some staggering economic losses could have been worse. Certain countries didn't have that kind of institution. And this is where we stop. And if you have any questions, thank you very much. Thank you very much.