 Often, as a policymaker, you work a little bit in the dark. It's hard to know is what you're doing having an impact. Ultimately, it's about the policies, how they impact the lives of ordinary South Africans in a way that meaningfully can contribute to growth in the future. If you don't have good data, you can't answer any of these questions. Pretoria, South Africa On the 20th floor of the National Treasury, there sits an ordinary desk. On this desk, data on every firm in the country. For the first time in Africa, a government has opened up this data to researchers. The hope is that research can help lead to better policy. For us, data is very important and my team is responsible for economic analysis. We are a consulting unit to other units of the National Treasury and government as we do work for the National Planning Commission or the Department of Environmental Affairs or the Department of Energy. So as a unit which is highly dependent on data to do our work, data is very important. What data allows us to do is to evaluate government policy, to look back in time and understand what has been the impact of government policy. Did we get the value for money that we expected to get and how can one redesign those programs going forward to make better use of these resources? One key area where data and evidence have been missing in South Africa is on firms. Firms are the growth drivers, the job creators in our economy that are paying taxes which we then use to pay for schools, to pay for infrastructure. Economists in South Africa, the academic researchers, the policy makers did not have access to a view of how these firms behave, who they are, which sectors they are in, do they participate in international markets, are they small, are they big, what are their age profiles, are they fast growing or slow growing, what is their productivity. So that information just was not available. We were unable to internalize that information in our policy making. The National Treasury, together with you and you Wider, have embarked on a multi-year program to address this problem. Rather than find new data, the solution they have come up with is to follow international best practice and open up an existing pool of data, tax data on each firm, data that are already being collected yearly by the South African Revenue Service. It's being collected as part of an administrative process, but it can be used by researchers to generate empirical research. Research and policy working together may sound nice, but it is easier said than done. With you and you Wider, however, the project had a partner that had done it before. We know the academics in the country that are interested in doing this kind of work. We know the policy makers who are interested in the output to want to know results to key questions. What Wider brought was confidence to both of those sides that this could be done. We had done it before in Vietnam. We were able to organize a large workshop. We brought policy makers from all across the government, really at the technical levels, people who are really interested in research. And we brought leading academics who indicated these are the kinds of policy questions that we can answer if we have access to this data. We were able to back that up with international experts who went and said, in Vietnam, we went and looked into this particular question at the request of, say, the Ministry of Planning and Investment. We arrived at this conclusion and this policy action was taken. So both sides had the confidence that if they went and invested a lot of time and resources in terms of bringing these data into the analytical sphere, that something really useful would come out. A broad coalition was built, one where every party was invested and protected. SAAS is the South African Revenue Service. It's the South African Government's Tax and Customs Authority. So we collect all the revenue that actually is used to fund the South African Government's activities. Trust and best practice, I think, are very important. Confidentiality of taxpayer information is absolutely inviolate. That's an international norm. So there were many mechanisms that needed to be put in place to ensure that we could not only make sure that taxpayer confidentiality was protected, but that any perception that it could possibly be misused or used against taxpayers could be avoided. My name is Catherine McLeod. I'm the Chief Director of Macroeconomic Policy at the National Treasury. My unit is responsible for looking at structural reforms and growth at the exchange rate and monetary policy and also issues relating to the labour market. So what do we do to get people into jobs? How do we resolve the unemployment crisis? This sort of project, because it's so big, can't be done by one institution alone. It really needs to be done in partnership. You're a little bit nervous. Will the data, if it's released, cause us problems? What if the research comes out and it's telling us something that we don't want to hear? One of the keys to success here has been that we've had the South African Revenue Services, National Treasury and other interested government departments really involved in setting up the architecture of how this project would work. You can sit and research the question that you have at hand. What you can't do is you can't identify individuals, you can't identify companies, and more importantly you can't take the data from the machine. So the data sits at the National Treasury or at the South African Revenue Service and researchers can use it only at our premises. Tax administrative data that gets collected by the South African Revenue Service for administrative purposes has been converted into a firm level panel. The income statement information, the balance sheet information, the tax information in the data has been converted to give us a window into the life of a firm over time. And so this data, having been anonymized, having been quality assured, having been checked, is then placed on these computers for use by researchers. Of course it's not just one firm, it's hundreds of thousands. And so collectively that really provides a very useful picture of what is actually happening in the business sector in South Africa. The desk at the Treasury with its newly available data is a game changer for research and policy. What the tax data allows us to do is actually understand what individuals are earning, which firms they're working for and how long they've been working there. It's really a breakthrough in terms of trying to understand what is happening in the labour market. And the tax data allows us to understand the behavior of an individual firm and watch how that changes over time and we can start to attribute that to specific policy. The problem that we're facing in South Africa is that youth unemployment is currently around 44%, which is almost twice that of adult unemployment. The government's introduced this incentive to try to encourage companies to hire young people who face a very high rate of unemployment and I'm using this data to see if it's had that effect, if it has encouraged more firms to hire young workers. There's a lot of work to be done to understand how it's been taken up and why certain firms are taken up and others have not. This tax data is incredibly useful. It's much better than any other data we could have to try to evaluate this kind of a topic. And the end goal of all of this is to improve policies so that we can try to improve outcomes in our economy such as helping young people get a job. Policymakers can decide what research is going to help us as we're moving forward to understand the data, to understand firms, understand jobs and ultimately understand what sort of policies should we be making or could we be making. Most of the research is published on the UNU-wide website. What we haven't advertised is that a part of the research which is on the website, we have had a number of policy dialogues which are closed sessions for senior policymakers. The work on the employment tax incentive, on the learnership allowance, the work on capital versus employment incentives and how they affect labour costs has directly fed into a treasurer's policy decisions. We have received very positive feedback on the entire programme and the achievements over the last four years. A long-term aim of the project is to inform not only policy, but to build capacity both within the research community in South Africa and also in the policy departments themselves. I've seen my economists in my unit grow dramatically from the exposure that they've gotten with these international researchers across departments and with more senior policymakers really opened up. Based on the success of the project, we are hoping to expand the access to the dataset to the broader research community. Making the data broadly available, we believe, will increase the amount of research and the quality of research exponentially. Ultimately what this broad collaboration gives us is it allows the national treasury to make the right kind of decisions to create better opportunities for South Africans. This is one of the most innovative projects of its kind in Africa. With the success in South Africa, other countries in the region are looking to follow.