 Thank you very much. I'm very pleased to be making this presentation on behalf of the research team that worked on this very topic, which is an assessment of inequality estimates the case of Ghana. And my presentation is going to be following this outline. So I'm going to provide a brief introduction where I try to provide some brief context to the study and state our main objective for the study. And I also like to provide a much broader context of what the narrative of inequality has been in Ghana, what the story has been over the years. And then I would get into the the weight inequality narrative focusing more on the kinds of data that they use and then the inequality story that emerges from the UNU wider research team analysis of their data. And then what I would do then at the end of the to conclude my talk would be to share ASUS narrative of inequality based on what we do with the analysis then equality analysis we carry out using the national level data. So I mean we would all agree that measuring actually measuring inequality is definitely critical when it comes to researchers trying to understand you know the trends and then for effective policymaking right. We usually know that when you have less reliable and inconsistent and inequality measures it makes it more difficult for us to see the trends in terms of monitor whether inequality is increasing or decreasing. Sometimes the inequality estimates that we get even for the same country is different using you know different measures some like Carlos already mentioned sometimes we have consumption inequality sometimes you have income inequality even for the same country using the same data set is different. Another difficulty that we find is that is usually not easy trying to undertake comparative studies maybe comparing inequality across different countries in sub-Saharan Africa or even for different regions of the world it's very difficult because the measures are not the same. So this project that the UNU wider team has embarked on is definitely a good step in terms of they trying to make data more accessible to facilitate inequality research across the globe and particularly with their work on the with companion which tries to standardize the measures of income inequality which will make it possible for us to be able to carry out these cross-country analysis or even for the same country over time we get the same measure and we be able to see whether inequality is decreasing or increasing but what we do observe from the with companion is at least for the case of Ghana is a fact that we do find some disparities in terms of the measure of income inequality that has been estimated by the UNU wider team and what we also do at ASA and so basically for this particular paper what we try to do is to investigate the extent to which you know the observed disparities that we find from the with companion and from our own research how those we could actually consider those variances or those disparities to be reasonable so in the paper what we do is to explore in detail the inconsistencies particularly with the national surveys that we use in terms of how inequality is measured and then we try to describe the trends using data from 1992 to 2016 2017 and then what we do then is to compare those inequality estimates with what the with companion actually does so just to give you a broad idea of what the inequality story has been in Ghana for data has actually shown or has been well documented that for the past decade or so Ghana has been recording very impressive economic growth rate but unfortunately this rising economic growth has also been associated with increasing inequality and what we actually find in our case is the fact that when we decompose the inequality estimate into within inequality and between inequality largely for Ghana we find that inequality seems to be a within you know issue because the subgroups the the inequality coefficients that we find are higher within groups compared to between so I'm just going to show the trends for consumption inequality labor market inequality particularly focusing on wage employment and then I would also give a broader perspective considering access to social amenities and I would like to mention here that even though I stated earlier that the data starts from 1992 to 2017 for this particular presentation I'm only focusing on the last three rounds of the Ghana living standards survey and the reason is that these three surveys are more comparable in terms of the instruments that are used to collect the data so it makes it easier for us to actually compare you know what is going on in terms of inequality so we do find that inequality is increasing over time as we can see from the graphs on the left panel and as I mentioned earlier when we disaggregate by different subgroups urban rural regional levels and also educational levels we do see that inequality in Ghana is mainly within groups so that tells us that there are no particular group that is driving the inequality that we find in Ghana is mainly based on systemic factors within the economy and what I show on the right panel is just giving you an idea of the heterogeneity in terms of the genie coefficients right so we see that over the three waves of the survey some regions have recorded declined in inequality whilst other regions have also recorded increases in inequality so basically what this means is that there are differences in terms of the inequalities that we find and it's also interesting to note that the regions that have high levels of poverty are the ones that are actually associated with increasing inequality in Ghana so here I'm showing wage inequality and as I mentioned earlier this is mainly for paid employees and then we see that for the three waves of the data in wage inequality seems to be decreasing and this is mainly because of a policy that the country had sometime in 2009 where you know the government tried to you know bridge the gaps in terms of wage wages for different people or categories of people and I also wanted to share what the asset inequality was looked like over the three waves so we see that between 2006 and 2013 there was an increase in asset inequality but then this dipped in 2016 2017 and here mainly I just want to show the regional differences with when it comes to access to social and economic services for instance and many things so we see regional differences when we look at things like access to electricity and also access to sanitation the for again relatively richer regions seems to have more equal access compared to other poor regions who do not have access and the same goes to sanitation now I want to talk a bit about the data that we use at ASA and what and UNU wider team also makes use of in their analysis mainly the original and also the with companion all the data that are being used by these two different groups if I can say that is based on the Ghana Living Standards Survey which is the main nationally representative data set that is available for inequality studies overall we have a total of seven rounds starting from 1988 to 89 and the latest round that we have is in 2017 now it's I have to mention that all the rounds are not completely comparable as I mentioned earlier it's only that last three rounds that are comparable and like Carlos mentioned in his introductory marks some measures almost every year has a different like you know measure so it becomes difficult actually monitoring inequality measures so some of the years would have expenditure per capita others would have adult equivalence other in some years income will be measured as gross others to be mentioned at net as net income others don't even have those data at all and in the last two rounds for instance additional items had been added to the consumption basket in terms of the computation of inequality consumption expenditure so as I mentioned the most recent three rounds have more identical instruments and are more comparable and so that's what we actually rely on so what is the narrative inequality night that is coming out of the with original and then the the with companion and I'm going to leave the you know the whole methodology of the with companion later on for us to discuss but basically on the left-hand panel the blue line is showing what we call the with original which is just you know you and you wider just trying to compile all the databases for inequality measures for different countries so basically they also tap into the national surveys so the blue line is what you would have and this is this would be the consumption inequality but what the orange line is showing is a standardized income inequality that is measured by the UNU wider team where they do this conversion from consumption inequality into income inequality and they use this methodology and for countries like Ghana and Kenya who were not particularly in part of the data what they do is to use this predictive and regressions to be able to project to see you know convert basically the consumption inequality into income inequality so as you would expect you know income inequality is definitely higher than consumption inequality so that's no surprise there but what we see is that at least for the two for the consumption inequality which is depicted by the blue line and then for income inequality which is from the with standardized data and we do see that over time inequality whether consumption or income is actually increasing and the fact that they are actually parallel tells us that the rate of increase or the slope is actually similar so they are actually growing at the same pace what's at what we do on the other side is just to bring in the national estimate of consumption inequality and like I said and this would be very similar to what Pofcarnet actually reports for some reason the national the Ghana sasika service reports actually shows a slight difference even though overall this is actually very related or closer to what we find with the Pofcarnet but then again we see the with standardized estimates being above indicating that income inequality rates estimates are higher than the consumption inequality now what do we find now even though and you know income estimates are not usually the preferred choice for the Ghana sasika service because of the obvious reasons right trying to get income data is always very difficult so they are more comfortable reporting consumption inequality compared to income inequality but what we did was that once the data was available we calculated the Gini coefficients using the income aggregates that were reported in the data and that's what is presented in the gray grayish line so the blue line is still showing the consumption inequality by Pofcarnet and then the orange line is still showing the income inequality that is estimates that is reported by the with standardized so basically we wanted to find out how our estimates from the national service would compare with what the UNU wider team had done using the with standardized and when we started that we actually saw that there was some disparity from the beginning but then as time went on this gap actually closed but so we thought okay then the with standardized was doing a good job of predicting income inequality given that we're also using national service but then beyond 2006 we actually saw this huge gap right in terms of what we were finding as income inequality compared to what the UNU wider team was actually finding and we did not really understand why right so this this is what we find from our analysis that's you know to some extent there are some you know from the beginning there was good predictive power but then at some point it wasn't really 15 anymore so just to conclude my talk I see the zero sign we do we acknowledge how useful this database is in terms of you know allowing us to be able to do comparative analysis when it comes to income inequality so this you know database that is being compiled by you and you are that is definitely a step in the right direction now what we actually need to do going forward is to really have a deeper understanding with more research is actually required to understand the best ways to actually compute these income inequality particularly for countries where consumption inequality or consumption estimates are what are being used so I would conclude by saying that definitely the wood companion makes a very good attempt at providing income inequality measures but we still need to understand more because the the methodology that they use is actually based on other African countries particularly South Africa which is very very much different from you know Ghana so we wouldn't be able to you know project what we are finding in South Africa for instance to other countries like Ghana so thank you very much