The Uganda Map of Financial Inclusion visualizes several financial service providers at the region and district level within Uganda. Specifically, the map contains layers mapping savings and credit cooperatives (SACCOs), credit institutions, commercial banks and both credit-only microfinance institutions (MFIs) and deposit-taking microfinance institutions (MDIs). There are also several demographic background layers that provide greater context surrounding the financial inclusion landscape. The following sections explain the data substantiating each layer, what each layer represents and any important assumptions made on the part of MIX while cleaning and preparing these data for mapping. All data used for mapping, raw data and any relevant questionnaires or further notes on the data can be downloaded and redistributed freely below as well.
Three layers within this map offer insight into the SACCOs operating in each district in Uganda. The data for this map derives from the 2011 SACCO Registry maintained by the Ugandan Ministry of Trade, Industry and Cooperatives (MTIC) Department of Cooperative Development. The registry contains a large amount of data on SACCOs that are active, dormant and under liquidation. The active SACCO layer’s circles are scaled based on the number of active SACCOs only, but the interactivity contains data on the number of SACCOs that are active, dormant and under liquidation in each district and region. The total SACCO Saving Account Value and Outstanding Loan Value layers represent the district and region totals for all active SACCOs. Each of these layers has a significant amount of data in its interactivity representing the savings / outstanding loans of individuals, institutions, cooperatives and groups. The individual data is broken down into male / female (m/f) and adult / youth (a/y) categories. Please note that the totals for the males and females are the sums of both the adult and youth males and adult and youth females, respectively. Similarly, the adults and youths totals are the sums of males and females of each age group. Detailed assumptions about how MIX manipulated the raw data can be found in the “sacco_assumptions.txt” file included in the download provided here. MIX has also provided both the raw SACCO registry data (and its accompanying questionnaire) and the data substantiating the SACCO map layers within this download.
The credit-only MFI layer represents the total number branches, headquarters and field offices in each district and region for each MFI based on the 2011 Association of Microfinance Institutions of Uganda. These data include MFIs that only give credit and as such do not include MDIs (see below) or FSPs like SACCOs or cooperatives that are within the raw data. A complete list of the financial service providers omitted is provided in the “omitted_fsps_amfiu.csv” file in the download here. Detailed assumptions about how MIX manipulated the raw data can be found in the “mfi_assumptions.txt” file included in the download provided as well.
The MDI layer visualizes branch level locations of deposit-taking microfinance institutions. In particular, the data stem from the branch and headquarter information of four financial institutions listed within the 2011 AMFIU directory:
This layer combines the branches of several different financial providers that are defined to be licensed credit institutions by the Bank of Uganda. The Postbank data comes from Postbank Uganda and the National Social Security Fund branches are from their own map. MIX extracted the Opportunity Bank data from the AMFIU directory mentioned above.
The Commercial Banks layer shows the number of commercial banks per district and region in Uganda from aggregated data from commercial banks across Uganda; though every effort was made to gather data from as many branches as possible due to the ever-changing nature of websites and bank listings, our listing may not represent all commercial banks operating in Uganda. A list showing the acronyms used in the pie chart and their corresponding full names can be found in this google document.
The map shown on the initial load of this page shows the combined points of service for all FSPs (Active SACCOs, MFIs / MDIs, Credit Institutions and Commercial Banks) to allow for a more immediate comparison of all FSP datasets.
The population density contextual layer is sourced both from the 2002 Ugandan Census and Afripop. Each pixel of this heat map represents an estimation the number of people per 100 square meters based on both the 2002 Ugandan Census and satellite image interpretation, according to Afripop’s methodology. The data appearing within the interactivity (the box appearing on hover) of this layer is from the district reports of the 2002 Ugandan Census only.
The distance to health facilities, schools, water source, under 18 population and average household size contextual layers come from the 2002 Ugandan census, all of which go down to the county level. There was a substantial increase in the number of districts in 2010 (from 56 to 112). However, the counties remained largely unchanged and the district reports of the 2002 Ugandan Census have county level information for all of these indicators. MIX used the Ugandan Electoral Commision’s 2011 breakdown that details which counties are in which districts to calculate the census values for the new districts.
Please note also that the color scheme for the under 18 layer uses an absolute scale, rather than percentiles (see below) to highlight the fact that the vast majority of the country has more than 50% of its population under the age of 18. The average household size layer uses an absolute scale with a similar justification.
Rural poverty data was obtained from Spatial Trends of Poverty and Inequality in Uganda: 2002-2005. The dataset provides detailed poverty statistics on non-urban regions within Uganda at the sub-county level. Please note the scaling of the poverty levels on this map does not extend from zero to one-hundred percent.
Data for this layer can be found at the Socioeconomic Data and Applications Center at Columbia University. We utilized the Urban Extents dataset and overlaid that with the population density dataset to create boundaries of urban regions. Through the use of a conversion factor from square arc degrees to square kilometers, we calculated the population density for regions on the map; populations falling within the urban boundary were considered urban population and the populations outside of such regions were considered rural. Please see the GRUMP and Gridded Population of the World dataset and methodology for more detailed information. Given that urban regions were ascertained using nighttime lights data, it is likely that the urban regions without electricity were categorized as rural in this dataset; it is therefore plausible that the rural population is overestimated or, conversely, the urban population is underestimated by this methodology for this particular country.
Infrastructure Data was gathered from Infrastructure Africa. Main roads, power transmission lines, and power plants are included on this map and are available for many Sub-Saharan African countries.
Except where noted above and for some outliers, all the colors and dots sizes displayed in this map are scaled based on the 25th, 50th, 75th and 90th percentiles of the respective dataset they represent. This guarantees a view into the relative extremes of each dataset, rather than employing an absolute scale. The 25th, 50th, 75th and 90th percentiles for the shaded contextual data are displayed in the legend of each layer.
The map files representing national, regional district, county and sub-county boundaries are sourced from Global Administrative Areas (GADM), an international database of geographic files (shapefiles) covering every country in the world. The data are maintained by contributors working at the International Rice Research Institute and the University of California, Berkeley’s Museum of Vertebrate Biology. The shapefiles are drawn primarily from the Centers for Disease Control and the UN Geographic Information Working Group’s “Second Administrative Level Boundaries” project.
The original GADM shapefile is current as of 2008. As mentioned in the 2002 Ugandan Census section, there was a substantial increase in the number of Ugandan districts in 2010, though the counties remained largely the same. MIX used the Ugandan Electoral Commision’s 2011 breakdown to construct an updated district layer from the GADM shapefile. MIX also created the regional level shapefile from the original GADM based on geohive’s breakdown and changed three district names based on the AMFIU directory, described in the “mfi_assumptions.txt” file.