Calculating a Poor/Non-Poor Capital Expenditure Ratio: Applying a generalised hexagon model zone
Spatial targeting refers to the deliberate focus of particular actions on a particular spatial area in order to achieve desired outcomes or objectives. This concept is a recent, but essential, addition to the traditional planning approach followed by municipalities in delivering on their mandate and can easily be identified in legislative and policy documents such as the Spatial Planning and Land Use Management Act of 2013 (SPLUMA), The Integrated Urban Development Framework (IUDF), and the Capital Expenditure Framework (CEF) guidelines.
The CEF guideline requires municipalities to report on capital expenditure towards poor and non-poor households within the municipality – from here on referred to as the Poor/Non-Poor Capital Expenditure Ratio.
Various attempts were made to determine the Poor/Non-Poor Capital Expenditure Ratio. Three of these attempts worth mentioning, which are still useful but in a different context, includes the Deprivation Index, Indigent Index, and the municipally adjusted Gini-Coefficient.
The Deprivation Index is a spatial index which identifies the most deprived areas within the municipality using household income, size, dwelling type, access to water, access to energy, access to refuse removal, and access to sanitation. This however was done using the Small Area Layer (SAL) and was not related to the actual dwelling structure distribution within the analysis area.
The Indigent Index took a similar approach than the Deprivation Index in using the SAL as the spatial representation of the data, however accessing data as defined by the National Framework for Indigent Policies is not an easy task – with some data sets not publicly available.
The Gini-Coefficient shows how much a specific zone within the analysis area varies from the norm within the analysis area making it difficult to differentiate between a high income classes and a low income class as they both vary from the norm to the same degree. Additionally, how the query zones were defined was also dependant on the SAL – thus not reflecting a true spatial distribution of households.
An alternative solution to calculating the Poor/Non-Poor Capital Expenditure Ratio is represented in the figure attached to this article – Calculating a Poor/Non-Poor Capital Expenditure Ratio: Applying a generalised hexagon model zone .
The first step is to generate a 500m hexagon grid within the analysis zone. The second step is to identify the household distribution by disaggregating household data as per SAL to the dwelling frame data – data which is also accessible from Mapable (Pty) Ltd. The thirds step is then to aggregate the number of households per hexagon zone in order to standardise the analysis zones. At this point in the calculation process the analyst identified the number of households per hexagon zone, together with the household attributes linked to that hexagon. The fourth step is then to identify the household income ratio per hexagon, using an income of less than R5000 per month as an identifier for poor households, and more than R5000 per month as an identifier for non-poor households. The next step in the calculation process is to identify the proportion of capital expenditure per hexagon. This is done by splitting the capital per project according to the proportion of which the project intersect with each hexagon using the reporting module of CP3. The result is thus the amount of capital spent per hexagon. The last step in the analysis is to multiply the household income ratio per hexagon of step four with the capital expenditure per hexagon. A future article will deal with the output of this process and how it can be interpreted.
This method of identifying the Poor/Non-Poor Capital Expenditure Ratio is but one approach towards measuring a municipality’s intent towards spatial transformation. It attempts to take into consideration the spatial reality to an above normal level of accuracy, whilst at the same time utilising relatively easily accessible tools and data sets.