New model predicts future poverty trends to help South Africa design effective policies

‘Income deprivation measures alone are not enough to plan around poverty, a multi-dimensional analysis is needed’

Johannesburg, South Africa: 20 March 2024 – Poverty, in all its dimensions, is a complex challenge that defies simplistic solutions.  Recognising the limitations of traditional income-based measures, Applied Development Research Solutions (ADRS) Global has launched a groundbreaking report that expands the scope of poverty analysis and offers insights into future trends.

The paper, Multidimensional Poverty: Future Proof with Linked Macro-Micro Modelling, co-authored by Dr. Asghar Adelzadeh and Ludwe Ngangelizwe, proposes a groundbreaking approach to predicting future multidimensional poverty (MDP) in South Africa – and planning for it. 

In 2014, Statistics South Africa produced the first official measures of multidimensional poverty in South Africa. Our study advances the approach to produce forward-looking projections on multidimensional poverty for the first time, Dr. Adelzadeh said. 

The outcome is a policy tool that can be used to design anti-poverty policies and assess their impact considering factors beyond income deprivation, such as education, health, and living conditions.

This provides South Africa with access to the world’s first  forward-looking MDP tool, a global breakthrough say the authors. 

“Our dynamic innovation is important internationally because it can be replicated in other countries, to increasingly integrate multi-dimensional poverty into policy design using evidence-based and advanced analytical tools of modelling,” said Dr. Adelzadeh. 

It can also specifically help other African countries integrate these tools to the local and global fight against poverty.

“Until today, this has not been done,” Dr. Adelzadeh added, speaking at the launch of the model at the Gordon Institute of Business Science (GIBS).

“We have now used our model to produce the current trajectory for deprivation of various population groups in South Africa and to test the likely impact of several economic and social policies on MDP indicators and want to share it to support government’s policy-making efforts.”

Past processes no longer as powerful

Traditionally, studies on poverty have focused on measuring deprivation and identifying factors that contribute to it. 

This study goes a step further by proposing a way to forecast future trends in multidimensional poverty. 

It considers factors such as education, healthcare, living conditions, and access to assets, to create a comprehensive picture of poverty beyond just income levels.

This innovative method can be used to design more effective anti-poverty policies and assess their potential impact from 2024 to 2030.

“By seeing and understanding projections of multidimensional poverty, policymakers are better able to make the type of adjustments today that can favourably impact what poverty looks like in the future,” said Ms. Ngangelizwe.

A powerful tool: The DIMMSIM Model

The study utilises a model developed by ADRS Global called DIMMSIM (Dynamic Integrated Macro-Micro Simulation Model) which integrates both macro (national) and micro (individual/household) data. 

This allows the model to project future economic trends and their impact on poverty indicators at a national and sub-national level, as well as by demographics like gender, race, and region.

The model’s power lies in its ability to simulate – dynamically not statically – the effects of various policy scenarios. 

The study explores five potential policy interventions, including fiscal and monetary policies, private investment, public employment programmes, and social grants. 

By analysing how these – and other policies – influence multidimensional poverty indicators, policymakers can gain valuable insights into the potential effectiveness of different anti-poverty strategies.

Scaled data analysis, robust predictions

The model leverages data from a comprehensive South African General Household Survey, encompassing over 125,000 individuals across nearly 62,000 families. 

This robust dataset ensures the model’s predictions are grounded in real-world data, leading to more reliable forecasts, making it an “extremely valuable tool for policymakers” in South Africa, said top leaders at the paper launch event today at GIBS.

By predicting future poverty trends and evaluating the impact of potential policies, the DIMMSIM model can empower them to design more targeted and effective anti-poverty strategies, by ensuring a big picture approach to combating poverty. 

“With this forward-looking tool South Africa is potentially able to lead our global counterparts in showing how multidimensional poverty can be integrated into policy dialogue and most importantly, into policy planning and policy design,” said Dr. Pali Lehohla, former Statistician General and now co-director of the Economic Modelling Academy with Dr. Adelzadeh.

“Today, using the forward-looking tool we can show what and how policies can be designed to provide more effective anti-poverty policies. This is not only important for combating poverty in South Africa but also for achieving the sustainable development goals, because all indicators that are components of multidimensional poverty are closely related to the SDGs.”