Above the line

Solving the poverty problem

THE INTERNATIONAL FOCUS on eliminating extreme poverty globally has grown since the late twentieth century, even among those who don’t view its elimination as utopian but simply as plausible. The most internationally recognised of these attempts came through the Millennium Development Goals, which were succeeded by the Sustainable Development Goals (SDGs, which began to be implemented in 2016 to be achieved by 2030). The former included a goal to halve the proportion of people living in extreme poverty between 1990 and 2015, while the latter aims (among other things) to end poverty everywhere in all its forms, including extreme poverty.

Indeed, since the 1990s, substantial reductions in extreme poverty have been achieved – one billion fewer people lived in extreme poverty in 2015 than in 1990, with significant progress recorded by countries in East Asia (particularly China), the Pacific and South Asia. However, since 2015, the pace of this progress has slowed, in part because poverty is now concentrated in sub-Saharan Africa, which has experienced both slower rates of poverty reduction than other regions and high population growth.[i]

Worse, 2020 was the first year in which the number of people living in poverty increased in many countries – a result of the widespread and dramatic impacts of the economic slowdowns associated with the COVID-19 pandemic. In 2020, as many as 115 million more people may have fallen into extreme poverty as a result of the pandemic, with a projected further increase of up to thirty-five million in 2021. This would bring the total number of people newly living in extreme poverty around the world to as many as 150 million. [ii]

While much is made of potential opportunities to ‘build back better’ post-Covid, the substantial increase in the number of people living in extreme poverty since the onset of the pandemic means that any goals to eliminate extreme poverty will now be even more difficult to achieve.[iii]

One argument that might facilitate original thinking about how to most effectively reach those living in extreme poverty – and make a difference to their situation – is to change the way poverty is measured; this can change how the problem is approached and how responses to the problem are designed. Complementing ‘traditional’ measurement approaches with a broader focus across multiple dimensions of poverty can help shed more light on the lived experience of people living in poverty. Changing what’s measured, and how, can’t by itself change the incentive of governments and policymakers to respond, but it can help more clearly illustrate specific aspects of people’s living and illuminate cumulative impacts of multiple deprivations that are likely to play out now and in the future. All of which could transform thinking about ways to tackle poverty alleviation and reduction. One example of a multi­dimensional approach that demonstrates greater complexity, detail and nuance around the experience of living in poverty is the Individual Deprivation Measure (IDM), a gender-sensitive measure of individual-level poverty.

The IDM was tested in South Africa, that continent’s second-largest economy, as part of a research process to develop and refine its metrics and potential. South Africa has both higher poverty rates than many other middle-income countries[iv] and reportedly the highest inequality in the world. These high levels of poverty and inequality are a result of persistently high levels of unemployment since the country’s first democratic elections in 1994 and economic growth rates that have been too low to increase employment levels. Estimates made since the start of the COVID-19 pandemic suggest that it may be one of the most affected countries on the continent[v] and that it may be at least five years before South Africa can return to 2019 levels of economic activity.[vi] In his 2021 State of the Nation Address, President Cyril Ramaphosa acknowledged sharp declines in economic activity and unemployment and a resultant rise in poverty and inequality.[vii] By late May 2021, South Africa had recorded more than 1.6 million confirmed Covid cases and 55,000 known Covid deaths.[viii]


POVERTY IS GENERALLY recognised as an inability to achieve a minimum standard of living in terms of meeting basic consumption needs for an individual or household.[ix] Its measurement has most often reflected the economic conditions of households. International poverty measures have this monetary focus, with the World Bank’s extreme poverty line being defined as income or consumption expenditure (the latter is preferred to the former) that falls below US$1.90 per person per day.[x] In 2014, it was estimated that almost 19 per cent of South Africans, or 10.4 million people, lived below this extreme poverty line.[xi]

Internationally comparable poverty lines such as this one do enable levels of poverty to be compared between countries and changes to be tracked over time, which is useful for assessing progress towards goals such as the SDG on poverty or the World Bank’s goal of reducing extreme poverty to 3 per cent of the world’s population by 2030. They help us understand where progress has been made and where poverty persists. However, such monetary measures do not really increase understanding of what it’s like to live in extreme poverty or how these experiences might differ from country to country.

Most countries also set national poverty lines designed specifically for local circumstances – and most of these are also monetary measures. For example, the South African food poverty line measures the most severe poverty in the country and was set at ZAR585 per person per month (approximately US$1.17 per person per day) in 2020, the estimated cost of purchasing a standard basket of food to meet the minimum required daily energy intake.[xii] Around one quarter of South Africa’s population – almost fourteen million people – were unable to meet these basic food needs in 2015, declining from a proportion as high as 34 per cent, or 16.7 million people, in 2009 in the wake of the global financial crisis.[xiii]

In recent decades, many have recognised that poverty measures can be improved, in particular by examining the multiple dimensions across which individuals experience poverty rather than just by measuring their income or consumption. Multidimensional measures can include a range of material, social, psychological and other dimensions, which means they have the potential to communicate a more complete picture of lived experience. Different combinations of dimensions can be measured, the selection of which can be made to use data that is already available (such as the UN Development Program’s Human Development Index), or to reflect public or expert consensus on what matters, or by determining the aspects that people living in poverty themselves view as important.

The IDM measures fourteen dimensions of poverty at the individual level, with dimensions selected on the basis of expert opinion and the values of people living in poverty across six countries in Africa and the Asia-Pacific. In 2019, survey data was collected on these fourteen social and economic dimensions from more than 8,500 individuals aged sixteen years and over in South Africa. Through this, it’s possible to examine a small number of indicators selected from the IDM dataset to illustrate and illuminate some aspects of the lived experience of those living in severe poverty – how their basic needs are met (or not) and also to provide a picture of some of the non-economic aspects that are rarely measured elsewhere.

It’s important to note that the data reported is for the bottom 20 per cent of the IDM dataset so that this group is roughly equivalent (by proportion) to those living in extreme poverty according to the international poverty line and the South African food poverty line. Selecting these people to talk about doesn’t imply that others don’t also experience multidimensional poverty – it’s just not as severe as this. Furthermore, it’s those experiencing extreme poverty that poverty reduction goals such as the SDGs target.

Looking closely at the individuals experiencing extreme poverty, more than eight out of every ten people experienced severe food insecurity – they had been hungry but didn’t eat and/or had gone without eating for a whole day (or more) because of a lack of money or other resources for food in the month prior to the survey. These figures – already dreadful – are thought to have worsened significantly since the start of the COVID-19 pandemic.

But energy for cooking is also essential to food security: without it, food cannot be cooked and meals may have to be skipped.[xiv] Smoke from unclean fuels can also cause respiratory illnesses. Among the most severely deprived, one in five people only had access to unclean energy sources (mostly wood), with only enough resources to meet their needs some of the time (or less). Indeed, only a third of this group had access to clean cooking energy sources and had enough to meet their needs at least most of the time.

While 85 per cent of individuals use electricity as their main source of light in the home, just 66 per cent use electricity for cooking and 32 per cent for heating, likely because of its cost (and the cost of stoves and other appliances) and unreliable supply. The South African state electricity provider periodically imposes intermittent ‘loadshedding’, or rolling outages, because of supply problems.

Around 40 per cent of the most severely deprived did not have enough clean drinking water to meet their needs at least most of the time. Depending on where they lived, these individuals could either access clean sources of drinking water but only sometimes had enough – or they had to rely on unclean water sources. One in twenty relied on unclean sources of water and still did not have enough to meet their drinking water needs at least some of the time.

Of the most severely deprived group, only around a third used a flush toilet at home that had enough water to function (though this was not necessarily a private toilet in their house), and almost 40 per cent used another type of ‘improved facility’, such as a ventilated improved pit latrine or a composting toilet. Access to improved toilet facilities such as these reduces the spread of disease and deaths (for example, from diarrhoea, cholera and other causes) and can reduce malnutrition, which is associated with poor sanitation.[xv]

Though not a measure of basic material needs, the need for literacy and numeracy is easily understood, as are the implications (especially for employment) of their absence. The most severely deprived South Africans are less likely to be literate and numerate than not – as determined by people completing very simple reading, writing and mathematical tasks during the survey. Fewer than a third were classified as being both functionally literate and numerate, and a quarter were classified as both functionally illiterate and innumerate. The remainder were partially so.

People were also asked about their ownership of basic but necessary household and personal items – adequate cooking utensils, cutlery and crockery, bedding, and water-storage containers for those without water piped to their houses. Only around one third of the most severely deprived had enough of all four of these basic items.

Mobile phones and internet connections are also key personal assets in the twenty-first century: only three-quarters of the most severely deprived South Africans owned their own mobile phone (14 per cent had no access to a mobile at all), and just 15 per cent could access an internet connection. However, these figures do not imply that individuals with phones and/or internet connections had money to buy airtime or data, or that they were able to access a quality connection (in terms of reception, data speeds and so on) or even electricity to charge a device.

A rarely measured aspect of poverty is the ownership of clothing and footwear. Among the most severely deprived group in South Africa, around four in ten did not meet a threshold set as the minimum acceptable standard in a humanitarian emergency – a really very basic measure of whether an individual owns two complete changes of clothing and two pairs of footwear. Clothing is plainly important not only for people to be able to move about in public (outside of their homes) with dignity and without shame, but it’s also important to have enough clothing that can be worn for work, whether for manual labour or in more formal settings, such as shops or offices.

Historically, unemployment in South Africa has been extremely high, hovering at around 25 per cent since the first democratic elections in 1994.[xvi] However, the picture for those seeking employment, whether in the formal or informal sectors, is even worse among those who are most severely deprived. The unemployment rate for working-age people (those aged between sixteen and sixty-four) in this group was 50 per cent overall and 58 per cent for youth aged between sixteen and twenty-four.

From this type of individual-level data across these indicators (and many more), a more complete picture emerges of what severe multidimensional poverty can mean for people. The data also enables a closer examination of how different groups may experience poverty differently – for example, depending on their gender, their age, or whether they live in rural or urban areas. This type of analysis can be important for designing and targeting responses, potentially increasing their efficiency and effectiveness. Men had worse outcomes than women in the severely deprived group in terms of clothing and footwear – a quarter of men did not own two complete changes of clothing and two pairs of footwear compared with 13 per cent of women. By contrast, among those of working age who were not in the labour force, more than six in ten women reported reasons for being unable to join the labour force and thus not being able to work compared with four in ten men (those remaining being out of the labour force by choice). There were also important differences between age groups, with just 6 per cent of youth classified as functionally illiterate and innumerate compared with 22 per cent of those aged twenty-five to sixty-four and 69 per cent of those sixty-five and over. Finally, only 2 per cent of individuals in rural areas used a flush toilet and had enough water to do so compared with 55 per cent of urban respondents – one example of the lack of public service delivery experienced particularly, but not exclusively, by those living in rural areas.

Individual-level measurement also allows the collection of data on issues that are relevant only to certain sections of the population. Of the women in the most severely deprived group who had menstruated in the six months prior to the survey, just over half reported not having access to sufficient sanitary products, and more than a third reported that they missed activities due to stigma associated with menstruation. Almost six in ten of the women who lacked sanitary products despite needing them missed social activities, school or work as a result of the shortage. For a young woman who reports regularly missing school because she doesn’t have enough sanitary products, there are likely to be negative impacts on her educational achievement, with potentially substantial cumulative negative impacts on her income-earning potential across her lifetime.

While this data was chosen to highlight some of the multiple deprivations experienced by those living in severe poverty in South Africa – and their implications are terrible enough – the COVID-19 pandemic is likely to have worsened the situation across multiple dimensions, including health, employment and, consequently, income and food security. Long-term negative impacts on education for young people are also anticipated because of school closures and a lack of access to electronic devices and internet connections for most poor families, aside from the difficulties with home schooling that illiterate parents and carers face. Further, the difficulties faced by those who rely on social networks to provide support for basic needs have been compounded by the mobility restrictions and any negative economic impacts felt by those who provide such support.


IN ADDITION TO generating a more complete understanding of multidimensional deprivation, another strength of collecting data at the individual level is that it can help to avoid some of the problems associated with many of the monetary measures of poverty. Though many poverty measures are reported at an individual level, they are derived from data collected at the household level – a single person in each sampled household is interviewed, and data is collected on the household composition (how many people live there, each individual’s age, gender, and so on) and any other data necessary. The total resources of each household are then divided among the household members either equally to generate per capita measures[xvii] – or using a measure of ‘equivalence’ that takes into account the number and age of household members. Children are scaled as a proportion of adults and adjustments are made to take account of economies of scale – it does not cost double to meet the needs of a two-person household compared with a person living alone.[xviii]

These calculations assume that all household members have equal needs and equal (or proportionate) access to the household resources. In reality, individual-level access to resources depends on those individuals’ bargaining power within the household, and so household-based measures can result in substantial underestimates of inequalities, particularly by gender and age.[xix] This is important because many policymakers assume that targeting poor households means they will be able to reach those living in the most extreme poverty. However, evidence such as that indicating that a substantial proportion of underweight women and undernourished children in sub-Saharan Africa are not found in the poorest households[xx] suggests this assumption is questionable.

Collecting data at the individual level avoids the need for such adjustments. In this way, the different experiences of individuals – whether men or women, young or old, those with or without disabilities – that arise from their differing access to household resources can be directly observed from the data, and more specific and accurate targeting of policy responses is thus feasible.


THE WAY POVERTY is measured does not automatically determine the policy response taken, but it does strongly affect the way problems are understood and solutions are devised. Since the mid-twentieth century, poverty has typically been measured using monetary approaches and has commonly been associated with a lack of economic growth. Policymakers across the world have therefore tended to advocate for increasing economic growth rates to achieve poverty reduction (increasing employment and income-generating opportunities) and, to a lesser extent, building human capital (education and skills). But policy responses also need to consider redistribution to ensure that some of the benefits of economic growth are directed towards those experiencing poverty[xxi] if inequality is not to worsen further.

In South Africa, redistribution has taken place through the expansion of social grants – mostly the child support grant and the old-age pension – with recipients increasing from 2.6 million in 1996 to 17.7 million in 2018.[xxii] However, while these measures make important contributions to poverty alleviation, the experience in South Africa has shown that economic growth does not necessarily translate to higher employment and reduced poverty,[xxiii] and substantial and permanent reductions in the country’s very high levels of poverty have not yet been achieved.

A more widespread adoption of multidimensional poverty measures could open pathways to more radical and creative ways to tackle the many different aspects of poverty. The wider focus of multidimensional poverty doesn’t preclude attention to economic growth or redistribution or efforts to increase employment and/or income-generating opportunities – income remains critical to the purchase of a range of goods and services that are necessary to improve living standards – but it can increase the visibility of what needs to be achieved and a multitude of ways that policy responses could be targeted to more effectively respond to different needs. For example, this data demonstrates the widespread need for improved service delivery in South Africa, but it also shows that the required focus of that delivery could be more effectively targeted in different locations. In the broadest sense, improving individuals’ access to clean energy and water sources is of critical importance in rural areas, while in urban areas, a majority of people already have this access, which makes improving the reliability of electricity and piped water supplies more relevant to many people living in towns
and cities.

Multidimensional measures can also show the ways in which interactions between dimensions or indicators move together or in opposite directions, information that could potentially feed the design of policy responses to address them jointly. Following the examples above, the explicit incorporation of clean (and sustainable) energy access into food security programs, or the systematic requirement to provide improved sanitation and sanitary products (for menstruation) in education policies – as well to improve the quality of education offered – present themselves as obvious positive targets to reach.

Just as multiple deprivations faced by individuals make it more difficult to escape poverty, individual measurement can yield opportunities for specially designed responses targeting specific combinations or profiles of poverty experienced by certain groups. This could be done by the same agencies (government, donor agencies, non-government organisations) in the way poverty reduction programs are already designed and implemented, but using a wider range of evidence available to inform the design of their policies and programs.

Because IDM data has only been collected once in South Africa, it cannot yet say anything about movements in and out of poverty – understanding these dynamics requires repeated data collections over time. However, other research conducted in South Africa indicates that there is little intergenerational mobility at the bottom of the income distribution scale and that around a quarter of the country’s population will move above and below the highest national poverty line over time. This suggests that children’s situations are largely determined by their parents, there is very little intergenerational improvement for those at the bottom, and change for those in the middle of the distribution is often not secure – both being driven by inequality of opportunity.[xxiv]

Historically, efforts to reduce poverty have often resulted in targeting those just below the poverty line to move them just above it – a comparatively low-cost and easy approach. This can generate apparently quick successes, but it may barely change the quality of life of those involved nor assure that their transition above that line is maintained. Think, for example, about the difference in living standards of someone moving from $1.85 per day to $1.95 per day. Improvements achieved in this way mean reductions in poverty rates can be achieved without any gain to the poorest of the poor. Interventions to reach the poorest of the poor typically involve relatively higher costs because of the severe and multiple deprivations they face and the rural locations in which they often live. In addition, given that these groups often have the least political influence, few countries with high poverty rates embrace strategies to target them.[xxv]

But the eradication of extreme poverty is plausible, even feasible, though a world without extreme poverty should not qualify as anything utopian – it is, after all, a modest goal that has been being framed as achievable for decades. But its achievement would improve, often dramatically, the lives of almost one in ten people globally[xxvi] – those who currently live in extreme poverty. While measures such as the IDM that take into account the individual and multidimensional nature of poverty can improve the evidence base for new policy, they can’t change policymakers’ incentives to design and implement successful poverty eradication policies. Ultimately, identifying and positively impacting policymakers’ incentives to respond effectively is critical to achieving even such a modest goal as this and taking these critical but small steps towards a more just world.


The 2016–2020 Individual Deprivation Measure program was a partnership between the Australian National University, the International Women’s Development Agency and the Australian Government through the Department of Foreign Affairs and Trade. Thanks to the ANU IDM team, particularly Sharon Bessell, Janet Hunt, Trang Pham and Mandy Yap.




[i] World Bank, 2018, Poverty and shared prosperity 2018. Piecing together the poverty puzzle. World Bank: Washington, DC.

[ii] World Bank, 2020, Poverty and shared prosperity 2020: reversals of fortune. World Bank: Washington, DC; Egger, D., et al., 2021, Falling living standards during the COVID-19 crisis: Quantitative evidence from nine developing countries. Science Advances, 7(6): p. eabe0997; Lenhardt, A. 2020, Pushing people further into poverty: the impact of Covid-19 in lower- and middle- income countries. ODI Comment Available from: [accessed 17 December 2020].

[iii] World Bank, Poverty and shared prosperity 2020.

[iv] Sulla, V. and P. Zikhali, 2018, Overcoming poverty and inequality in South Africa: an assessment of drivers, constraints and opportunities. World Bank: Washington, DC.

[v] Sulla, V., 2020, Poverty and equity brief. South Africa. World Bank: Washington, DC.

[vi] UNDP, 2020, COVID-19 in South Africa. Socio-economic impact assessment. United Nations in South Africa. UNDP: Pretoria.

[vii] Ramaphosa, C. 2021, State of the Nation Address 2021. Available from: [accessed 18 February 2021].

[viii] WHO. 2021, WHO Coronavirus (COVID-19) Dashboard, South Africa. Available from: [accessed 24 May 2021].

[ix] World Bank, 1990, World development report. World Bank: Washington, DC.

[x] Ferreira, F.H.G., et al., 2016, A global count of the extreme poor in 2012: data issues, methodology and initial results. Journal of Economic Inequality, 14, pp. 141–172.

[xi] World Bank, Poverty and shared prosperity 2020: reversals of fortune.

[xii] Statistics South Africa, 2020, National poverty lines, P0310.1. Statistics South Africa: Pretoria.

[xiii] Statistics South Africa, 2017, Poverty trends in South Africa. An examination of absolute poverty between 2006 and 2015, Report No.03-10-06. Statistics South Africa: Pretoria.

[xiv] WFP, 2019, Energy for food security. Enhancing people's food security with improved energy access. World Food Programme: Rome.

[xv] WHO. 2019, Sanitation key facts. Available from: [accessed 7 April 2021].

[xvi] GRSA, 2019, Towards a 25 year review. 1994 to 2019. Government of the Republic of South Africa: Pretoria.

[xvii] Ferreira, F.H.G., et al., 2016, A global count of the extreme poor in 2012: data issues, methodology and initial results. Journal of Economic Inequality, 14, pp. 141–172.

[xviii] UNECE, 2017, Guide on poverty measurement. UN Economic Commission for Europe: New York, Geneva.

[xix] Klasen, S. and R. Lahoti, 2020, How Serious is the Neglect of Intra-Household Inequality in Multidimensional Poverty and Inequality Analyses? Evidence from India. Review of Income and Wealth: p. 27.

[xx] Brown, C., M. Ravallion, and D. van de Walle, 2019, Most of Africa's Nutritionally Deprived Women and Children are Not Found in Poor Households. The Review of Economics and Statistics, 101(4): pp. 631–644.

[xxi] Edward, P. and A. Sumner, 2019, The end of poverty. Inequality and growth in global perspective. Cham: Palgrave Macmillan.

[xxii] GRSA, 2019, Towards a 25 year review. 1994 to 2019. Government of the Republic of South Africa: Pretoria.

[xxiii] Plessis, C.d. and M. Plaut, 2019, Understanding South Africa. Jacana: Auckland Park.

[xxiv] Liebbrandt, M., S. Schotte, and R. Zizzamia, 2018, Tackling persistent poverty and inequality: a dynamic perspective, in Confronting inequality. The South African crisis, M.N. Smith, Editor. Fanele: Auckland Park.

[xxv] Ravallion, M., 2020, On the origins of the idea of ending poverty, NBER Working Paper 27808. National Bureau of Economic Research: Cambridge, MA.

[xxvi] World Bank, 2020, Poverty and shared prosperity 2020.

Get the latest essay, memoir, reportage, fiction, poetry and more.

Subscribe to Griffith Review or purchase single editions here.

Griffith Review