Research

Journal Publications

The labor market and poverty impacts of covid-19 in South Africa South African Journal of Economics, (2023)
(joint with Ihsaan Bassier, Joshua Budlender, and Ronak Jain)

Abstract We use newly-released South African data to present the first estimates of COVID-19-related employment and poverty impacts in a developing country. We observe a 40% decline in active employment. Half of this comprises job terminations, suggesting persistent labor market effects. Initially vulnerable groups are disproportionately affected. Exploiting the dataset's panel dimension and comparing lockdown incomes of job losers to re-weighted job retainers, we estimate that 20-33% of job losers fall into poverty. Only 20% of those temporarily not working received the intended relief, while a third of job losers had no access to any major form of social protection.

A poverty dynamics approach to social stratification: The South African Case World Development, Volume 110 (2018)
(joint with Simone Schotte and Murray Leibbrandt)

Abstract The wave of upbeat stories on the developing world’s emerging middle class has reinvigorated a debate on how social class in general and the middle class in particular ought to be defined and measured. In the economics literature, most scholars agree that being middle class entails being free from poverty, which means being able to afford the basic things in life – not only today, but also tomorrow. In consequence, there is an increasing tendency to define the middle class based on a lack of vulnerability to poverty. In this paper, we strengthen and expand on these existing approaches in three ways: First, we incorporate the differentiation between the middle class and a (non-poor) vulnerable group into a broader social-stratification schema that additionally differentiates between transient and chronic poverty. Second, in estimating the risk of poverty, we employ a multivariate regression model that explicitly allows for possible feedback effects from past poverty experiences and accounts for the potential endogeneity of initial conditions, unobserved heterogeneity, and non-random panel attrition – four factors insufficiently addressed in existing studies. Third, we highlight the value of paying attention to these conceptual and modelling issues by showing that class divisions based on monetary thresholds inadequately capture a household’s chances of upward and downward mobility. We then apply our conceptual framework to the South African case. We find that only one in four South Africans can be considered stably middle class or elite. Access to stable labor market income is a key determinant of achieving economic stability. A lack of jobs as well as the prevalence of precarious forms of work drive high levels of vulnerability, which in turn constrains the development of an emergent middle class – not only in South Africa but potentially also in other parts of the developing world that face similar labor market challenges.

The livelihood impacts of COVID-19 in urban South Africa: A view from below. Social Indicators Research, Volume 165, 1–30 (2023)
(joint with Simone Schotte)

Abstract This paper investigates the impact of the COVID-19 pandemic and related policy measures on livelihoods in urban South Africa. Using qualitative research methods, we analyse two rounds of semi-structured phone interviews, conducted between June and September 2020 in the township of Khayelitsha, Cape Town. We contextualise these by presenting a snapshot of the nationwide dynamics using quantitative panel data. Our findings describe how the shock of the COVID-19 pandemic has deepened the economic vulnerability which preceded the crisis. Survivalist livelihood strategies were undermined by the economic disruption to the informal sector, while the co-variate nature of the shock rendered social networks and informal insurance mechanisms ineffective, causing households to liquidate savings, default on insurance payments, and deepen their reliance on government grants. In addition, the impact of the pandemic on schooling may deepen existing inequalities and constrain future upward mobility.

Snakes and Ladders and Loaded Dice: Poverty Dynamics and Inequality in South Africa South African Journal of Economics, Volume 90, Issue 2, 2022
(joint with Simone Schotte and Murray Leibbrandt)

Abstract Longitudinal surveys allow us to understand how markers of (dis)advantage determine present material welfare and economic upward or downward mobility over time. In this paper, we use five waves of panel data to empirically assess the extent and dynamics of poverty in South Africa between 2008 and 2017. Investigating the correlates of poverty entries and exits, we analyse how multidimensional inequalities in terms of household- and individual-level characteristics relate to these dynamics and identify markers of vulnerability. We utilise these markers to classify the South African population into five strata characterised by their present and future risk to poverty.

Locked down and locked out: Repurposing social assistance as emergency relief to informal workers World Development, Volume 139, 2021
(joint with Ihsaan Bassier, Joshua Budlender, Murray Leibbrandt and Vimal Ranchhod)

Abstract The COVID-19 pandemic presents a particular challenge to countries with high levels of labour market informality. Informal workers and their households are especially vulnerable to the negative economic consequences of the pandemic and associated lockdown measures, while the very fact of their informality makes it difficult for governments to quickly provide targeted economic relief. Using South Africa as a case study, we examine how an established social assistance system – not originally designed to support informal workers – can be re-purposed to provide emergency relief to these workers and their households. We examine how expansions of this system on the intensive margin (increasing the value of existing social grants) and extensive margin (introducing a new feasibly-implemented grant) can be used to mitigate this COVID-19-associated poverty. We compare the efficacy of the different policies by using pre-pandemic nationally representative household survey data to project how a negative shock to informal incomes can be mitigated by the different social grant measures, with a particular emphasis on poverty impacts. We find that an intensive-margin expansion of the existing Child Support Grant is complementary to the extensive-margin introduction of a new Special COVID-19 Grant, and that this combined policy intervention performs best out of the options considered. However conclusions as to this “optimal policy” are not simple technical determinations. We show that these conclusions are in fact sensitive to both unavoidable technical assumptions about how resources are consumed and shared within the household, as well as to normative value judgments about which populations to prioritise and how to value poverty reduction spillovers amongst the non-targeted group. While our approach helps identify a range of sensible policy approaches, there is no escaping the limits to our knowledge or the issue of normative goals – a finding likely applicable to a broad range of empirical policy analyses.

Are We Really Painting the Devil on the Walls? Polarization and its Drivers in Sub-Saharan Africa in the Past Two Decades Journal of African Economies, Volume 31, Issue 2, 2022
(joint with Vasco Molini, Michele Fabiani and Fabio Clementi)

Abstract The development path of Sub-Saharan Africa (SSA) over the past two decades has been characterized by sluggish poverty reduction occurring alongside robust economic growth. Applying polarization measures to comparable survey data from 24 SSA countries, we find that there has been a generalizable increase in polarization over the past two decades—and in particular, an increased concentration of households in the lower tail of the welfare distribution of SSA countries. The polarization process is further analyzed by identifying the main drivers and singling out the effect of different covariates at different points in the consumption distribution. This investigation reveals that the drivers of polarization are relatively similar across SSA: demographic, urban/rural, regional variables and access to basic infrastructure are found to be the most important drivers of polarization in many countries.

Is employment a panacea for poverty: A mixed-methods investigation into employment decisions in South Africa World Development, Volume 130, 2020 \

Abstract Unemployment is a key determinant of poverty in South Africa and labour market inequalities reflect deep-rooted socio-economic inequalities. In a context of high rates of poverty and unemployment, we would expect a job loss to be associated with a decline in wellbeing. Using nationally representative panel data and original qualitative data collected in Cape Town, I find that, on average, this hypothesis holds. However, this aggregate effect conceals heterogeneities in the relationship between labour market transitions and wellbeing which are of special analytic interest. In particular, this study focusses on those cases which go against the grain of the overall labour market-wellbeing nexus – that is, cases in which black, urban youths turn down or quit wage work. An analysis of these examples helps illuminate how disadvantaged workers face non-negligible disincentives to certain forms of low-skill employment and reveals the circumstances under which these disincentives may outweigh the disincentives to unemployment. To aid this investigation, I develop a model which analyses the welfare effect of job losses as being jointly determined by the strength of outside options and disincentives to work. Using qualitative data, I provide evidence in support of this model and show that, under certain circumstances, transitioning out of employment will be the welfare optimising choice for workers: Younger workers with no dependants and with alternative sources of support can be said to have stronger outside options, and are especially likely to turn down or quit low-quality jobs. Older workers, with dependants and without alternative sources of support, are more likely to accept and persist in low-quality jobs. This study argues that understanding the complexity of the incentives that workers face and which inform labour market choices will be indispensable in effectively designing policies which aim to reduce inequalities in labour markets – in South Africa and beyond.

Book chapters

Tackling persistent poverty and inequality: A dynamic perspective In Confronting Inequality: The South African Crisis, 2019
edited by M.S. Smith. Jacana Media
(joint with Simone Schotte and Murray Leibbrandt)

Selected working Papers

Earnings inequality over the life-course in South Africa. AFD Research Paper 160
(joint with Vimal Ranchhod)

Abstract Earnings inequality is usually calculated from a distribution which is measured at a point in time. However, because we typically observe a positive age-earnings profile, a part of cross-sectional inequality is explained by age-related differences in earnings across age cohorts. When inequality is computed using earnings measured over the lifetime, these age-specific differences are averaged out. However, there are also factors that may drive up inequality in earnings measured over time relative to cross-sectional inequality – for instance, low cross-sectional earnings are likely to be correlated with low wage growth and longer spells of unemployment, thereby compounding inequality. Using South African data, we investigate how these dynamic processes act simultaneously but over different time scales to both moderate and exacerbate inequality over time. Because the available panel data in South Africa spans only nine years, straightforwardly constructing a measure of lifetime earnings is not possible. We circumnavigate this challenge by constructing a synthetic lifetime panel by stitching together relevantly similar individuals across successive age cohorts. We use this synthetic panel to compute inequality of lifetime earnings and compare this to inequality of earnings measured over the medium-term (2-9 years), and to inequality measured at a point in time. We find that inequality of lifetime earnings, which reflects the effect of the age/earning relationship, is lower than inequality of contemporaneous earnings. However, inequality of earnings measured over two to nine years, which is more sensitive to inequalities in short-term employment dynamics, is substantially higher than point-in-time estimates.