Good morning everyone.
Introduction
I am honoured to launch the findings of the fourth Wave of the National Income Dynamic Study (NIDS) today. As many of you are aware, in 2012, government adopted the National Development Plan which sets out the long-term path for our country with the objectives of eradicating poverty and reducing inequality and unemployment. The NDP is built on the progress that was made since the transition from the unequal apartheid system to democracy through targeted government policies. It is important that we evaluate whether these policies are effective and to identify the gaps and the measures to address them. To do this, government requires data sources that provide empirical evidence.
In this regard as Government, we initiated a study to provide us with an enhanced understanding of the socio-economic dynamics of South Africa. In 2006, the Presidency commissioned the Southern African Labour and Development Research Unit (SALDRU) based at the University of Cape Town to undertake a panel study - the first National Income Dynamics Study (NIDS). At the time South Africa joined developed countries such as the UK and the US and developing countries such as Mexico and Indonesia in having a national panel survey.
When it was initiated, it was apparent that South African society was undergoing rapid economic, political and social change, and government needed to have a better understanding of these dynamics, in terms of mobility, poverty and the household. The aim of the National Income Dynamic Study is to provide empirical data on the changing dynamics of South African society.
The study tells us a comprehensive story about what is happening to all South Africans across the income distribution spectrum, from rich to poor. It provides us with empirical evidence of the unfolding story of the changes, both positive and negative, in the lives of all South Africans. It is important that we recognise that the NIDS is South Africa’s only national panel or longitudinal study of individuals of all ages. It is a national research resource which is an example of government and academia working together through the partnership with SALDRU at UCT as the Implementing Agency.
The NIDS is a unique dataset in that it provides information about the same set of households across time, that is, from wave 1 in 2008 to wave 4 in 2014/15. The study is undertaken at two to three year intervals with the first or the base wave undertaken in 2008.
This means we have eight years of data about our respondents including the changes or the dynamics in their life circumstances. For example, we collected data on children’s education status; as well as data on their parents’ level of education. Using this information as the longitudinal study unfolds from one wave to the next, it reveals the effect of this on the structure of households in South Africa, including on the living conditions and the well-being of household members. As government, we are therefore able to use the NIDS data to improve the evidence-base for policy analysis and policy-making so that we design policies that have an impact for the better on peoples’ lives. In this way, the NIDS data will be able to tell us who is getting ahead and who is falling behind over time and, most importantly, the reasons for this.
I’ve mentioned that this study is the only longitudinal panel in the country, and permit me to digress for a few minutes to explain the importance of this aspect of the study. As government we use a wide range of data sources to measure the impact of our programme of action including both regular and ad hoc survey data mainly undertaken by Statistics South Africa (Stats SA). A panel study is different to a cross-sectional study, such as Stats SA’s Income and Expenditure Survey (IES) or General Household Survey, in that while those surveys provide information of households at a particular point in time, the panel or longitudinal study provides information about the same households over time. Longitudinal data allow for analysis and the measurement of differences from one period to another, that is, we can use the information to locate the causes of social phenomena and identify connections between events that are widely separated in time.
In short, panel data provide more detailed information and offer a more precise measure of both the direction and the magnitude of change at the individual level. Importantly this household panel data can contribute to identifying the features that define poverty and the factors that influence the transmission of income and wealth.
Main findings
Poverty
So, what have we learnt from the NIDS: The main findings of the NIDS Waves 1 to 4 is that with regard to poverty, of those who were poor in Wave 1, almost three quarters remained poor in Wave 4. We can conclude that these people are effectively trapped in poverty and will require policy interventions directed at addressing the structural issues in the economy. Then there are a number of households whose fortunes change and they either move into or out of poverty. These households will require different policy interventions that focus on these transitory periods to support them when they fall into poverty.
When we examine these transitions more closely using the data, the findings show that changes to household composition is the largest trigger of poverty entry and exit, and that increasingly the income from government grants acts as the main trigger for an exit from poverty for about one quarter of the NIDS sample. The analysis shows that access to the labour market within the household is the single most important determinant of poverty entry and exit. Again, with panel data it is possible to understand the impact of family size on consequent income or expenditure dynamics.
Inequality
The nature of the dataset also provides information on the education levels and earnings of both parents as well as their adult children. It allows us to record mobility across generations to determine, for instance, whether children are better-off compared to their parents. What we have found is that education and earnings mobility follow two different paths. While there has been an increase in the years of schooling, this unfortunately, has not translated into getting better jobs. The correlation between the earnings of parents and their adult children in South Africa is high in general, but is particularly strong for low-wage earners and very high earners. This means that the advantage of having a more educated younger generation has not yet led to a reduction in earnings inequality. This lack of intergenerational mobility in earnings is an important element in our understanding of the persistence of inequality in South Africa.
Middle Class
With regard to vulnerability and the middle class, the NIDS finds that the middle class in South Africa is relatively small. The study defined middle class by using a monthly per capita expenditure range of R2 920 to R10 678 based on January 2015 prices. Using this threshold the middle class in South Africa is only about 15 per cent of the total population. There has, however, been considerable demographic transformation within that band of the middle class, with Africans now outnumbering whites by about 2 to 1. However, this must be understood in the context that Africans are still underrepresented in the middle class in general, taking into account the overall population dynamics.
Youth Employment
In respect of youth employment, the NIDS finds that individuals who were most consistently employed across the waves also were also those who had the highest level of education. Two-thirds of those who found employment between Wave 3 and Wave 4 remained employed. The use of networks was the dominant method of securing this employment. Almost a third (32%) found their main regular job in the Wholesale and Retail sector, and 24 % found a job in the community, social and personal services sector.
The lessons from previous NIDS waves have been no less insightful. For example, NIDS Wave 3 revealed that when educational expenditure increased, the no-fee school policy alleviated the circumstances of poor households by decreasing educational spending for these households. It also found that the recipients of the child support grant (CSG) aged 15-19 years old are 6% more likely to be enrolled in school than non-recipients. These findings point to the positive impact of government interventions including social grants to ensuring educational access.
NIDS Wave 3 reminded us that migration remains a feature of the post-apartheid landscape. It told us that South Africans are making choices about their lives, choices about moving or staying: one in seven people in the sample population moved. This was mainly to look for employment and to improve their lives and the lives of their children.
As a national resource, it is important that we extract as much value as possible from the NIDS and its very rich data. And because we are mindful of the capacity constraints in respect of quantitative analysis in South Africa, the DPME through the NIDS contract with SALDRU has provided nine scholarships to post-graduate students in NIDS Wave 4. In fact, since 2006 we have awarded 55 scholarships, at honours, masters, doctoral and post-doctoral level. In addition, the DPME has collaborated with SALDRU, thorough our Programme to Support Pro-poor Policy Development to train 196 government officials and academics in advanced panel analysis.
As I conclude, I am pleased to inform you to inform you that the NIDS dataset is freely available, and I want to appeal to policy-makers to ask the pertinent questions and for the academic community to be responsive in utilising the NIDS data. This would ensure that the government’s investment is maximised, and in this way the research evidence can inform our policy choices so that we achieve our goals of eradicating poverty and reducing inequality.
Thank you 
