Social grants to reduce poverty feature prominently in the campaign promises of political parties in South Africa's 2024 national and provincial general elections, set for 29 May.
The country's social grants system is one of the largest in Africa in terms of number of beneficiaries. Research shows that this has helped reduce poverty. About 26 million to 28 million social grants have been paid every month to children, older persons, people with disabilities and the unemployed. The country's population is 62 million.
While some political parties propose expanding the grants system, others propose increasing the amounts. On the campaign trail, President Cyril Ramaphosa hinted at the possibility of a loss of grants if another party came to power. This was condemned by some parties and commentators as "fearmongering" and taking advantage of voter fears about their livelihoods, especially among rural voters.
But do grant recipients reward political parties with votes?
We have been studying voting behaviour since 2017 using a nationally representative sample. In five surveys, the latest in 2023 - Factors determining voter choice in South Africa's 2024 national general elections - we have asked respondents about the reasons for their political party choices. The research was coauthored with our colleague Jaclyn de Klerk, a statistician.
Across all five surveys, we found that receiving a grant per se did not influence voters' choice of which party to support. What we found in the four waves of the research from 2017 to 2020 was that grant recipients would be more likely to vote for the governing ANC than the opposition party if they feared losing their grants.
But in the 2023 survey we saw a change - the voters didn't feel this way anymore. Our hypothesis is that this was due to the fact that grants were paid to younger people after an expansion in payouts during COVID. Allegiance to the ANC was not part of this cohort's makeup.
Grants and voting
South Africa's social grants system for children, older persons and people with disabilities expanded significantly from 7.2% of the population in 1995 to 47% in 2022, in keeping with the constitutional right to social assistance.
The latest survey (2023) was prompted by the introduction of the Social Relief of Distress grant in 2020 to mitigate the financial and employment impacts of the coronavirus pandemic on vulnerable individuals and households. The grant incorporated new groups of welfare recipients - "welfare outsiders" such as youth and unemployed adults who were not previously receiving social grants. The grant has been extended annually until March 2025.
There is limited empirical evidence of what the effects of social grants are likely to be on voting, especially by the new recipients - the voting adults and youth.
Despite the reach of the grants, the monetary values are small, often below the food poverty line, which limits their impact.
Our surveys from 2017 to 2020 did not find that receiving a grant influenced support for the incumbent African National Congress (ANC). What mattered across all four successive surveys in 2017, 2018, 2019 and 2020 was the fear of loss of a grant if the incumbent party lost the election. We found grant recipients would be more likely to vote for the ANC than an opposition party if they feared losing their grant.
Given the highly competitive nature of the 2024 election - different polls show that the governing ANC is unlikely to win a clear majority needed to form a government on its own - we were particularly interested to know whether receiving a grant and fear of loss of a grant were likely to be predictors of party choice.
Also, whether our previous findings would hold true in a context of grant expansion since 2020, and because of the extraordinarily high unemployment rates, persistent poverty, poor service delivery and a stagnant economy.
The latest study
We analysed a nationally representative random sample of 3,511 respondents. The data was collected by the public opinion surveys company Ipsos Public Affairs using face-to-face interviews conducted in people's home languages. The questions included why the respondent would choose a particular political party, with 14 optional responses.
Additional questions included:
- fears of loss of a grant if the person voted for another party
- level of trust in institutions such as the presidency, parliament and in the ANC government
- type of grant received.
Do grants matter?
The most interesting change in 2023 over the previous four waves from 2017 to 2020 was the fact that the fear of losing a grant if a person voted for an opposition party was no longer a significant factor in voter choices. It was the first such finding since the study commenced in 2017.
This was the case despite the expansion of the social grants, and the introduction of the social relief of distress grant.
In fact, the expanded base of grant recipients may be the reason these findings differ from previous surveys.
Before the onset of the COVID-19 pandemic, the largest number of grant beneficiaries were women receiving the child support grant and older persons receiving an old age pension.
Now, many grant recipients are young and unemployed and may have different party loyalties to older generations. Further, the fear that recipients would lose their grants if they voted for an opposition party has abated considerably compared to our earlier surveys, where this factor was highly significant.
More telling was that the majority of grant recipients in the latest survey were more likely to vote for an opposition party (53%) than for the ANC (47%). This is a major shift compared to 2020 when only 26% of grant recipients selected an opposition party.
It may be that the expansion of the grants since 2020 has created a sense of comfort and security among recipients about their right to social assistance. And the fear of loss of a grant if they voted for the opposition was not a concern to the same extent as before.
Since the research began in 2017, more opposition parties such as the Democratic Alliance, the Economic Freedom Fighters, and new parties such asAction SA and Rise Mzansi have endorsed social grants as important for poverty reduction.
The change in the public discourse away from the negative effects of grants on work seeking may be filtering through to grassroots voters, leaving them feeling less stigmatised. Previously there were claims that grants created dependence on the state and increased childbearing. Now they are seen more as reducing poverty.
Socio-economic issues remained uppermost in the minds of the respondents in the study. The top four reasons for their party choice were that the party would create jobs (55%), improve people's lives (49%), improve service delivery (48%) and pay social grants (44%).
Grants are only one set of factors that influence voter choice. Participants' views about governance, corruption, party loyalty, party leadership and trust in the presidency of Ramaphosa also held sway.
We conclude that a combination of other factors will drive voter decisions in the upcoming election.
Leila Patel, Professor of Social Development Studies, University of Johannesburg
Yolanda Sadie, Emeritus professor in Politics, University of Johannesburg