Women are more likely than men to register to vote.
Political parties fail dismally to appeal to women voters. Some litter their manifestos with rhetoric meant to attract women but few offer a targeted and tangible package of policies that address the many problems that affect women, such as access to housing, child support grants and reproductive health services; improving primary education; and reducing gender-based violence.
This is surprising. Women are more likely than men to register to vote. At present, 14.3-million women are registered with the Electoral Commission for next year’s 2019 elections, compared with 11.7-million men. In the 2014 elections, the gender gap in actual participation was significant. Of all people who voted, 57% were women and 43% men. Given the proportionately higher levels of registration for 2019, we can expect a similar pattern next year. Women, therefore, are a well-defined constituency, with a specific policy niche. The focus on service delivery and socioeconomic rights and the movement away from traditional party loyalties by women confirm the findings of an earlier study by the Centre for Social Development in Africa as reported in the Mail & Guardian of October 19.
A recent nationally representative study on gender, culture and politics indicates that not focusing on women voters is a missed opportunity. A project of the South African Research Chairs Initiative’s (SARChI) chair in gender politics, it was the first of its kind, with a sample size of 1 300, and conducted in mid-2018 by Citizen Surveys.
Findings suggest there is a significant difference in how women and men think about and respond to politics. First, although women voters are more likely to be registered to vote, they are less likely to be partisan. In other words, they are less likely to have a strong, enduring party loyalty. So, far more women than men will be late deciders. Men, on the other hand, are more likely to go into an election with their minds already made up. Political parties will have to work that much harder to clinch the women’s vote.
Apart from the turnout gender gap, there also appears to be a partisan gender gap. When asked, “If an election was held tomorrow, which party would you vote for”, women voters as a group were slightly less likely to vote for the ANC (58%) than males (63%) but definitely far less likely to vote for the Economic Freedom Fighters (9%) than their male counterparts (16%). But women voters were more likely to vote for the Democratic Alliance (17%) compared with men (12%) and also more likely (16%) to support one of the smaller parties than men (9%).
The data also gives details of the gender distribution of votes for each party. The ANC is likely to enjoy a fairly equal distribution of its votes from both gender groups with 49% from women and 51% from men.
But the DA draws almost two-thirds of its votes from women (61%), as do the smaller parties (collectively) at 65%. The EFF, however, is the mirror opposite of the DA, with the majority of its votes coming from men (61%) and only 39% from women. The EFF and the ANC have displayed patriarchal and even misogynous tendencies at times, despite claims to support particularly marginalised women.
There was a third discernable gender gap or difference in the choice of issues used to inform the vote decision, according to the data. The same battery of variables to predict ANC vote choice (being the largest and ruling party) were tested among men and women separately. A logistic regression suggests that the only variable that predicted ANC vote choice (if an election was held tomorrow) among men is party identification (being a partisan predicts a vote for the ANC). No other variables were statistically significant.
Among women, however, a number of variables were significant. First, rural women are more likely to vote for the ANC than urban women. The government’s handling of nine critical areas (job creation, reducing crime, improving health services and education, promoting access to water and to land, service delivery protests, fighting corruption and empowering women) predicted ANC vote choice. Unsurprisingly, if women perceived the ANC government had performed badly in these areas they were less inclined to vote for the party. But being a recipient of a child support grant was positively associated with a vote for the ANC.
To understand the importance of gender policy issues, respondents were asked whether they would vote for a party that prioritised women’s rights, to which 73.8% of men and 86.2% of women said yes, indicating another gender gap.
The ANC is most likely to benefit according to these sentiments because the relationship between prioritising women’s issues and “the party you would vote for if there was an election tomorrow” shows that 87% of the respondents would vote for the ANC, 34% for the DA and 16% for the EFF.
It also suggests that the ANC takes its women support base for granted and it does not get punished by women voters when it does not deliver on women’s issues, such as decreasing gender-based violence, and increasing access to healthcare and child care.
Although the evidence shows that women make rational calculations about policy issues when they vote, it is less evident that women use the power of their votes to force their parties to respond to gender issues. This will only happen when the women’s vote becomes politicised. In other words, parties that really want to cash in on the women’s vote need to raise public consciousness about the electoral power that women possess to sway an election. And women should start to vote more strategically.
One of the problems is, of course, for delivering on policy issues for women, so the choice women have is limited.
Religion continues to be politicised and a potential source of mobilisation during elections. In the past decade, there have been politicians who have appealed to the religious sentiments of voters. In the run-up to the 2009 national elections and again in 2016, former president Jacob Zuma told supporters that the ANC would “rule until Jesus comes back”. Similarly, Zuma promised ANC supporters a place in heaven during voter registration campaigns for the 2011 municipal elections.
By evaluating how religious respondents consider themselves to be, their belief in religion and identification with specific religious denominations as well as how regularly they attend religious services, a nuanced measure of religiosity was established. In line with the findings of other studies, the data shows that an overwhelming majority (88%) of South Africans are religious but 86% of women respondents were recorded as being very religious compared with 67% of men.
Religious women are also considerably more conservative than either their religious male or nonreligious women counterparts.
But to what extent do these levels of religiosity influence political choice? The data from the study shows that religious respondents are more interested in politics that nonreligious South Africans, with religious men expressing the most interest in politics. The majority of religious South Africans, men and women, also consider themselves to be close to a political party.
When asked about their party preference if there was a national election tomorrow, the majority of men (56%) or not religious (53%) indicated that they would vote for the ANC. But women who are religious are more than twice as likely to vote for the ruling party (56%) compared with their nonreligious counterparts (20%).
By better understanding these gender differences in electoral behaviour, political parties could be better able to direct their messages to the large pool of women voters. They should bear in mind, though, that women follow politics with a more critical eye on service delivery.
Collette Schulz-Herzenberg specialises in South African voter behaviour; Cindy Lee Steenekamp focuses on political behaviour and African politics; Amanda Gouws focuses on gender politics and political behaviour. The authors are at Stellenbosch University