The Criterion Report

The Criterion Report is a quarterly market research survey conducted by the Institute of Race Relations into voter preferences, attitudes and the South African political landscape.

THE CRITERION REPORT VOL 2; No 2

17 March 2019

The Criterion Report is home to the findings and insights drawn from market research conducted by the South African Institute of Race Relations (IRR) into voter preferences, attitudes and the nature of the South African political landscape.

IRR February 2019 Election Poll: ANC Government Performance

In line with the IRR’s objective to become an authority on political market research, the Institute has undertaken its second full survey of the electoral landscape to supplement the first full survey carried out in September 2018, and a “snap poll” carried out in December 2018.

The IRR’s February Election Poll was in the field during the last two weeks of February 2019. Below is a summary of two sets of questions: First, how voters perceive the ANC national government to have performed on six core issues; second, which of the big three political parties – the ANC, DA and EFF – voters believe is best placed to deliver on the six core issues.

Part A comprises two summary tables of all the data, for ease of reference, along with the IRR’s analysis of those findings. Part B comprises the full data sets for each question.

The banner headline findings are:

1. NATIONAL ANC GOVERNMENT PERFORMANCE:

- Among all voters, the ANC national government has a negative net satisfaction score on all of six core issues. On Economic Growth, the ANC national government receives a net negative score of -56%; on Crime it scores -41%; on Education -11%; on Healthcare -13%; on Corruption -45% and on Land Reform -22%.

- The highest net dissatisfaction rate on all issues is recorded among minority voters and DA Voters. Black Voters give negative net satisfaction scores to the ANC government on Economic Growth (-48%); Crime (-31%); Corruption (-33%) and Land Reform (-7%), and marginal positive net satisfaction scores on Education (+5%) and Healthcare (+3%).

- EFF Voters are universally dissatisfied with the ANC national government performance on all six issues, but more sympathetic than DA voters: Economic Growth (DA Voters: -85%, EFF Voters: -79%), Crime (DA -74%, EFF -63%), Education (DA -53%, EFF -16%), Healthcare (DA -57%, EFF -24%), Corruption (DA -77%, EFF -60%), Land Reform (DA -67%, EFF -46%).

- ANC Voters are the most sympathetic to the ANC national government performance, but more dissatisfied than satisfied on three of the six issues: They give net negative satisfaction scores on Economic Growth (-36%); Crime (-16%); Corruption (-22%) and Land Reform (-7%), and net positive satisfaction scores on: Education (+5%); Healthcare (+3%) and Land Reform (+6%).

2. SERVICE DELIVERY PERCEPTIONS:

- Among all voters, across five of the six issues at a national level, the ANC is selected as the party best placed to deliver on that particular issue. The only exception, as a national issue, is fighting corruption, on which a majority of all voters (30.1%) stated that the EFF would do the best job.

- Significantly, in Gauteng, the DA was chosen as best placed to deliver on three of six issues, among all voters: on Economic Growth and Job Creation (34.9%), on Fighting Crime (34.7%) and on Fighting Corruption (33.1%).

- Among Black voters, the ANC emerged as the dominant choice to deliver best on all issues at national and provincial levels, bar three, where the EFF emerged with a majority: Fighting Crime, among black voters in Gauteng (36%) and Fighting Corruption, both nationally (38%) and in Gauteng (38%).

- The DA universally dominated at every level and on every issue among minority voters, all with a very high majority.

3. NATIONAL ANC GOVERNMENT PERFORMANCE vs SERVICE DELIVERY PERCEPTIONS - ANC VOTERS ONLY:

- 51% of ANC Voters are “very unsatisfied” with the ANC national government’s performance on fighting corruption. 47% of ANC Voters also believe the ANC best - out of the ANC, DA and EFF - at fighting corruption.

Methodology:

The poll was conducted between 12 and 26 February 2019. The sample was fully demographically representative and comprised only registered voters. A total of 1,611 respondents were questioned. The national margin of error is 3.3%. Supplementing this were two fully demographically representative sub-samples for Gauteng (sample size: 502 registered voters) and the Western Cape (sample size: 405 registered voters).

The margin of error for the Gauteng sub-sample was 3.8%, and for the Western Cape sub-sample, 5.9%. The confidence level is 95%. The poll was conducted telephonically, using a single frame, random digit-dialling sampling design. Briefly: The sampling frame consists of every potential cell phone number in existence in South Africa, from which a probability sample is drawn. This approach ensures that every number stands an equal chance of being included in the study, which is the most basic condition that must be met for survey results to be generalizable to the population from which a sample is drawn. The poll was conducted by Victory Research. A full explanation of the methodology follows at the end of this document.

Interpretation:

This poll is not a prediction. It is a snapshot in time, in this case of the electoral market between 12 and 26 February 2019. Likewise, the numbers presented in the poll are not absolutely definitive. A 3.3% margin of error means, for example, the DA – which comes out with 22% – could be on 18.7% or 25.3%. A confidence level of 95% means we are confident 95% of the time the findings will never vary more than 3.3 percentage points up or down from reality. When reporting on the poll, it is important to bear these parameters in mind. A number of the tables have in summary a positive or negative “Net Score”.

This is calculated by subtracting the total score of one indicator (strongly/somewhat approve) from the other (strongly/somewhat disapprove). Finally, the most valuable aspect of any poll is its ability to identify trends and patterns, particularly over time. One should thus avoid ascribing absolute authority to any given single, isolated finding.

Structure of Full Diagnostic:

This document is separated into two sections. For each question we provide the national results and the results for both the Gauteng and Western Cape sub-samples. Each question has also been disaggregated by six standards cross tabulations: “All Voters”, “Black Voters”, “Minority Voters”, “ANC Voters”, “DA Voters” and “EFF Voters”, in order to track the responses of each of these key groups to any given question both at national and provincial level. Every question has a table, followed by an overarching analysis of the findings contained in that section. Regarding the tables, the Key Findings have been presented objectively and without comment. The end-of-section analysis comprises the CRA’s political analysis of what those findings mean.

Reading the Table Headings:

The table headings are designed to allow you to discern the information you are looking at. Here is an example of a table heading:

TABLE 1.13: [National] Government Performance: Education [All Voters/Black Voters/Minority Voters]

The square parenthesis indicate whether that table has information drawn from the national sample or one of the two provincial samples. In this case, it is from the national sample: [National]. Data from the two provinces is identified by: [Provincial]. Then follows the section theme (Government Performance) and the specific indicator for that question (Education). Finally, also in square parenthesis but not in bold text are the relevant cross tabulations for that table. In this case the question has been cross tabulated by: “All Voters”, “Black Voters” and “Minority Voters”.

SUMMARY TABLE 1: Net Government Performance [All Voters/All Gauteng/Western Cape Voters]

Explanation: We asked all respondents how satisfied or unsatisfied they were with the job the ANC government is doing on six core issues, and out of 100 (0 = totally unsatisfied; 100 = totally satisfied). Respondents could indicate they were either “very satisfied”, “somewhat satisfied”, “neutral” (i.e. neither satisfied nor unsatisfied); “somewhat unsatisfied” or “very unsatisfied”. In order to summarise the individual findings for each question (all of which are contained at the end of this document), we created a “net satisfaction score” for each answer. This was done by subtracting the combined “satisfied” score (somewhat/very satisfied together) from the combined “unsatisfied” score (somewhat/very unsatisfied). If the net score is negative, it indicates more voters in that category were unsatisfied with the ANC government’s performance than satisfied. If the net score is positive, it indicates more voters in that category were satisfied with the ANC government’s performance than unsatisfied.)

Key Findings:

- Among all voters, there is no issue which does not receive a negative net satisfaction score, and all of which range between -11% (for education, which fares best) and -56% (for economic growth and job creation, which fares worst). That trend is most pronounced among minority voters, for which the highest net satisfaction rate is -66%, for education.

- Significantly, ANC voters, who are responding to the performance of a national administration under the control of the party they support, give negative net scores for three of the six issues: Economic growth (-36%), Crime (-16%) and Corruption (-22%). They give small but positive net satisfaction scores on Education (+16%), Healthcare (+13%) and Land Reform (+6%)

- Gauteng ANC voters are generally more positive about the party’s performance in national government, with the exception of crime (-20% in the province, compared to -16% nationally). On the other issues, the difference is more in the ANC government’s favour: Economic Growth (-12% versus -36%), Education (+42 versus +16%), Healthcare (+21% versus +13%), Corruption (neutral versus -22%) and Land Reform (+21% versus +6%).

Explanation: We asked all respondents which of the big three parties – the ANC, DA or EFF – they felt would do a better job on six core issues. Below is a summary of each of the individual questions (all of which can be found at the end of this document). For each category, we have listed the majority response. So, for example, among all voters nationally, on the issue of Economic Growth, 36.3% felt the ANC would do the best job out of the ANC, DA and EFF. To see the second and third place scores for this question, you would need to look at the individual question – in this case TABLE 2.1. That shows that, among all voters, after the ANC’s 36.3%, the DA came second with 29.2% of all voters and the EFF third with 22.4% of all voters. But those last two scores do not appear on the summary table – only the biggest majority score in each case is given.

Key Findings:

- Among all voters, across five of the six issues at a national level, the ANC is selected as the party best placed to deliver on that particular issue. The only exception, as a national issue, is fighting corruption, on which a majority of all voters (30.1%) stated that the EFF would do the best job.

- Significantly, in Gauteng, the DA was chosen as best placed to deliver on three of the six issues, among all voters: on Economic Growth and Job Creation (34.9%), on Fighting Crime (34.7%) and on Fighting Corruption (33.1%).

- Among Black voters, the ANC emerged as the dominant choice to deliver best on all issues at national and provincial levels, bar three, where the EFF emerged with a majority: Fighting Crime among black voters in Gauteng (36%) and Fighting Corruption, both nationally (38%) and in Gauteng (38%).

- The DA universally dominated at every level and on every issue among minority voters, all with a very high majority.

Explanation: This table constitutes the views of ONLY ANC Voters. For each of the five issues for which it is possible to compare ANC Government Performance perceptions with Service Delivery Perceptions, we have taken only those ANC Voters who indicated they were “very unsatisfied” with the ANC government’s performance and compared the result to those ANC Voters who said the ANC was best placed to deal with that issue. So, one is comparing deep dissatisfaction levels with ANC performance with the belief that the is ANC best equipped to deal with the given issue, among ANC Voters. (For each of these numbers, see the individual tables at the end of the document.)

Key Findings:

- Across the board, and despite being significantly unsatisfied with the ANC’s performance in national government on these five issues, on the same five issues, ANC voters believe the ANC best placed to deliver on them. On only one of the issues – Fighting Corruption – are more ANC voters very unsatisfied with the ANC’s performance, than believe the ANC best placed to deliver on that issue. Only on Education and Healthcare is there a significant gap between dissatisfaction levels and the belief that the ANC is best placed to deliver on that issue, in favour of the latter.

- That pattern holds for Gauteng, where a majority of ANC voters believe the ANC is best placed to deliver on those issues, with the exception of Fighting Corruption, where less than 50% of ANC voters (46%) believe the ANC is best placed to deliver on each issue.

- There are two sets of trends that are worth considering in this data set. The first is the degree to which ANC voters themselves are dissatisfied with the ANC government’s performance on these six core delivery issues, and that, of them all, even when compared to corruption, it is economic growth and job creation that elicits the greatest levels of dissatisfaction – more evidence of how powerful this issue is and that, if any political party can offer meaningful solutions to it, they will stand a chance of making a significant impact on the electorate.

- The second trend worth noting is the disjuncture between the levels of dissatisfaction between DA voters and EFF voters about the ANC government’s performance in national government. Both are damning across the board, but the EFF considerably less so: Economic Growth (DA -85%, EFF -79%), Crime (DA -74%, EFF -63%), Education (DA -53%, EFF -16%), Healthcare (DA -57%, EFF -24%), Corruption (DA -77%, EFF -60%), Land Reform (DA -67%, EFF -46%). That suggests a certain affinity among EFF voters towards the ANC which is more pronounced than among DA voters. It is a remarkable difference considering that two of those issues are corruption and land reform, around which the EFF has built its brand over the past few years.

- Satisfaction levels among EFF voters in Gauteng increase further still, which is also worth noting.

- One particular provincial rating is worth noting: in Gauteng, ANC voters give a net satisfaction score of +21% to the national government on healthcare, far above the +13% given nationally by ANC voters. Although not a national issue technically (most voters will not make a distinction between national and provincial competencies), this is a remarkable score given what happened at Life Esidimeni and the subsequent commission of inquiry into the deaths of 144 people.

ANALYSIS: Party Delivery Perceptions

- The summary table for this section illustrates two things: First, how strong the DA’s delivery brand is among minority voters and its own supporters. Second, how relatively weak the ANC’s delivery brand is nationally, among its own supporters and, particularly in Gauteng, in comparison to the DA.

- With regard to the DA: among both minority voters and DA voters, its brand as a party of delivery, on all issues, is marked not only by the universal majority it secures across the board, but by the high percentages it generates – typically in the 70% range.

- With regard to the ANC, among ANC voters, it too secures a majority across the board, even on Fighting Corruption, but the percentages are generally lower, typically in the 50% range. That suggests its brand as a party of delivery less stable.

- That is further evidenced by how small many of the majorities are that the ANC secures among all voters on all six issues nationally – all in the 30% and 40% range.

- In Gauteng, in which the ANC is on 41.6% on the provincial ballot, its numbers dip below a majority on three issues – Economic Growth and Job Creation, Fighting Crime and Fighting Corruption – all in favour of the DA, a sign that its brand as a party of delivery in that province is particularly shaky. Of the three issues it loses out on to the DA in that province, most remarkable is Economic Growth and Job Creation. Fighting Crime and Fighting Corruption are mainstays of the DA’s opposition brand, and it makes sense that the DA would gain a majority on those first. But Economic Growth and Job Creation is a very difficult issue to wrest control of from a governing party. This appears to be symptomatic both of how critical that issue is to voters and how profound dissatisfaction levels are across the board with the ANC’s performance on it.

ANALYSIS: Delivery Satisfaction versus Delivery Perceptions: ANC Voters

- A significant number of ANC Voters seem to be able to hold two contradictory views simultaneously: First, a deep unhappiness with the ANC national government performance and, second, the belief that the ANC government is best placed to deliver on that particular issue.

- Some of this might be explained by the election of Cyril Ramaphosa. That is, there is a belief that, over the last decade, the ANC has failed to deliver but that, in Ramaphosa, the ANC is now best placed to deliver reform. However, it is unlikely that explains this seeming contradiction entirely, as the broad structure of the ANC and its leadership at national and provincial level remains unchanged. There would appear to be a well-set cognitive dissonance in place among a significant number of ANC voters.

Gareth van Onselen

Head of Politics: IRR

The PDF of the report, including additional tables, can be accessed here. 

Issued by the IRR, 17 March 2019