Reach the People You Need to Know. SOUTH AFRICAN CITIZENS SURVEY CORE REPORT – QUARTER 4 2018 TABLE of CONTENTS

Reach the people you need to know. SOUTH AFRICAN CITIZENS SURVEY CORE REPORT – QUARTER 4 2018 TABLE OF CONTENTS

SECTION PAGE

ONE: Background and Sample 4

TWO: Key Findings 8

THREE: Direction of the Country and Economic Perceptions 12

FOUR: Political Parties and Key Leaders 16

FIVE: Political Leaders’ Election Campaign Activity 29

SIX: The Electoral Commission, Voter Registration, and Estimated Voter Turnout 35

SEVEN: Most Important Problems Facing the Country 52

EIGHT: Trust in Institutions 56

THE SURVEY THE SAMPLE

 Citizen Surveys has been carrying out the South African Citizens Survey  Designed as a complex, multi-stage stratified probability sample, based on (SACS) since July 2015. the Census EA sample frame.

 Face-to-face interviews are conducted on computer tablets with a  Sampling sites are chosen at random across all provinces and metro, nationally-representative sample of 1,300 South African adults (i.e. 18 urban and rural areas, with probability proportionate to population size, years of age and older) per month. based on the latest Stats SA estimates of the population aged 18+.

 Interviews are conducted in English, Zulu, Xhosa, Afrikaans, Sotho, Sepedi,  Results are reported quarterly on a total of 3,900 respondents, which and Setswana. produces results with a margin of error margin of error of ±1.5 percentage points at the 95% confidence level.  The SACS covers a range of current issues. Citizen Surveys invites researchers from universities and non-profit organizations to access the  Weights are applied to ensure the sample represents the most recent data free of charge – subject to usage conditions – by visiting our website national population with respect to province, race, gender, age, and area. (https://citizensurveys.net/sa-citizens-survey and clicking on Request Our Data), or contacting us on +27 (0)21 447 4484.

1,300 INTERVIEWS PER MONTH 3,900 INTERVIEWS PER QUARTER

 Over the course of the South African Citizens Survey, just over 11,000 Enumerator Areas were covered in order to ensure that all South Africans – from the cities, towns, villages and remotest regions of South Africa – had the opportunity to have their voices heard.

 Alongside is a map of all the Enumeration Areas where just over 50,000 interviews have been conducted over the past three years,.

 This coverage, combined with our scientific sampling and rigorous quality controls, means that the data used for this report represents all South Africans, and can be relied upon for crucial and strategic decisions.

 Citizen Surveys, founded in 1996, operates both nationally and internationally and has a broad selection of clients including international corporates, public institutions, government departments, academic institutions, NGOs, and development agencies.

 We specialize in research that has a social impact, and in large-scale national household surveys, continuous tracking studies, customer satisfaction surveys and segmentation modelling.

 Our robust, scientific methods and quality management systems deliver superior quality information and insights that can be trusted.

 Most of our business stems from repeat business and client recommendations.

 Some of our clients include:

 During Q4 of 2018, the proportion of South Africans who believed the  Levels of trust in state institutions grew slightly during this quarter, country is headed in the right direction dropped from 37% to 28%. although the overall trend remains one of decreasing trust.

 Perceptions of a weakening economy and deteriorating personal living  Trust in the Electoral Commission grew the most (+11%) to 59%. conditions continued during this quarter.  Trust in SARS dropped from a high of 64% in Q2 of 2017 to its current

 Only one-in-four (24%) of South African adults reported an level of 56%. This is most likely due to the reputational damage SARS improvement in their personal living conditions. has suffered.

 Similarly, less than one-in-five (17%) believed the economy had  The Public Protector (53%) and the National Prosecuting Authority improved over the past 12 months. (50%) have also seen strong declines in public trust over time, as has the South African Police Service (42%).  Despite this, many South Africans remain hopeful about the future: just over half (58%) believed that their livelihoods would improve over the next  Perceptions around the state of the economy and prospects of an 12 months. economic recovery in addition to continuing revelations of corruption, looting and state capture continue to undermine the confidence of South  South Africans are less optimistic about the economy, however, with Africans in the political process and contribute to the belief that the just under half (48%) of adults believing it will improve in the next 12 country is heading in the wrong direction. months.

 Both of these indicators have also steadily declined over time, in line  The decline in the credibility of the pillars of South African society, coupled with the global downturn continuing to take its toll. with the dynamics that are threatening to overshadow the achievements of our democracy, are more than just evidence of a worrying trend that does not bode well for the future of our country.

 President Cyril Ramaphosa enjoyed job performance ratings of 68% in April, May, and June. His rating has since declined, dropping to 60% in Q4 of 2018. Despite this, Cyril Ramaphosa remains the most popular political party leader polled over the past 12 months.

 The biggest challenge for Ramaphosa is whether he will be able to manage this economic crisis and retain the public’s confidence into the 2019 elections, as much more time may be needed to fix the deep-rooted problems that face him.

 The favourability score of Mmusi Maimane, the opposition DA leader, has been declining gradually over time, to just 29% in the last quarter of 2018 .

 In contrast to the other major party leaders, Julius Malema’s popularity– has grown steadily over time, despite his populist stance and polarising nature. In Q4 of 2018, Julius Malema’s favourability rating at 30% was on a par with that of Musi Maimane.

 The Q4 of 2018 results reflect a further decline in favourability towards the ANC, as the proportion of South Africans who “like” the party dropped from 64% in the second quarter of 2018 to 56% in the last quarter.

 Those who prefer the ANC to other political parties remained at 44%.

 The ANC, however, continues to benefit from Cyril Ramaphosa being at the helm of the party and the country. Over the past 12 months, support and preference for the ANC has seen a net growth from 43% to 56% and from 37% to 44%, respectively. INSERT PICTURE(S) HERE  Those South Africans who feel close to the DA have remained largely unchanged over time, at 7%, while those who like the DA has remained between 30% and 34%.

 Favourability towards the EFF has remained flat, at 29%. Those who prefer the party above others dropped slightly to 10%, while its partisanship score (South Africans who feel close to the EFF) declined to 5% in Q4 of 2018.

 During the first half of 2018, belief that the DIRECTION OF THE COUNTRY WOULD YOU SAY THE COUNTRY IS GOING IN THE RIGHT DIRECTION OR IN THE WRONG DIRECTION ? country was headed in the right direction grew noticeably, to 34% in Q1 and 37% in Q2. Economy enters  The technical economic recession announced Cyril Ramaphosa technical becomes Thuli recession & in the previous quarter, along with the president Madonsela’s Commissions of continuing revelations of state looting and State Capture Enquiry into Report released State Capture state capture has eroded some of the Finance Minister previous optimism. Pravin Gordhan 41% dismissed 39% 38%  In Q3 of 2018, belief that the country was 36% 37% 34% 34% headed in the right direction dropped to RIGHT 28% 28%, and during this quarter to 27%. DIRECTION 27% 23% 23% 29%

19%

Q4 2015 Q1 2016 Q2 2016 Q3 2016 Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

 We asked South Africans whether they PAST ECONOMIC PERFORMANCE THE ECONOMY HAS IMPROVED OVER THE PAST 12 MONTHS believed the economy had improved or deteriorated over the past 12 months, and whether the economy would improve in the next 12 months or not. 24% 25% 21% 22% 19% 19% 20% 20% 17%  Fewer than one-in-five (17%) South Africans felt that the economy had improved during Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 this quarter. This continues the downward trend seen since Q2 of 2018.

 There is much greater optimism about the FUTURE ECONOMIC PERFORMANCE economy improving: during this quarter, just THE ECONOMY WILL IMPROVE OVER THE NEXT 12 MONTHS under half (47%) of all South African adults 61% believed that the economy would improve. 58% 52% 42% 46% 47%  Like the past performance, however, this 35% 36% 34% perceived future performance has also steadily declined over time, from a high of

61% in Q1 of 2018. Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

 We also asked South Africans whether they PAST LIVING CONDITIONS MY PERSONAL LIVING CONDITIONS HAVE IMPROVED OVER THE PAST 12 MONTHS believed that their personal living conditions had improved or deteriorated over the past year, and whether these would improve in

the future. 28% 26% 25% 23% 25% 28% 26% 26% 24%  Roughly one-in-four (24%) South African adults felt that their personal living Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 conditions had improved in the past 12 months – a slight decline from the previous quarter’s result. FUTURE LIVING CONDITIONS  Looking forward, just under six-in-ten South MY PERSONAL LIVING CONDITIONS WILL IMPROVE OVER THE NEXT 12 MONTHS Africans (58%) believe that their personal 67% living conditions will improve over the next 12 57% 62% 60% 58% 54% 45% months. 41% 45%

 Much like the views on the future of the economy, however, this trend has been

declining steadily over time. Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

 The President of South Africa’s job approval rating jumped to 64% when Cyril Ramaphosa took the helm in March 2018, and maintained monthly highs of 68% in Q2 of 2018 (April, May and June). Since then, the President’s approval rating has declined slightly on aggregate (despite increases in September and Nov 2018). There remains optimism – as seen previously – but that optimism is being tempered by the political realities.

PERFORMANCE OF THE PRESIDENT THE PRESIDENT IS DOING HIS JOB WELL / VERY WELL 68% 68% 68% 65% 64% 64% 61% 62% 59% 60%

46% 41% 41% 39% 40% 40% 39% 37% 35% 36% 36% 33% 35% 33% 33% 32% 33% 30% 27% 24% 25% 24% 22% 23% 23% 22% 22%

Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 2015 2016 2016 2016 2016 2016 2016 2016 2016 2016 2016 2016 2016 2017 2017 2017 2017 2017 2017 2017 2017 2017 2017 2017 2017 2018 2018 2018 2018 2018 2018 2018 2018 2018 2018 2018 2018

JACOB ZUMA | CYRIL RAMAPHOSA

Original data and written content: © 2018 Citizen Surveys. All rights reserved. 17 DEMOGRAPHIC PROFILE: Q4 2018 SOUTH AFRICAN ADULTS ( 18+ ) WHO BELIEVE THAT THE PRESIDENT IS DOING HIS JOB “WELL / VERY WELL”

POPULATION GROUP LIVING STANDARDS MEASURE TOTAL SOUTH AFRICANS LM BLACK 10 1,344,000 (6%) 22.43-million GT 17,836,000 (80%) 2,551,000 (11%) 7,162,000 (32%) 09 2,007,000 (9%) COLOURED 61% 1,798,000 (8%) 08 1,716,000 (8%) MP INDIAN NW 1,891,000 (8%) 2,892,000 (13%) 575,000 (3%) 07 1,729,000 (8%) 7,476,000 (33%) WHITE 06 2,222,000 (10%) FS KZ 3,945,000 (18%) 05 NC 1,184,000 (5%) 2,390,000 (11%) 441,000 (2%) AGE GROUP 04 2,143,000 (10%)

18-24 4,708,000 (21%) 03 545,000 (2%) EC 02 281,000 (1%) 25-34 6,093,000 (27%) 2,449,000 (11%) WC 01 82,000 (0%) GEOTYPE 2,634,000 (12%) METRO 35-44 4,639,000 (21%) 9,794,000 (44%) GENDER

FEMALE URBAN 45-54 3,060,000 (14%) 11,100,000 (49%) 4,198,000 (19%)

MALE RURAL 55+ 3,930,000 (18%) 11,331,000 (51%) 8,438,000 (38%)

 Cyril Ramaphosa is still ranked as the most AFFINITY TOWARDS POLITICAL LEADERS DO YOU HAVE A FAVOURABLE OR UNFAVOURABLE OPINION OF … ? popular political leader by a significant margin, despite a decline in his favourability rating from 64% in Q2 to 55% in Q4 of 2018. 64% 60%  Mmusi Maimane’s favourability score stood 56% 55% at 29% in this quarter and has remained relatively flat over the past 18 months. 46% 47% 44%  Julius Malema’s popularity continues to grow 38% 38% 35% and stood at 30% in Q4 of 2018. 37% 37% 35% 35% 35% 33% 31% 31% 32% 32% 31% 29% 30% 35% 34% 34% 34% 28% 32% 30% 31% 29% 22% 30% 29% 29% 27% 28% 25% 25% 25% 22% 22% 21% 20% 19% 19% 16%

Q4 2015 Q1 2016 Q2 2016 Q3 2016 Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

JACOB ZUMA CYRIL RAMAPHOSA MMUSI MAIMANE JULIUS MALEMA

 We use three measures to gauge political party affinity: feeling close to a FEEL CLOSE TO THE PARTY party (partisanship), likeability, and preference. Likeability is rated from 0 39% 38% 32% 35% 33% to 10, where 0 means a strong dislike and 10 means a strong liking. ‘Like 8% 7% 7% 7% 7% the party’ is calculated as those who score the party at 6 or higher. ‘Prefer 4% 5% 5% 7% 5% the party above others’ is calculated as the party which obtains the Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 highest rating in relation to the other political parties.

 The proportion of South Africans who feel close to the ANC decreased LIKE THE PARTY 58% 64% 58% slightly to 33% during this quarter, while the DA’s partisanship level 50% 52% 53% 56% 42% 43% remained flat, at 7%, and the EFF’s partisanship remains around 5%. 34% 30% 31% 29% 29% 32% 32% 31% 34%  Favourability towards the ANC dropped from 64% in Q2 to 56% during 31% 29% 23% 29% 16% 15% 19% 20% 23% this quarter, while the DA and the EFF remained at around 31% and 29% Q4 2015 Q1 2016 Q2 2016 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 respectively . PREFER THE PARTY ABOVE OTHERS  Preference for each of the three major political parties remained relatively 50% unchanged at 44% for the ANC, 13% for the DA and 10% for the EFF . 44% 45% 44% 47% 43% 44% 36% 37%

21% 18% 18% 19% 19% 12% 13% 16% 12% 7% 8% 12% 9% 11% 9% 9% 11% 10% Q4 2015 Q1 2016 Q2 2016 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

 The proportion of South African adults who feel close to the ANC declined FEEL CLOSE TO THE PARTY from 39% in Q1 of 2018 to 33% in the current quarter. 39% 38% 32% 35% 33%  While the party’s likeability rating jumped to 64% in Q2 of 2018, it dropped 33% Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 off to 58% in Q3 and to 56% in Q4.

 A combined measure of those who both like and prefer the ANC above all LIKE THE PARTY

other political parties grew slightly to 41%. 58% 50% 52% 53% 64% 58% 56% 42% 43%  What follows are demographic profiles of those who feel close to the ANC. 56% Q4 2015 Q1 2016 Q2 2016 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018  When compared to the previous quarters we observe a decline in support amongst poorer Black adults, in the 25-34 and 45-54 age categories. PREFER THE PARTY

Support also declined in the Urban and Metro areas, with the largest drop 50% 44% 45% 36% 37% 44% 47% 43% 44% occurring in Gauteng and the Eastern Cape (and marginal growth in the 44% Western Cape). Q4 2015 Q1 2016 Q2 2016 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

LIKE & PREFER THE PARTY

48% 41% 43% 40% 44% 40% 41% 41% 32% 33% Q4 2015 Q1 2016 Q2 2016 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

POPULATION GROUP LIVING STANDARDS MEASURE TOTAL SOUTH AFRICANS LM BLACK 10 181,000 (1%) 12.40-million GT 11,708,000 (94%) 1,779,000 (14%) 2,961,000 (24%) 09 579,000 (5%) COLOURED 33% 455,000 (4%) 08 622,000 (5%) MP INDIAN NW 1,131,000 (9%) 1,239,000 (10%) 143,000 (1%) 07 806,000 (7%) 4,558,000 (37%) WHITE 06 91,000 (1%) FS KZ 2,456,000 (20%) 05 NC 705,000 (6%) 2,536,000 (20%) 293,000 (2%) AGE GROUP 04 1,899,000 (15%)

18-24 1,603,000 (13%) 03 592,000 (5%) EC 02 228,000 (2%) 25-34 3,625,000 (29%) 1,633,000 (13%) WC 01 43,000 (0%) GEOTYPE 553,000 (4%) METRO 35-44 2,949,000 (24%) 3,949,000 (32%) GENDER

FEMALE URBAN 45-54 1,696,000 (14%) 6,903,000 (56%) 2,437,000 (20%)

MALE RURAL 55+ 2,525,000 (20%) 5,494,000 (44%) 6,012,000 (48%)

 The proportion of South Africans who feel close to the DA remained flat at FEEL CLOSE TO THE PARTY 7%, while those who like the party grew slightly to 31%. 7% 8% 7% 7% 7%  The DA’s racially-diverse support base served as a unifying factor against 7% the Zuma administration. In post-Zuma politics, the diverse interests of its Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 constituencies manifested in the internecine struggles that have LIKE THE PARTY undermined the party. 34% 29% 29% 32% 32% 31% 34% 30% 31%  The decline in the preference for the DA is evident over the past year, 31% dropping from highs of around 19% a year ago to 12% in the last quarter Q4 2015 Q1 2016 Q2 2016 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 of 2018. The combined measure of ‘like and prefer’ the DA mirrors this PREFER THE PARTY trend, with a decrease from 18% to 10% over the past 12 months. 21% 18% 18% 19% 19% 16% 13% 12% 12% 13% Q4 2015 Q1 2016 Q2 2016 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

LIKE & PREFER THE PARTY

20% 17% 16% 17% 18% 11% 14% 11% 10% 11% Q4 2015 Q1 2016 Q2 2016 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

POPULATION GROUP LIVING STANDARDS MEASURE TOTAL SOUTH AFRICANS

BLACK LM 10 250,000 (10%) 2.46-million GT 886,000 (36%) 158,000 (6%) 642,000 (26%) 09 297,000 (12%) COLOURED 7% 878,000 (36%) 08 256,000 (10%) MP INDIAN NW 140,000 (6%) 07 296,000 (12%) 153,000 (6%)

WHITE 06 816,000 (33%) 626,000 (25%) FS KZ 05 347,000 (14%) NC 74,000 (3%) 137,000 (6%) AGE GROUP 04 128,000 (5%)

18-24 345,000 (14%) 03 51,000 (2%) EC 02 23,000 (1%) 25-34 639,000 (26%) 147,000 (6%) 01 WC GEOTYPE 871,000 (35%) METRO 35-44 516,000 (21%) 1,154,000 (47%) GENDER

FEMALE URBAN 45-54 453,000 (18%) 1,217,000 (49%) 666,000 (27%)

MALE RURAL 55+ 511,000 (21%) 1,247,000 (51%) 643,000 (26%)

 Between 2016 and 2017, Mmusi Maimane’s performance score with respect to improving the image of the DA soared to 53%. LEADER IS IMPROVING THE PARTY'S IMAGE AMONGST VOTERS

 However, this started to decline from the beginning of 2018, most likely 51% 53% 46% 46% 47% 47% 46% 43% 41% 43% 44% due to the internal party conflicts and the water crisis in the Western 37% 41% Cape. 44% Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4  Furthermore, Maimane’s favourability ratings have remained relatively flat 2015 2016 2016 2016 2016 2017 2017 2017 2017 2018 2018 2018 2018 over the past year, at around 31%.

LEADER’S FAVOURABILITY RATING

37% 35% 37% 32% 34% 31% 33% 32% 32% 31% 28% 29% 29% 27% Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 2015 2016 2016 2016 2016 2017 2017 2017 2017 2018 2018 2018 2018

 The proportion of South African adults who feel close to the EFF has FEEL CLOSE TO THE PARTY remained flat over the past 12 months. 4% 5% 5% 7% 5%  While the EFF has steadily increased in popularity, its growth has 5% remained unchanged (around the 30%) over the past three quarters. Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

LIKE THE PARTY

23% 23% 31% 29% 29% 16% 15% 19% 20% 29% Q4 2015 Q1 2016 Q2 2016 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

PREFER THE PARTY

7% 8% 12% 9% 11% 9% 9% 11% 10% 10% Q4 2015 Q1 2016 Q2 2016 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

LIKE & PREFER THE PARTY

7% 7% 11% 8% 10% 8% 9% 10% 10%

Q4 2015 Q1 2016 Q2 2016 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

POPULATION GROUP LIVING STANDARDS MEASURE TOTAL SOUTH AFRICANS

BLACK LM 10 104,000 (5%) 2.00-million GT 1,910,000 (95%) 251,000 (13%) 702,000 (35%) 09 165,000 (8%) COLOURED 5% 31,000 (2%) 08 45,000 (2%) MP INDIAN NW 126,000 (6%) 21,000 (1%) 07 331,000 (17%) 157,000 (8%)

WHITE 06 682,000 (34%) 40,000 (2%) FS KZ 05 446,000 (22%) NC 102,000 (5%) 262,000 (13%) 36,000 (2%) AGE GROUP 04 217,000 (11%)

18-24 462,000 (23%) 03 11,000 (1%) EC 02 25-34 659,000 (33%) 242,000 (12%) 01 WC GEOTYPE 124,000 (6%) METRO 35-44 501,000 (25%) GENDER 821,000 (41%)

FEMALE URBAN 45-54 198,000 (10%) 782,000 (39%) 458,000 (23%)

MALE RURAL 55+ 181,000 (9%) 1,220,000 (61%) 724,000 (36%)

 Julius Malema’s efforts at improving his party’s image have been successful, especially when viewed across time: from 22% in Q1 of 2016 to LEADER IS IMPROVING THE PARTY'S IMAGE AMONGST VOTERS 43% in the current quarter. 43% 40% 39% 35% 36% 37% 29% 30%  Malema’s growing favourability score is significant, nearly doubling from a 26% 28% 27% 28% 22% base of 16% in the first quarter of 2016 to 30% in the final quarter of 43% 2018. Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 2015 2016 2016 2016 2016 2017 2017 2017 2017 2018 2018 2018 2018

LEADER’S FAVOURABILITY RATING

31% 29% 30% 25% 25% 25% 21% 20% 22% 19% 22% 19% 16% 30% Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 2015 2016 2016 2016 2016 2017 2017 2017 2017 2018 2018 2018 2018

Original data and written content: © 2018 Citizen Surveys. All rights reserved. 28 SECTION FIVE POLITICAL LEADERS’ ELECTION CAMPAIGN ACTIVITY POLITICAL LEADERS’ ELECTION CAMPAIGN ACTIVITY CYRIL RAMAPHOSA AND JACOB ZUMA ( JANUARY 2019 )

 We asked respondents "If the National Election FAVOURABILITY AMONGST POTENTIAL ANC VOTERS was held tomorrow, who would you vote for?" IF THE ELECTIONS WERE HELD TOMORROW, 20.1m PEOPLE WOULD VOTE FOR THE ANC

 Respondents were then given a computer CYRIL 14.6m 5.6m tablet and asked to cast their vote, RAMAPHOSA 72% 28% simulating a ballot. JACOB 7.2m 12.9m  We then examined the favourability ratings of ZUMA 36% 64% both President Ramaphosa and ex-President FAVOURABLE NOT FAVOURABLE Zuma among those 20.1-million people who would vote for the ANC.

 Roughly 14.6-million (72%) of these ANC PREFERENCE AMONGST ANC SUPPORTERS JANUARY 2019 supporters viewed Ramaphosa favourably, while 7.2-million (36%) viewed Zuma favourably, and some seeing both favourably.

 In terms of preference, 13.1-million (65%) of 13.1m 2.9m 4.1m the 20.1-million ANC supporters preferred 65% 15% 20% Ramaphosa over Zuma. Interestingly, 2.9-million (15%) preferred them equally. RAMAPHOSA OVER ZUMA PREFERRED EQUALLY ZUMA OVER RAMAPHOSA

 In terms of preferences across the provinces, PREFERENCE AMONGST ANC SUPPORTERS Ramaphosa is preferred in all provinces JANUARY 2019 except for KwaZulu-Natal, where 51% of ANC supporters prefer Zuma over Ramaphosa. Gauteng 68% 14% 18%

 It is worth noting the following in the KwaZulu-Natal 23% 26% 51% provinces where Ramaphosa is preferred Eastern Cape 79% 10% 11% over Zuma:

 Ramaphosa enjoys the highest preference Western Cape 89% 7% amongst ANC supporters in the Western Free State 74% 16% 9% Cape (89%), which is also the province with

the lowest preference for Zuma. Limpopo 72% 17% 11%

 Ramaphosa has a lower level of support in Mpumalanga 77% 9% 14% Gauteng (68%), compared to the other

provinces. North West 73% 8% 19%

Northern Cape 72% 19% 10%

RAMAPHOSA OVER ZUMA PREFERRED EQUALLY ZUMA OVER RAMAPHOSA

 The SACS also asked about the effect that ex- President Zuma would have on the voting behaviour of ANC supporters, were he to be JACOB ZUMA’S EFFECT ON ANC SUPPORTERS JANUARY 2019 deploying on the ANC's campaign trail.

 Of the 20.1-million ANC supporters, 6.7- million (34%) would be more likely to vote for 10.6m (53%) 2.8m (14%) 6.7m (34%) LESS LIKELY NO EFFECT MORE LIKELY the ANC if Zuma was on the campaign trail. TO VOTE FOR TO VOTE FOR THE ANC THE ANC  Some 10.6-million (53%) would be less likely to vote for the ANC in this scenario, however, which is an important finding.

 We asked respondents "If the National Election FAVOURABILITY AMONGST POTENTIAL DA VOTERS was held tomorrow, who would you vote for?" IF THE ELECTIONS WERE HELD TOMORROW, 6.4m PEOPLE WOULD VOTE FOR THE DA

 Respondents were then given a computer MMUSI 3.8m 2.6m tablet and asked to cast their vote, MAIMANE 59% 41% simulating a ballot. HELEN 4.5m 2.0m  We then examined the favourability ratings of ZILLE 70% 30% both DA Leader Mmusi Maimane and former- FAVOURABLE NOT FAVOURABLE Western Cape Premier Helen Zille among the 6.4-million people who would vote for the DA.

 Roughly 3.8-million (59%) of these DA PREFERENCE AMONGST DA SUPPORTERS FEBRUARY 2019 supporters viewed Maimane favourably. Helen Zille enjoyed higher levels, however, with 4.5-million (70%) viewing her favourably.

 Roughly 2.0-million (30%) of the 6.4-million 2.0m 1.1m 3.4m DA supporters preferred Maimane over Zille. 30% 17% 52% On the other hand, 3.4-million (52%) DA supporters preferred Zille over Maimane. MAIMANE OVER ZILLE PREFERRED EQUALLY ZILLE OVER MAIMANE

 In terms of preferences across the two PREFERENCE AMONGST DA SUPPORTERS provinces where the DA has a strong voting FEBRUARY 2019 presence:

 In the Western Cape, 62% of DA Gauteng 23% 39% 38% supporters prefer Zille over Maimane, while 22% prefer the two equally, and 15% Western Cape 15% 22% 62% prefer Maimane over Zille.

 In Gauteng, 39% of DA supporters prefer MAIMANE OVER ZILLE PREFERRED EQUALLY ZILLE OVER MAIMANE both figures equally, followed closely by 38% who prefer Zille over Maimane.

 Given these levels of preference for Zille, it is HELEN ZILLE’S EFFECT ON DA SUPPORTERS FEBRUARY 2019 unsurprising that her presence would mean that 3.7-million (58%) DA supporters are more likely to vote for the DA. 1.9m (29%) 841k (13%) 3.7m (58%)  This is most likely due to her long association LESS LIKELY NO EFFECT MORE LIKELY TO VOTE FOR TO VOTE FOR with the party, especially as its leader. As a THE DA THE DA result, the two are closely-associated.

 In Quarter 4 of 2018, trust in the IEC grew significantly to 59%, from its all- TRUST IN: ELECTORAL COMMISSION (IEC) time low of 48% in the previous quarter HOW MUCH TRUST, IF ANY, DO YOU HAVE IN… ?

 It would appear that as the IEC engages with the public in the leadup to 67% the general election, more adult citizens of South Africa have begun to 62% 62% 59% trust the IEC. 57% 54% 54%  The degree of trust in the IEC has also improved, with an increase in those 49% 48% who say they have “a lot of trust”. 39%  Those South Africans who said that they have a lot of trust in the IEC 37% has dropped since the beginning of 2018, however, from 31% to 25% in 32% 32% 32% 31% 31% Q2 of 2018 and Q3 0f 2018. 25% 25% 30%  Those with some trust in the IEC declined from 37% in Q1 0f 2017 to 28% 29% 25% 25% 25% 22% in Q4 of 2017, and have largely remained at this level until this 22% 23% 23% quarter.

 In this quarter, both degrees of trust saw growth.

Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

SOME TRUST / A LOT OF TRUST A LOT OF TRUST SOME TRUST

POPULATION GROUP LIVING STANDARDS MEASURE TOTAL SOUTH AFRICANS LM BLACK 10 671,000 (4%) 14.97-million GT 11,704,000 (78%) 1,513,000 (10%) 4,082,000 (27%) 09 1,368,000 (9%) COLOURED 40% 1,534,000 (10%) 08 1,206,000 (8%) MP INDIAN NW 1,371,000 (9%) 280,000 (2%) 07 1,973,000 (13%) 865,000 (6%)

WHITE 06 4,722,000 (32%) 1,447,000 (10%) FS KZ 05 2,821,000 (19%) NC 791,000 (5%) 2,638,000 (18%) 328,000 (2%) AGE GROUP 04 2,550,000 (17%)

18-24 3,155,000 (21%) 03 459,000 (3%) EC 02 448,000 (3%) 25-34 4,581,000 (31%) 1,218,000 (8%) WC GEOTYPE 01 57,000 (0%) 2,159,000 (14%) METRO 35-44 2,904,000 (19%) GENDER 6,192,000 (41%)

FEMALE URBAN 1,776,000 (12%) 45-54 8,180,000 (55%) 3,119,000 (21%)

RURAL 55+ MALE 6,785,000 (45%) 5,654,000 (38%)

 Of the 37-million South African adults (18+ years), roughly 60% (or 22.2-million potential voters) claimed that they are certain that they are both registered to vote and have a stamp in their ID book. 37-million SOUTH AFRICANS AGED 18+  Some 6% (2.4-million) potential voters believed that they are registered,

but were uncertain of whether their green ID document had been

:| :| :| :

stamped or not. :

) (  Another 18% (6.7-million) potential voters intended to register, while about 3% (1.1-million) adults are not sure yet. 22.2m 2.4m 6.7m 1.1m 4.5m  A substantial number of eligible voters (12%, or 4.5-million) are not 60% 6% 18% 3% 12% interested in registering to vote, however. REGISTERED + POSSIBLY INTEND TO UNCERTAIN / DON’T INTEND STAMPED ID REGISTERED + REGISTER DON’T KNOW TO REGISTER  Being registered to vote is a precondition to voter turnout. UNSURE OF STAMPED ID

10.2-million 28% POTENTIAL VOTERS TO OF SOUTH AFRICAN BE REGISTERED ADULTS

POPULATION GROUP LIVING STANDARDS MEASURE TOTAL SOUTH AFRICANS

BLACK LM 10 229,000 (5%) 4.53-million GT 3,590,000 (79%) 414,000 (9%) 1,350,000 (30%) 09 376,000 (8%) COLOURED 12% 438,000 (10%) 08 454,000 (10%) MP INDIAN NW 441,000 (10%) 110,000 (2%) 07 622,000 (14%) 395,000 (9%)

WHITE 06 1,600,000 (35%) 387,000 (9%) FS KZ 05 548,000 (12%) NC 218,000 (5%) 661,000 (15%) 87,000 (2%) AGE GROUP 04 471,000 (10%)

18-24 1,543,000 (34%) 03 109,000 (2%) EC 02 117,000 (3%) 25-34 1,430,000 (32%) 304,000 (7%) WC GEOTYPE 01 ,000 (0%) 655,000 (14%) METRO 35-44 746,000 (16%) GENDER 2,162,000 (48%)

FEMALE URBAN 312,000 (7%) 45-54 2,039,000 (45%) 670,000 (15%)

MALE RURAL 55+ 494,000 (11%) 2,486,000 (55%) 1,693,000 (37%)

POTENTIAL VOTERS TO BE REGISTERED  We asked of the 28% (10.2-million) potential voters to be registered about the challenges 5.7m 4.6m they may face in voter registration: 56% 44% FACE NO CHALLENGES FACE CHALLENGES  How likely are you to take time off to register to vote? REGISTRATION CHALLENGES  How likely are you to pay for transport or HOW LIKELY ARE YOU TO HOW LIKELY ARE YOU TO PAY FOR HOW LIKELY ARE YOU TO walk in order to register? TAKE TIME OFF TO TRANSPORT OR WALK IN ORDER GIVE YOUR RESIDENTIAL REGISTER ? TO REGISTER? ADDRESS TO THE IEC ?  How likely are you to give your residential address to the IEC? 5% 8% 7% 9% 8% 26%  About 79% (7.8-million) of these potential 14% 35% 8% 33% voters will be able to take time off to register. 10%

 About 68% (6.7-million) of these potential 9% voters will pay for transport or walk to register. 44% 42% 42%  About 75% (7.4-million) of these potential voters would provide the IEC with their proof NEITHER LIKELY VERY LIKELY LIKELY UNLIKELY VERY UNLIKELY of address. NOR UNLIKELY

POPULATION GROUP LIVING STANDARDS MEASURE TOTAL SOUTH AFRICANS

BLACK LM 10 223,000 (5%) 4.56-million GT 3,809,000 (84%) 374,000 (8%) 1,402,000 (31%) 09 413,000 (9%) COLOURED 12% 310,000 (7%) 08 289,000 (6%) MP INDIAN NW 389,000 (9%) 80,000 (2%) 07 389,000 (9%) 228,000 (5%)

WHITE 06 1,574,000 (35%) 357,000 (8%) FS KZ 05 1,014,000 (22%) NC 194,000 (4%) 802,000 (18%) 116,000 (3%) AGE GROUP 04 474,000 (10%)

18-24 1,202,000 (26%) 03 106,000 (2%) EC 02 66,000 (1%) 25-34 1,520,000 (33%) 479,000 (11%) WC GEOTYPE 01 8,000 (0%) 573,000 (13%) METRO 35-44 735,000 (16%) GENDER 1,863,000 (41%)

FEMALE URBAN 681,000 (15%) 45-54 2,368,000 (52%) 900,000 (20%)

MALE RURAL 55+ 419,000 (9%) 2,189,000 (48%) 1,794,000 (39%)

SOUTH AFRICAN ADULTS ( 18+ )  Whether they are registered or not, voters may also face voting day challenges. 23.3m 13.8m 63% 37%  We ask two questions in order to establish whether FACE NO CHALLENGES FACE CHALLENGES potential voters believe that they may encounter voting day challenges: EASE OF VOTING VOTER PATIENCE HOW EASY IS IT FOR YOU TO GET TO A HOW LIKELY ARE YOU TO WAIT TO PLACE  How easy is it for you to get to a local polling LOCAL POLLING STATION? YOUR VOTE, EVEN IF THERE IS A LONG QUEUE? station? and 26.2m 12.7m  How likely is it that you would wait to place your Very easy Very Likely 71% 34% vote, even if there is a long queue at the voting station? 5.9m 12.7m Fairly easy Likely 16% 34%  The respondent would be deemed able to

overcome voting day challenges if they answered in 1.7m 2.3m Neither Likely Fairly difficult nor Unlikely the affirmative to both questions. 5% 6%

 Using these criteria, 63% or 23.3-million citizens 1.4m 3.8m Very difficult Unlikely would be able to overcome voting day challenges, 4% 10% while 37% may face challenges at the polls. 1.8m 4.3m Don't Know Very Unlikely 5% 12%

POPULATION GROUP LIVING STANDARDS MEASURE TOTAL SOUTH AFRICANS

BLACK LM 10 780,000 (6%) 13.77-million GT 11,376,000 (83%) 1,053,000 (8%) 3,518,000 (26%) 09 1,122,000 (8%) COLOURED 37% 896,000 (7%) 08 1,111,000 (8%) MP INDIAN NW 1,125,000 (8%) 369,000 (3%) 07 1,453,000 (11%) 927,000 (7%)

WHITE 06 4,401,000 (32%) 1,126,000 (8%) FS KZ 05 2,312,000 (17%) NC 852,000 (6%) 2,616,000 (19%) 249,000 (2%) AGE GROUP 04 1,505,000 (11%)

18-24 3,211,000 (23%) 03 545,000 (4%) EC 02 500,000 (4%) 25-34 4,365,000 (32%) 2,088,000 (15%) WC GEOTYPE 01 38,000 (0%) 1,340,000 (10%) METRO 35-44 2,472,000 (18%) GENDER 4,841,000 (35%)

FEMALE URBAN 1,576,000 (11%) 45-54 7,411,000 (54%) 2,980,000 (22%)

MALE RURAL 55+ 2,144,000 (16%) 6,356,000 (46%) 5,946,000 (43%)

 In order to gauge the motivation to vote amongst South African adults, we asked: “Thinking about the 2019 national VOTER MOTIVATION elections, how much do you want to vote in next year’s THINKING ABOUT THE 2019 NATIONAL ELECTIONS, HOW MUCH DO elections?” YOU WANT TO VOTE IN NEXT YEAR’S ELECTIONS?

 Overall, 72% or 26.6-million South African adults are

motivated to vote, while the remaining 28% or 10.5-million 3% 12% citizens have little or no motivation to vote. 37-million SOUTH AFRICANS AGED 18+

 Of the 26.6-million citizens who are motivated to vote, 13% roughly 20.0-million South African adults (54%) expressed a 54% great deal of motivation to vote, while 6.6-million (18%) said 26.6-million (72%) MOTIVATED TO VOTE that they wanted to vote quite a lot. 18%

 In analysing those who are not motivated to vote, we 10.5-million (28%) observe that roughly 13% (4.8-million) said that they have NOT MOTIVATED TO VOTE not much motivation to vote, while 12% (4.3-million people)

have no intention to vote at all. DON’T KNOW / A GREAT DEAL QUITE A LOT NOT VERY MUCH NOT AT ALL REFUSED

VOTER READINESS SEGMENTATION MODEL VOTER READINESS MODEL (VRM)

 We set about modelling the impact of voter registration, voting-day READY VOTERS challenges and the motivation to vote in order to establish the impact Those citizens who are registered to vote, motivated to vote and are capable these factors may have on voter turnout. of overcoming all voting day challenges.  In total thirteen scenarios were considered, various combinations of voter motivation, registration challenges and voting day challenges. LIKELY VOTERS Those citizens who are motivated to vote, are uncertain whether they are  These thirteen scenarios were distilled into four prototype voter readiness registered to vote, however they are capable of overcoming registration segments. challenges as well as capable of overcoming all voting-day challenges.

CHALLENGED VOTERS Those while they are motivated to vote are likely to encounter registration and / or voting day challenges.

UNMOTIVATED VOTERS They are not yet motivated to vote regardless of whether they will face registration and / or voting day challenges.

 In order to estimate voter turnout, we take into account all preceding  When we apply these segments to the overall adult population, we see preconditions: motivation to vote, registration status, registration that those South Africans aged 18 and older who are most likely to vote challenges, and voting-day challenges. Collectively this allows us to divide total 17.6-million: South African adults into four segments: 37-million SOUTH AFRICANS AGED 18+ VOTERS WHO ARE MOTIVATED, VOTERS WHO ARE MOTIVATED, REGISTERED, AND CAN OVERCOME UNCERTAIN OF THEIR REGISTRATION, CHALLENGES AND CAN OVERCOME CHALLENGES READY VOTERS LIKELY VOTERS 14.5m (39%) 1 2 1 READY VOTERS 17.6m MOST LIKELY TO TURN OUT AT UNMOTIVATED VOTERS CHALLENGED VOTERS 3.2m (9%) THE POLLS VOTERS WHO ARE NOT YET MOTIVATED VOTERS WHO ARE MOTIVATED, 2 LIKELY VOTERS TO VOTE, REGARDLESS OF REGISTRATION UNREGISTERED, AND LIKELY TO OR VOTING CHALLENGES FACE CHALLENGES 9.0m (24%) 3 CHALLENGED VOTERS 4 3 10.5m (28%) 4 UNMOTIVATED VOTERS Original data and written content: © 2018 Citizen Surveys. All rights reserved. 46 SCENARIOS: POLITICAL PARTIES AND VOTER SUPPORT IF THE GENERAL ELECTION WAS HELD TOMORROW WHO WOULD YOU VOTE FOR?

 We asked “If the elections were held tomorrow, which party would you vote VOTER TURNOUT SCENARIOS AND PARTY SUPPORT for?” We placed a mock ballot on the computer tablet and gave the tablet  Using the Voter-Readiness Model (VRM), we set about analysing various to the respondent to select the party they would vote, simulating the voter turnout scenarios. voting day experience in terms of the ballot and its privacy.  Based on these turnout scenarios, we analysed the outcome of the  A total of ten political parties were listed, with an option for ‘other mock ballot vote cast by respondents. political party’ provided. 1. Total Turnout: If the entire adult population of 37-million were to vote,  If the respondent, after going through the exercise, did not chose a what would the election outcome be? political party, the interviewer would then record their choice as ‘undecided’. 2. High Turnout: If Ready Voters, Likely Voters, and Challenged Voters who overcame most of their challenges (a 79% voter turnout) were to cast  Using the political party voting choice, we set about analysing various their ballots, what would the election outcome be? voter turnout scenarios. 3. Medium Turnout: If Ready Voters, Likely Voters and Challenged Voters who overcame some of their challenges (a 75% voter turnout) were to In these scenarios, voter turnout is calculated vote, what would the election results look like? as a proportion of the 26.7-million* voters 4. Low Turnout: If only Ready and Likely Voters were to cast their ballots, registered with the IEC, as at 31 January 2019. representing a voter turnout of 66%, what would the outcome be? * www.elections.org.za/content/Voters-Roll/Registration-statistics/

 If the entire voter base (i.e. 37-million South Africans aged 18+) were to TOTAL VOTER TURNOUT vote tomorrow: ALL SOUTH AFRICAN ADULTS ( 18+ )

 The ANC would receive 52% of the vote (19.4-million);

 The DA would receive 17% of the vote (6.4-million); 52% 19.4m  The EFF would receive 11% of the vote (4.2-million); and  Roughly 5% (1.7-million) would vote for the smaller parties. 17%  The undecided vote is of significance in modelling potential election 6.4m results. We observe that undecided citizens constitute 14% (5.3-million) 11% people. 4.2m  We know that the entire voter population of South Africa will not go to the polls, however. It is therefore necessary to model potential voter turnout 5% scenarios using the VRSM model. 1.7m

14% 5.3m 37.0m

ANC DA EFF OTHER PARTY UNDECIDED

The High Voter Turnout scenario assumes The Medium Voter Turnout scenario assumes The Low Voter Turnout scenario assumes only Ready, Likely, and those Challenged voters who Ready, Likely, and those Challenged voters who Ready and Likely voters will vote, which results in can overcome most obstacles will vote. The can overcome some challenges will vote. This a 66% turnout (17.6-million voters) – a similar estimated turnout is 79% (21.0-million). place voter turnout at 75% (20-million voters). level to that of the 2014 general election.

HIGH VOTER TURNOUT MEDIUM VOTER TURNOUT LOW VOTER TURNOUT READY + LIKELY + CHALLENGED VOTERS READY + LIKELY + CHALLENGED VOTERS READY + LIKELY VOTERS ONLY WHO OVERCOME MOST CHALLENGES WHO OVERCOME SOME CHALLENGES

58% 58% 57% 12.3m 11.6m 10.1m

20% 20% 20% 4.0m 3.9m 3.5m

8% 8% 8% 1.7m 1.6m 1.4m 5% 5% 5% 1.0m 973k 917k

9% 21.0m 9% 20.0m 9% 17.6m 2.0m 79% 1.9m 75% 1.6m 66%

ANC DA EFF OTHER PARTY UNDECIDED

TOTAL VOTER TURNOUT HIGH VOTER TURNOUT MEDIUM VOTER TURNOUT LOW VOTER TURNOUT ALL SOUTH AFRICAN ADULTS ( 18+ ) READY + LIKELY + CHALLENGED VOTERS READY + LIKELY + CHALLENGED VOTERS READY + LIKELY VOTERS ONLY WHO OVERCOME MOST CHALLENGES WHO OVERCOME SOME CHALLENGES

52% 58% 58% 57% 19.4m 12.3m 11.6m 10.1m

17% 19% 20% 20% 6.4m 4.0m 3.9m 3.5m

11% 8% 8% 8% 4.2m 1.7m 1.6m 1.4m

5% 5% 5% 5% 1.7m 1.0m 973k 917k

14% 37.0m 9% 21.0m 9% 20.0m 9% 17.6m 5.3m 2.0m 79% 1.9m 75% 1.6m 66%

ANC DA EFF OTHER PARTY UNDECIDED

UNDECIDED VOTERS IMPLICATIONS FOR THE ANC, DA, EFF

 The decisions made by undecided voters on election day will be key to the  The ANC garners 52% of the vote at a 100% voter turnout. At a 66% voter electoral performance of the three major political parties. turnout, the party receives 57% of the vote, and at 75% voter turnout they receive 58% of the vote.  Within the entire adult population, 14% (5.3-million) are undecided as to which political party they would vote for.  The DA receives 17% of the vote at a total voter turnout, which grows to 20% of the vote at both the 66% and 75% voter turnout scenarios.  In the High, Medium, and Low Turnout scenarios, the proportion of undecided voters drops to 9%.  It therefore appears that both the ANC and the DA would benefit most from a lower voter turnout.  This means that those citizens who are likely to go to the polls, already have a party in mind that they intend to vote for.  For the EFF, a total voter turnout nets them 11% of votes. At a 66% voter turnout and a 75% voter turnout, they receive 8% of the vote, however.

TURNOUT SCENARIO RESULTS  It therefore appears as though the EFF would benefit the most if the entire adult population were to cast their ballots in 2019.  A key feature of the three turnout scenarios is that the election outcomes are all very similar (and in fact all fall within the sampling error of 1.5%).

 This therefore means that the only noteworthy changes in election results would occur if all eligible voters participated. This in effect means that the Unmotivated Voters are the ones who will shift the election outcomes.

Original data and written content: © 2018 Citizen Surveys. All rights reserved. 51 SECTION SEVEN MOST IMPORTANT PROBLEMS FACING SOUTH AFRICA THE MOST IMPORTANT PROBLEMS FACING SOUTH AFRICA RANKING OF IMPORTANCE

 Unemployment is the most important problem facing the MOST IMPORTANT PROBLEMS country – by a considerable margin – as expressed by 26.3- IN YOUR OPINION, WHAT ARE THE THREE MOST IMPORTANT PROBLEMS FACING SOUTH AFRICA ? million (72%) adult South Africans. Unemployment 72% 26.3m  Crime received the second-highest number of responses, as Crime 41% 15.0m indicated by 15.0-million (41%) of South Africans. Corruption 25% 9.1m  Corruption was the third-most important problem facing Poverty / Destitution 23% 8.3m SA, as cited by 9.1-million (25%) South Africans. While Housing 17% 6.3m “Crime” grew from the previous quarter, “Corruption” grew Water Supply 15% 5.6m even more. This is most likely owing to the continuing Education 11% 3.9m revelations of corruption, from State Capture to Steinhoff Roads 10% 3.6m and now Bosasa. Electricity Supply 8% 2.8m  Poverty / Destitution was the fourth-most cited problem, Economic / Income Inequality 7% 2.7m with 8.3-million (23%) of South Africans citing this issue. Basic Health 6% 2.2m

 Infrastructural issues – Housing, Water Supply, Roads, and Management of the Economy 5% 2.0m Electricity Supply – and Education were also prominent Immigrants / Xenophobia 5% 2.0m issues. Discrimination / Racism 4% 1.4m Wages / Income 3% 1.3m

 President Ramaphosa has emphasised that rooting out government PERCEPTIONS OF CORRUPTION corruption would be a national priority. This resonates with South IN YOUR OPINION, IS THE LEVEL OF CORRUPTION IN SOUTH AFRICA INCREASING, STAYING THE SAME, OR DECREASING? Africans, of whom 9.1-million (25%) said that corruption was one of the

most important problems facing South Africa. Zondo Commission of  Since April 2018, in the process of tackling corruption within government President Nugent Inquiry begins Ramaphosa Commission of (for example, the Nugent and Zondo Commissions), South Africans have admits SOEs are Inquiry begins ‘sewers of 81% 81% become increasingly aware of the extent of State Capture and general 80% corruption’ corruption and mismanagement, and their adverse impact on the country 77% 74% and its economy. 72%

 In April, 59% of South Africans believed that corruption in South Africa is 67% growing. In May, just over two-thirds (67%) believed this, and this measure has continued to grow. November and December 2018 have seen the 59% highest levels, at 81%. INCREASING  Corruption is unlikely to be getting worse; instead, corruption is being brought out into the open. In order to restore trust, however, corrupt Apr 2018 May 2018 Jun 2018 Jul 2018 Aug 2018 Sep 2018 Nov 2018 Dec 2018 officials who broke the law must face serious consequences.

 During this quarter, South Africans PARTY ASSOCIATION associated the ANC most with providing I WOULD LIKE YOU TO TELL ME WHICH POLITICAL PARTY YOU THINK IS BEST AT… ? water, providing housing, maintaining roads, and providing electricity. STRONG ASSOCIATION  The DA was not strongly associated with

addressing any of the problems, which Don't Know Reducing poverty reflects its lack of clear policy and direction.

None Providing electricity  The EFF were most closely associated with Creating jobs

reducing crime and reducing corruption Reducing income Maintaining roads Reducing crime (despite the VBS Bank scandal). STRONG DA WEAK inequality STRONG ASSOCIATION ASSOCIATION ASSOCIATION Providing water  Finally, South Africans were unsure which party was best at reducing poverty. Improving education ANC Reducing corruption

Providing housing

EFF STRONG ASSOCIATION

RELIGIOUS LEADERS TRADITIONAL LEADERS

 Over the past two years, there has been a slow-but-steady decline in trust  Trust in traditional leadership follows a similar trend to religious in religious leadership in general, with this trend continuing in the current leadership, although from a lower starting level of 46%. While the decline quarter to a new low of 53% during this quarter. in trust is evident across the provinces and within the rural areas, trust in traditional leadership is slightly higher in rural areas.

TRUST IN: RELIGIOUS LEADERS TRUST IN: TRADITIONAL LEADERS HOW MUCH TRUST, IF ANY, DO YOU HAVE IN… ? HOW MUCH TRUST, IF ANY, DO YOU HAVE IN… ?

66% 64% 64% 61% 58% 58% 56% 58% 53% 47% 46% 45% 45% 44% 42%

35% 35% 36%

Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

ELECTORAL COMMISSION (IEC) CONSTITUTIONAL COURT

 In Q4 of 2018, trust in the IEC grew noticeably (to 59%), after a steady  Trust in the Constitutional Court grew during this quarter, to 59%, after a decline over time. Since its previous high of 67% was after the 2016 local dip in trust levels in the beginning of 2018. government elections, as the IEC engages with the public in the leadup to the election, we observe that trust in the IEC increases.

TRUST IN: ELECTORAL COMMISSION (IEC) TRUST IN: CONSITUTIONAL COURT HOW MUCH TRUST, IF ANY, DO YOU HAVE IN… ? HOW MUCH TRUST, IF ANY, DO YOU HAVE IN… ?

70% 67% 68% 65% 65% 65% 66% 62% 62% 59% 59% 57% 58% 56% 54% 54% 49% 48%

Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

PUBLIC PROTECTOR NATIONAL PROSECUTING AUTHORITY

 Busisiwe Mkhwebane was appointed by President Zuma in October 2016.  Trust in the NPA has been severely undermined under the directorship of Trust in the Public Protector’s office has since declined over time, Shaun Abrahams. The growth in trust in the current quarter could be nonetheless there has been growth in trust in the current quarter. linked to Advocate Shamila Batohi, the new head of the NPA appointed by President Ramaphosa.

TRUST IN: PUBLIC PROTECTOR TRUST IN: NATIONAL PROSECUTING AUTHORITY HOW MUCH TRUST, IF ANY, DO YOU HAVE IN… ? HOW MUCH TRUST, IF ANY, DO YOU HAVE IN… ?

69% 66% 64% 61% 61% 59% 57% 57% 58% 53% 52% 53% 53% 51% 50% 50% 47% 46%

Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

SOUTH AFRICAN POLICE SERVICE SOUTH AFRICAN NATIONAL DEFENCE FORCE

 While trust in SAPS has been declining steadily since Q4 of 2016, the most  The SANDF has in general enjoyed more public trust than the South significant drop has during the course of 2018, to the current level of 42%. African Police Service has. While trust in the SANDF has declined since Q1 of 2017, it grew to 59% during this quarter.

TRUST IN: SOUTH AFRICAN POLICE SERVICE TRUST IN: SANDF HOW MUCH TRUST, IF ANY, DO YOU HAVE IN… ? HOW MUCH TRUST, IF ANY, DO YOU HAVE IN… ?

72% 72% 69% 69% 68% 66%

59% 57% 56% 54% 52% 52% 50% 50% 51%

44% 42% 40%

Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018

SOUTH AFRICAN REVENUE SERVICE SOUTH AFRICAN BROADCASTING CORPORATION

 Trust in SARS declined marginally during 2017, but has declined more  Despite the spate of financial mismanagement, corruption, and state noticeably since Q1 of 2018 (most likely due to the state capture capture allegations, roughly two-thirds (66%) of South Africans still trust revelations). SARS did see a slight growth in trust during this quarter, the SABC, and this level has been largely consistent over time. however, to 56%

TRUST IN: SARS TRUST IN: SABC HOW MUCH TRUST, IF ANY, DO YOU HAVE IN… ? HOW MUCH TRUST, IF ANY, DO YOU HAVE IN… ?

68% 67% 66% 66% 64% 65% 64% 65% 64% 62% 64% 63% 62% 60% 62% 55% 54% 56%

Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018 Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018 Q3 2018 Q4 2018