A 'breakthrough' election? (I)

RW Johnson says that the DA does not have an easy path forward into 2019

Analyzing the “breakthrough” Election: 2016

South Africa's long-anticipated local elections of August 2016 have been hailed as a major breakthrough for the Opposition, particularly for the DA and its leader Mmusi Maimane. Without doubt this exaggerated response derived from the fact that public opinion has been through a period of unprecedented gloom, so any indication that there might be an end to the long, dark night of ANC rule was welcomed with huge sighs of relief.

The illusions of victory

Even so sober a publication as the Financial Mail talked enthusiastically of the “DA surge”. This seems an odd way to treat an election in which the DA has continued to trudge forward at the hardly electrifying rate of a 2.95% gain over five years and in which the main dynamic factor has been the entree en scene of the far left EFF with 8.19% and 761 council seats. This, rather than the DA's more modest push, was the principal reason why the ANC fell back by 9.74% since 2011.

It is, indeed, difficult for the DA to advance any faster for its gains now come from only two sources: a differentially higher turnout among minority voters and very gradual gains among African voters. The evidence this time suggests that both things happened – in many white suburbs voting lines stretched from dawn to dusk and the DA vote rose noticeably in many all-African wards.

It is not easy to say how much of this was really Maimane's personal triumph for he is still far less identified with his party in the public mind than the DP was with Tony Leon in the 1999 “Fight Back” election or even the DA with Helen Zille in 2009 or 2014. What can be said, more modestly, was that this was Maimane's first real test and he passed it.

The elections were accompanied by many and varying claims, many of them misleading. In the run-up IPSOS conducted weekly polls, carried by eTV, which showed the ANC trailing dramatically in all the big battleground metros. This attracted strong criticism from the ANC and especially the SACP, which held demonstrations against the polls at the eTV offices in Johannesburg, accusing IPSOS of a “regime change agenda”.

As if rattled by this, in the last week before the vote IPSOS brought out dramatically different figures, suddenly catapulting the ANC vote in Tshwane (Pretoria), for example, from under 30% to 47%. This was apparently achieved by altering the sample used in the last week.  All that one can say is that such behaviour by a polling agency would be unthinkable in any developed country. Everyone knows that such huge changes in opinion do not really occur from week to week, so either the earlier polls or the last one must be admitted to be faulty.

These weaknesses were compounded by IPSOS apparently failing to make much allowance for turnout factors. Indeed, the polls were always reported as if all respondents were intending to vote. South African voters are very prone to say they will definitely vote even though abstention rates are often high. On this occasion 57.97% voted compared to 57.6% in 2011.

When the results were known some commentators excitedly claimed that the ANC vote had gone down everywhere, in the countryside just as much as the towns. Others claimed to see the effect of patronage politics in the fact that the ANC vote held up better where the “premier league” ruled. Peter Bruce in Business Day excitedly warned that the buffeting the ANC had taken in 2016 was as nothing compared to the complete hiding that awaited it in 2019 and even held out the prospect of a President Maimane after 2019. None of these conclusions can be justified.

2019 and all that

Let us take the last claim first. It may be that the decline in the ANC vote shown in 2016 will continue but the ANC has always done better in national elections rather than local ones because it is better able to pull out its massive vote then. The DA has fared correspondingly worse in national elections. So, if existing patterns hold, one would expect the ANC to recover somewhat in 2019 compared to 2016. Secondly, the dynamic factor in 2016 as compared with 2011 was the EFF's entry into the mix.

But when we get to 2019 we will be comparing back with 2014 when the EFF was already in the mix. Thirdly, 2016 offered Opposition voters the enticing prospect of displacing the ANC from power in major towns and cities, but no such prospect will be on offer in 2019. Talk of Maimane leading a government after 2019 are thus hugely inflated. In 2014, after all, the DA gained 22.23% of the vote, way short of the 30% target Zille had set. Even if past trends continue – and DA gains will be harder now – 26% of the vote in 2019 and 30% in 2024 would be a good achievement. The ANC still gets more than twice that.

Of course, one cannot assume that the DA's regular march simply continues at 3% every five years. Its progress gets more difficult as it needs to hack into core ANC support. On the other hand a lot depends on whether the ANC is now seen as a declining party. Already party discipline has slipped considerably and individual ministers seem to do what they like – that is to say, a process of dis-aggregation appears to be under way.

If people cease to believe that the ANC is destined to “rule until Jesus comes”, this will inevitably hasten its decline. Such decline could well be uneven and rapid: as an iceberg melts, bigger and bigger lumps fall off. It is certainly too soon to think this is bound to happen but far bigger and more monolithic parties than the ANC have vanished away. To quote a Russian book on the Soviet Union, “Everything was forever and then it was no more”.

Table 1. Party swings by province, 2011-2016

                                          ANC                          DA                         EFF

Eastern Cape                    -7.04                         +3.30                        5.18

Free State                       -10.05                         +0.51                        9.69

Gauteng                          -14.62                        +3.90                       11.36

KZN                                 -0.34                         +2.90                        3.46

Mpumalanga                    -8.45                         -0.75                        9.39

N. Cape                            -5.52                         +2.75                       8.60

Limpopo                         -13.92                        +1.42                      16.73

North West                     -16.04                         -0.89                      15.54

W. Cape                           -7.88                         +5.22                        2.81

Overall national swing    -9.74                         +2.95

How well did the EFF do ?

Most breakaways from the ANC (the PAC, UDM, Cope) have enjoyed an initial popularity but have failed to sustain it and rapidly fall into oblivion. But the EFF went from 6.35% in 2014 to 8.19% in 2016. This was far from the doubling or trebling that they had promised – and no doubt this was a sharp lesson to Malema and Shivambu about the hard realities of the slog they are now engaged upon. But on the other hand they are now king-makers in Tshwane and Jo'burg, as also in Rustenburg and Thabazimbi. In addition they have clearly built a base of sorts in Marikana and Limpopo province.

Again, the EFF had made extravagant claims about their likely support in Limpopo and even at the last minute there were reports of polls showing them at 30% in the province. In fact they got 16.73%, a testament to the power of the tribal base, for Malema comes from Polokwane in Limpopo. This was, indeed, where the EFF peaked, with 28.21% of the vote. The DA took 10.98% (down from 11.38% in 2011) and the ANC 57.16%, down from 79.27% in 2011.

The EFF also managed to produce hung councils in two towns in Limpopo, something previously unimaginable in this ultra-safe ANC province:

 - Modimolle/Mookgopong where the DA won 7 seats, the EFF 6, and Others 2 against the ANC's 13. This municipality, which includes Naboomspruit and Nylstroom, has seen violent ANC factionalism, large scale protests occasioned by water cut-offs and in general bad governance.

- Thabazimbi, where the EFF is disputing the result. In this iron-mining centre (2m. tonnes are mined annually here), the ANC won 45.8% of the vote, the DA 22.21%, the EFF 20.24% and the Thabazimbi Residents' Association 8.14%.

Logically, both councils might thus produce DA mayors, an ironic result of the EFF surge. Overall, the EFF displaced the DA as the main Opposition in the province.

The EFF's most dramatic result, however, was in Rustenburg, a tribute to how successfully the party has identified itself with the martyrs of Marikana. Here the ANC won 43 out of the 89 seats in play, the EFF 24, the DA 14 and Others 8. Thus if the latter three were to get together this could have become the only municipality in the country with an EFF mayor. Perhaps most impressive is the fact that the EFF has built on the 20.22% it won here in 2014 and this time took 26.76%.

The DA advance was uneven

It did best of all in Nelson Mandela Bay (PE) where it gained 6.58% and the ANC lost 10.99%. The EFF took 5.12% and the United Front 0.94%. (Numsa, the UF's sponsors, thus received a hard reception into the world of party politics where the fabled political influence of the unions just melted away.) Even so, the DA's progress in NMB over the past decade has been striking – it has almost doubled its share of the vote.

Table 2. ANC  and DA municipal vote shares (%) in Nelson Mandela Bay, 2006-16

                    2006                    2011                           2016

ANC           66.53                   51.91                          40.99

DA              24.39                   40.13                          46.65

The DA also captured Kouga (Jeffery's Bay, Humansdorp), so it now has a commanding bridgehead in the Eastern Cape. Overall in that province the DA won 19.7%, up from 16.4% (+3.3%), while the ANC fell from 72.35% to 65.31% (-7.04%). The EFF took 5.18%. However, the DA's further progress here may be difficult: its most obvious target, Buffalo City (East London) remains ANC by a heavy margin.

The DA's other great success was in the Western Cape where its vote rose from 58.11% to 63.33% (+5.22) and the ANC fell from 34.1% to 26.22% (-7.88). The EFF took 2.81%. What this meant was that the DA gained control (sometimes in alliance with third parties) in great swathes of the province where its writ did not run before. Most notably, it gained control of the last municipality in the province in which the ANC had had a majority, Beaufort West – this despite the boast of the ANC mayor, Truman Prince that “only God can remove me from office”. In fact, as Table 3 shows, the DA has shown a remarkable progression in the town, perhaps through divine providence.

Table 3. ANC and DA municipal vote shares (%), Beaufort West 2006-16

                                  2006                2011                       2016

ANC                         40.56               50.17                      42.21

DA                            17.24               41.01                      48.99

Beaufort West is one of 8 hung municipalities in the Western Cape. All the others have DA majorities. Most of the 8 will end up with DA-led coalitions. Without doubt what we are looking at here is the progression of the DA among the province's Coloured electorate. Having won over this group first in Cape Town, the “DA effect” is gradually radiating out from the provincial capital into the smaller towns and the countryside.

With the conquest of Beaufort West it has now spread to the borders of the Northern Cape and the question is how far the DA can now spread into that province.  In fact in 2016 the ANC fell from 63.78% to 58.26% in the Northern Cape while the DA rose from 22.1% to 24.85%, and the EFF took 8.6%. COPE fell here from 11.79% to 2.54%. Thus the Northern Cape, despite its unpromising sociological profile, has already become the third biggest DA target after the Western Cape and Gauteng.

Table 4. DA and ANC municipal vote shares (%) in the Northern Cape, 2006-16

                                       2006               2011                     2016

ANC                               69.95               63.12                   58.26

DA                                  13.63               21.99                   24.85

The DA advance means that there are now 3 hung councils in the province: 

Nama Khoi (Springbok) where in 2011 the DA (35%)and COPE (15%) had actually won control of the council from the ANC (49%). In 2016, however, the COPE vote collapsed and the ANC picked up enough of it to rise to 56% in 2014. This produced a hung council with the ANC taking 8 seats, the DA 7 and Others 2.

Ubuntu (Victoria West) where the ANC now has 3 seats, the DA 2, and Others 2. (In 2011 the ANC had led the DA with 4 seats to 3, with 1 Other, producing a dead-heat. The ANC took power on a coin toss. This time that will not be necessary.)

Finally in Kgatelopele (Danielskuil) the ANC had led the DA by 5 seats to 2 with 1 Other, thus producing an easy ANC majority. 2016 ended that leaving the ANC with 3 seats to the DA's 2 with 2 Others.

The Cape Town wave

The DA's greatest success in the Western Cape came with its increased majority in Cape Town itself.

Table 5. DA and ANC municipal vote shares (%) in Cape Town, 2006-16

                   2006                     2011                     2016

ANC           37.91                    32.8                     24.33

DA              41.85                   60.92                    66.62

The Cape Town result is notable for several reasons. First, the DA result came as no surprise after a sustained period in which Capetonians have been made thoroughly aware by the City and the local media that Cape Town is enjoying the fruits of good governance – higher economic growth and lower unemployment, that there is a continuing inflow of migrants to the city both of affluent “semigrants” from the rest of South Africa and penniless work-seekers from the Eastern Cape, Zimbabwe, the DRC and elsewhere.

Its boom status is signalled by record house prices in Clifton/Camps Bay and above average house price increases elsewhere despite the zero-growth economy. The general feeling was “why change a winning formula?” and sufficient citizens have memories of the chaotic conditions reigning under the last ANC administration in 2001-06 for that to be a powerful theme.

Second, the ANC had always previously chosen a Coloured candidate for mayor because it was accepted that no party could win without pivotal Coloured support, yet this time the ANC selected an African candidate, Xolani Sotashe, a symbolic admission that it had given up all hope of winning and was now concerned merely to defend its heartland of Khayelitsha. For the DA attacked hard this time and achieved new highs in a number of Khayelitsha wards. Everywhere the ANC vote crumbled while the DA racked up record majorities in the suburbs.

The real point about the DA landslide is best seen through the demographic data. Cape Town was for long kept a city of whites and Coloureds. In 1944 the whites constituted 47% of the population, the Coloureds 46%, Asians 1% and Africans under 6%. With the collapse of the pass laws and then of apartheid there has been a huge African inflow so that by 2011 the population was 15.7% white, 42.4% Coloured, 38.6% African and 1.4% Asian with 1.9% Other.

Without doubt that inflow has continued apace over the last five years so that by 2016 it is likely that the biggest population group in Cape Town was Africans. One might have imagined that this would bring a ballooning ANC vote and yet the opposite has occurred, testimony to how completely the city now lives within a DA bubble.

The battle for Gauteng

Most attention was, naturally, on the battles in Gauteng, particularly those for the three big metros of Johannesburg, Ekurhuleni and Tshwane. Across the province as a whole the ANC fell from 60.48% to 45.86% – a loss of 14.62%. This was mainly due to the EFF which took 11.36%. The DA improved from 33.31% to 37.21% (+3.9%). Thus of Gauteng's nine councils, the DA easily held Midvaal and four more were hung (Johannesburg, Tshwane, Ekurhuleni and Mogale City (Krugersdorp)).

In the Big Three:-

JO'BURG. The DA went from 34.62% in 2011 to 38.37% (+3.75%) while the ANC fell from 58.56% to 44.55% (-14.01%) and the EFF took 11.09%.

TSHWANE: The DA rose from 38.74% in 2011 to 43.11% (+4.37%) while the ANC fell from 56.46% to 41.22% (-15.24%) and the EFF took 11.7%.

EKURHULENI: The DA rose from 30.29% to 34.15% (+3.86%), while the ANC fell from 61.63% to 48.64% (-12.99%) and the EFF took 11.23%.

Overall, the swing in fortunes was the highest of all in these large metropoles. The reason for this lay not only in the high EFF vote but in the fact that the DA gain here was considerably greater than the national average. This would appear to be in line with the phenomenon observed elsewhere (eg Paris, London) whereby the capital city exaggerates the national trend. However, only in Tshwane did the DA pull ahead of the ANC with a particularly large swing. 

Click here to read the second part of the article.