ANC has every reason for alarm over losing control in 2019 election
After Friday night’s dramatic and farreaching Cabinet reshuffle much has been said about the ANC’s electoral prospects for 2019 especially given widespread erosion of support for the party in the 2016 local government elections.
In response to the reshuffle the Thabo Mbeki Foundation issued a statement arguing that "concern and unease [about the state of the nation] was expressed by the general public during the August 2016 local government elections which among others resulted in significant loss of support for the ANC the dominant governing party … since 1994".
Dismissed finance minister Pravin Gordhan has also alluded to the loss of confidence apparent in last August’s elections. While the ANC acknowledged the loss somewhat cryptically at the time of the polls — and undertook to engage in a period of introspection — how worried should the party be about its support base? Especially now that three of the top six have expressed concern about the president’s most recent actions and as disunity poses a very real risk?
Data trends reveal a bleak picture of deteriorating support. At a national level the ANC’s performance in the 2016 local government elections was nothing short of disastrous. While the party polled at consistently above 60% between 2000 and 2014 support plummeted by 12 percentage points between 2014 and 2016 from 66% in 2014 to 54% in two short years.
To understand clearly the national tipping point below a 50% majority for the ANC it should be considered that of the 55.6million South Africans estimated by the 2016 Community Survey 13.4 million reside in Gauteng — almost one in four.
To make matters worse disclosures of a "black ops" programme alleged to have cost R50m with the objective of influencing voters in the local government elections by employing false news and propaganda prompted concerns as to why such a strategy if it existed had been contemplated at all.
Perhaps the answer lies in the fact that downward trends were already evident and feared to be difficult to reverse.
The rise of the EFF and its ability to quickly carve out 6% of support in the 2014 provincial poll — more comparable to local government election results than national due to requirements for voter registration — equated to a similar drop in support for the ANC in many provinces especially Limpopo the North West and Gauteng.
The EFF clearly represented a new and important variable that contributed to a trend of downward support for the ANC.
As inconceivable as it might have seemed a few years ago pressure on ANC support was seen in 2014 in key metros such as Nelson Mandela Bay Tshwane Johannesburg and Ekurhuleni as well as in Gauteng generally. Between 2014 and 2016 things got worse for the ANC with 10% of its support shed in five provinces: the Eastern Cape Free State the North West Gauteng and the Western Cape. The ANC did worst in the Western Cape with support dropping 12 percentage points between 2014 and 2016. Support for the ANC between 2000 and 2016 evaporated plunging almost 18 points — from 44% to a little more than a quarter of votes (26%).
The Western Cape is an example of just how badly things could go wrong for the ANC should the trend not be halted. Support for the party has shrivelled over 16 years and the 2016 polls showed that it could no longer mount serious competition for a 50% majority in any but one or two municipalities in the province. Leadership of the province itself had long been lost and it is now difficult to remember that it was ever ANCcontrolled. Will the same be said of other provinces before too long?
Simple linear trend lines suggest cause for serious concern at Luthuli House. At the same rate (not factoring in any accelerating deterioration of support) a reduction in support for the ANC between the 2011 and 2016 local government elections would leave the party without a majority in 2019 in the North West Gauteng and the Western Cape and also most significantly nationally.
To understand clearly the national tipping point below a 50% majority for the ANC it should be considered that of the 55.6million South Africans estimated by the 2016 Community Survey 13.4million reside in Gauteng — almost one in four — so it is not a constituency that should be dismissed lightly or underestimated in national elections.
Adding those provinces where the ANC has already lost or risks losing its majority — the North West Gauteng and Western Cape — there is a collective population representing 42% of South Africans. This clustering effect of the population can be lost when viewing provincial or local government election results on a onedimensional national map but understanding it could very well point to who sits in the Union Buildings.
The prospect of losing SA has never been one the ANC has had to face up to. While this analysis represents just two data sets (2011 and 2016 local government elections) and is based on a simple mathematical extrapolation it is a consistent downward trend that the ANC dismisses at its peril especially in such turbulent times.
It is slightly less dire for the ANC if 2014 provincial data are taken into account in the 2019 projection. "Only" in Gauteng and the Western Cape would the ANC lose its majority — the case in 2016’s local government elections — but national support would be perilously close to the 50% line not losing sight of the fact that Gauteng and the Western Cape are home to 35% of all South Africans (roughly one in three).
Another perhaps more meaningful way to analyse recent election data is to consider trend lines from when the EFF started to contest elections in the 2014 provincial and national and the 2016 local government elections. Opinions differ on whether the EFF will continue to galvanise support or whether this will plateau but like the 2011 and 2016 local government trend lines 2014 and 2016 election projections suggest the ANC will lose its majority in North West Gauteng and Western Cape as well as nationally and will face serious vulnerability in the Free State and KwaZuluNatal — although the latter is particularly difficult to extrapolate given that the National Freedom Party did not contest 2016 local government elections.
These trend lines even though just simple mathematical equations suggest that even before the current crisis faced by the ANC the party was likely to face difficult polls in 2019. Of course such analysis is by its nature linear and simplistic and trends can be reversed given material changes in determining variables. But they can also be accelerated with support going further downwards.
From a modelling point of view politics is an infinitely complicated set of variables and these can converge and contradict each other in many ways. There are many months ahead of the ANC’s national conference at the end of 2017 never mind national and provincial polls in 2019 and there is much scope for variables affecting the ANC’s performance to improve or deteriorate in the eyes of the electorate.
The time has come for the ANC to decide what these material variables will be; its future as the governing political party hinges on this.