By-elections and polls at odds

The ruling African National Congress is expected to emerge with guns blazing from its policy review conference next month in preparation for next year’s all-important local government elections. It seems the party will need some good ammo to talk away the effects of widespread service delivery failure, as the results of municipal by-elections this past year have shown that the Democratic Alliance (DA) has steadfastly gained ground mostly at the ANC’s expense, contrary to what opinion polls seem to indicate.

The DA gained particularly in its “home province” of the Western Cape, but also in some traditionally ANC-dominated provinces. Some of the biggest DA gains came in predominantly black areas. In fact, the only recent winning stand from the ANC recently has been in the virtual world of opinion polls – notably the latest Ipsos-Markinor poll.

But then, neither municipal by-elections nor opinion polls are ideal barometers of what to expect in an actual election. Come election day, there will always be surprises.

For now, however, the ANC must be concerned, and it shows in much of its rhetorical war-talk.

The results in 26 municipal by-elections held in several provinces two weeks ago indeed echo those of all held over the past year. The major trends show that the ANC and other parties continue losing ground to the DA in the Western Cape, despite the fact that before the most recent by-elections, the ANC dispatched  prominent ANC leaders such as the minister responsible for the National Planning Commission Trevor Manuel, former Education minister Naledi Pandor and Police Minister Nathi Mthethwa to help campaign.

In KwaZulu-Natal, the Inkatha Freedom Party (IFP) continues to lose ground to the ANC, while in the rest of the country the ANC more or less manages to hold its ground, with the DA making some inroads here and there.

The statistics of 28 by-elections in May show that the ANC won 18 or 64% of the wards, lost three to the DA (two in the Western Cape and one in the Free State) and one to an independent candidate (in Limpopo). It also gained one ward from the IFP in KwaZulu-Natal. For the ANC, this represented an overall loss of 10.7% in terms of the total number of contested wards.

Including its three gains, the DA won five wards, representing almost 18% of the contested wards. It also represented a 10.7% growth among the wards contested. The IFP won in three wards and lost one – all in KwaZulu-Natal – representing a 25% loss compared to its position before the by-elections.

Add to this the continuing, often violent, service delivery protests in most parts of the country, and the message to the governing ANC is probably even more serious than the formal statistics of the by-elections would suggest.

Altogether, over the past year, the DA gained seven seats from the ANC in the Western Cape, one in the Eastern Cape, one in the Free State and one in Gauteng as part of a coalition.

Its most important gains were the two previously ANC-controlled wards it won in black townships in the Western Cape - that despite the ANC Youth League (ANCYL) trying to capitalise at the DA’s expense on the “open toilet” saga in the most recent by-elections. The DA could be further strengthened before the 2011 local elections if the current initiative to form a coalition with the Independent Democrats (ID) is concluded successfully. It should benefit the DA, particularly in the Western Cape, Northern Cape and Free State.

Under a proportional voting system at national level, the role of local government as an indicator of trends becomes much more important than might otherwise have been the case. Local government has, in most instances, become the only direct exposure the broader voting public has to representatives of political parties.

The good news in a broader perspective is that apparently the South African public has not yet lost faith in the democratic process.

The trends and voting demographics of these by-elections have brought to the fore another important trend: South Africa may slowly be moving away from its race-dominated politics and race-based parties. However, one swallow - namely the DA taking two wards in Cape Town in which not a single white vote was cast - does not make a summer. But at least a start seems to have been made toward issue- and merit-driven politics.

The fact that the ANC’s much vaunted turnaround strategy for local government has so far produced little, if any, tangible results, could come back to haunt the ruling party in next year’s elections.

Meanwhile, the results of the latest Ipsos-Markinor poll show a somewhat different picture. This poll was conducted in May 2010 with a “nationally representative study of 3 386 South Africans of 18 years and older”. Ipsos-Markinor claims the sample “accurately mirrors the adult South African population and has a margin of error between 0.7% and 1.67%”.

The major finding of this poll is what it sees as the continued one-party dominance by the ANC with the party attracting a two-thirds majority, despite a 5% fall in support since the Ipsos-Markinor’s December 2009 poll. In the April 2009 general election, the ANC obtained 65.9% of the vote. In the December 2009 Ipsos-Markinor poll, support for the ANC stood at 70.5%. And in the latest Ipsos-Markinor poll, support for the ANC had fallen to 66%. While Ipsos-Markinor’s political analysts Helen Macdonald calls this 5.5% loss of support a slight decrease, such a shift is actually quite substantial for any political party in any scenario.

The poll puts the DA’s support at 13%, down 3.6% from its April 2009 position. However, these numbers do not reflect what has been happening in municipal by-elections in the real world since last year.

Other rather questionable results of this Ipsos-Markinor poll show increases for the Congress of the People (Cope), the IFP, African Christian Democratic Party (ACDP), United Christian Democratic Party (UCDP), and Pan-Africanist Congress (PAC) since December last year. While the increases in the case of the ACDP, UCDP and PAC are relatively small, these parties are nonetheless largely political non-entities or spent forces that are unlikely to show growth. And the growth suggested for Cope and the IFP certainly do not correspond with developments on the ground since December.