Wednesday's Local Government Elections will provide a fascinating report card on South Africa's maturing democracy, and the most important issue to watch is the question of the black floating vote.

For South Africa's democracy to flourish it is imperative that our elections become more issue-based and less driven by identity politics. Given the country's largely racialist political past, and the way this has allowed race and class to overlap, such an escape from the politics of race was always going to be gradual at best and impossible at worst.

The ANC's woeful service delivery record on local government level has ensured that a floating vote of disenchanted black voters exists for this election. The critical questions are what the size of this voting bloc is, what the extent of its unhappiness is, and how this unhappiness will manifest itself on election day.

Pollsters put the size of the discontented black vote at around 20% of black voters, though I'd argue the difficulty of polling in our country with its diverse social strata and big lie factor makes such a figure a bit of a thumb-suck.

One thing there is very little doubt about, however, is that there are many black voters who are disappointed with the ANC.

The trend has manifested itself in many ways: Service delivery protests, the sometimes violent disagreements about ANC candidates lists, the rise in the number of independent candidates formerly with the ANC, strong indications of a higher than expected black stayaway vote and finally, some indications of growing black DA support are some reasons to follow this year's election results closely.

I'd guess the most contested of these manifestations I have just advanced would be the final one: indications of black DA support, which in the past has been minimal, were calculated at between 2% and 3% in the 2009 national elections.

All I can say is that my colleagues and I at Rapport have covered sixteen municipalities in some depth, and have chanced upon black DA supporters in thirteen of these. When we did the same exercise in 2006 and 2009, we never met one. I know this is hardly empirical and I don't want to split hairs or get into arguments over it. We'll know by next week. I'm just raising it because it was such a noticeable change.

That aside, the unhappiness about service delivery, candidates' selection and voter apathy must surely be common cause. The consequences on election day will follow one of three routes.

Firstly, the unhappy voter can opt to give the ANC another chance, secondly they can stay away or thirdly they can change their vote.

In a brand-conservative country like ours people see it as a betrayal rather than a decision when you change preference. So, the expected preference for the floating black vote, given that it was previously an ANC voting bloc, will be to give the ANC another chance. This, I believe, will indeed be the majority choice.

The second option would be to not vote at all. When, after the election, we analyse whether this has happened, it would be important to compare apples with apples. This is a local government election where, as a global and local phenomenon, turnout is always much lower than for a national or provincial poll.

Turnout must be compared to 2006, and thus anything above 50% would be good turnout for a municipal election. Having said that, it seems a differential might come into operation. A voting differential occurs when a party mobilises its core base better than its opponent does.

In a low turnout, the voting differential is important, and I predict the turnout in areas with a DA supporting majority will be much higher than in predominantly ANC areas. This could have a huge influence in metros such as Cape Town and Tshwane.

If the floating black vote, made up of predominantly of former ANC voters, does stay away, it would obviously help all opposition parties in a proportional system. The extent to which this will happen will only emerge with the results by Thursday. However, I don't expect too much of it, because South Africans tend to get caught up in the moment and vote when the day is there. It is a novel privilege, and is valued.

Moving along to the third and truly interesting question: To what extent will the floating black vote opt to support the opposition? This really is the jackpot option, and I think we need to distinguish between two types of floating black voter: the middle class and the proletariat.

The proletariat are unhappy with service delivery as they are dependent on the state and they want more than they are getting, without anything to offer in return but the vote. I would therefore argue that it would be hard for opposition parties to honestly promise them much (given the constraints on the fiscus). If they choose not to vote ANC they would be more likely to stay away than change their vote.

This raises the issue of people losing faith in democracy as a tool for self-improvement, which in a country with the revolutionary potential our wealth gap represents is cardinal, but on which merits its own discussion.

The second grouping of middle class black floating voters are irritated with potholes, traffic lights that are out of order and general wasting of money by ANC councils, just as all taxpayers are. The DA's service delivery record is better than the ANC's, which is why they are running on that election message. The ANC has, in response, not raised a single example of a council they run exceptionally well.

The ANC knows, however, that its history as a liberation movement, and the floating black vote's historical allegiance to it, is its strongest drawcard, especially in a brand-conservative country. This explains much of their campaign. The choice between history and service delivery is the choice confronting floating middle class black voters, and their choice will be fascinating. 

All said, the middle class might be more likely to shift their vote to the DA than the proletariat because the DA spends taxes better. If so, this will show up in the metro votes, especially Johannesburg.

The long and the short of it is that this election will show whether the black floating vote will give the ANC another chance, stay away or switch their vote. I expect all three to happen, in that order of prevalence.

As for the results, I still believe the ANC and DA can both grow from their 2009 positions. The DA will benefit from people impressed by its own record and disgusted by the ANC's service delivery record, but the ANC must surely benefit from the weakening of Cope and the IFP.

I therefore fully expect the DA to grow, mostly in urban areas, the ANC to grow mostly in rural areas, and for South Africa to move closer to a two-party state in Wednesday's election. 

Jan-Jan Joubert is political editor of Rapport. This article was published with the assistance of the Friedrich-Naumann-Stiftung für die Freiheit.

Click here to sign up to receive our free daily headline email newsletter