Disc Dem

Disc Dem

Wednesday 26 August 2015

The Political Cost of Local Government is no longer Affordable

Our electoral system has been the subject of much discussion and debate with many people favouring a mixed constituency/proportional representation model. 

The mixed model is already used for Local Government elections in South Africa, but not for National Assembly elections.  However, looking at the numbers from previous Local Government elections, it is easy to conclude that Proportional Representation does not achieve its intended purpose, and is not affordable at local government level. 


The political cost of local government is far higher than its contribution to service delivery warrants, not only relative to the actual cost of paying too many politicians, but also to the indirect costs of corruption, attributable to the lack of accountability engendered by the patronage-based closed party list system. 

So, why doesn't our electoral system  work as was constitutionally intended? First of all, "one man, one vote” is a myth. It is worth repeating that we actually get three votes in local government elections if we are not with a Metro Council.  One constituency (Ward) vote for the individual candidate of choice, one Local Council PR party vote, and one District Council PR party vote. Metro Councils have one constituency (Ward) vote and one PR party vote.

Half the seats in local and metro councils result from direct constituency votes, and half are allocated from the PR votes.  So far, so good, but the value of proportional representation is only realised if constituencies have widely varying numbers of eligible voters, and are geographically widespread.[*]

As required by Schedule 1 of the Municipal Structures Act of 1988, the Municipal Demarcation Board (MDB) ensures that this does not happen.  The total number of councillors (Ward + PR) for a municipality are determined by each Provincial MEC for Local Government. This determination is then used by the MDB to calculate the number of Wards by dividing the total number of councillors for the municipality by two. 

The target number of registered voters per Ward is then calculated by dividing the total number of registered voters in the municipality by the number of Wards. The result of this calculation is used by the MDB for the physical configuration of Ward boundaries to ensure that every Ward has more or less the same number of registered voters.  

The maths involved are not complex, in fact it is a simple case of 1 + 1 = 2. 

As our electorate tends to be politically unsophisticated,  1 Ward vote + 1 Proportional vote will generally = 2 votes for the same party. 

Assuming that, as is generally the case, everyone casts their two votes in the same direction, and all Wards have more or less the same number of voters, then all we have created is a “winner takes all” outcome using two votes instead of one, with an end-result of appointing two councillors where actually one would suffice.

It is argued that proportional representation gives smaller parties the opportunity to participate and be heard but, in reality, minority parties have no influence on decision-making where a majority of seats is held by a single party.  In addition, there are many other avenues for participation that do not involve having to pay an ineffective, unnecessary councillor.

Proportional representation benefits minority parties only when a deadlock occurs between major parties, and a coalition is needed in order to govern.  

All that such coalition politics succeeds in doing is to put control of governance into the hands of a minority – sound familiar?  A classic example of this is the Laingsburg Municipality where the ANC and DA were tied on both Ward and Proportional seats, the balance of power being held by COPE with a single proportional seat, gained with the princely total of 556 votes.[†] 

In summary, the need for proportional representation at local government level is largely negated by the spatial design of Wards within each municipality, compounded by the generally low level of political sophistication of the electorate. 

Apart from not fulfilling any practical purpose, the system also provides an incubator for corruption  by removing accountability to the electorate. Candidates only need their party to place them high enough on their closed PR list to be appointed as a Councillor, regardless of their failure as a Ward candidate at the polls. Who knows what deals are done to secure these places? 

Aside from theoretical and generally idealistic arguments regarding the pro’s and con’s of proportional representation at Local Government level, a purely practical standpoint is that we simply cannot afford to keep paying so many politicians before even one cent is spent on service delivery.  


Analyses of previous local government elections have categorically proven that proportional seats make no difference to the majority standing of parties in any municipality countrywide (See http://bit.ly/1M6PNXf. So why persevere with such an idealistic system that is so clearly inappropriate, ineffective, unduly expensive and, most likely, corrupt?




* A classic example can be found in analysing the last UK national elections. With no proportional representation system, the UKIP party with around 13% of votes won only one seat. The Scottish National Party, with less than 5% of votes, is now the third largest party in Parliament with 56 seats, owing solely to their regional popularity in Scotland.   The UKIP party had national appeal, whereas the SNP was only regionally attractive, so the UKIP party lost out to a combination of variable constituency populations, and the geographically widespread distribution of their supporters.

[†] An interesting thought on this is that without proportional representation, the DA and ANC would have been deadlocked, so would be forced to work together to keep their municipality functioning. Perhaps, if given a chance, pragmatism and compromise could beat the norm of confrontational politics?