HomeOpinionProportional representation: The unknown unknowns

Proportional representation: The unknown unknowns

“There are known knowns. These are things we know that we know. There are known unknowns. That is to say, there are things that we know we don’t know. But there are also unknown unknowns. There are things we don’t know we don’t know.”

Opinion by Derek Matyszak

This well-known quote from former United States secretary of Defence Donald Rumsfeld comes to mind when considering the provisions relating to proportional representation which appear in the proposed constitution for the country.

It is apparent, at least as far as the main political parties are concerned, that issues relating to proportional representation presently fall with the group of unknown unknowns.

The proposed constitution beguilingly provides that 60 of the 80 members of the senate (six from each of the 10 provinces, elected in such a manner as to ensure a near gender parity), 60 of the 270 members of the National Assembly (again, six from each province, all of whom must be women) and 10 persons on each provincial council, will be elected on the basis of proportional representation.

For unknown reasons the members of the provincial council elected by proportional representation are omitted from this provision (Section 157).

The vacancies must be filled by persons belonging to the same political parties and who are of the same gender as those who previously held the seats.

The election of senators and provincial councillors is through a party-list system of proportional representation.

This is based on the votes cast for candidates representing political parties in each of the provinces in the general election for Members of the National Assembly and in which male and female candidates are listed alternately, every list being headed by a female candidate.

The formula used for those elected through proportional representation to the 60 seats reserved for women in the National Assembly is the same, bar one important difference — while there is obviously no need for a requirement that male and female candidates are listed alternately on the party lists, the relevant provision egregiously omits any mention of party lists entirely.

Nonetheless, the broad idea is this: suppose, merely by way of illustration, that of the total votes cast for the National Assembly seats in the constituencies lying in Masvingo province, MDC-T garners three-sixths, Zanu PF two sixths and MDC one-sixth.

Then the top three persons on the MDC-T list of candidates for the senate will gain seats, and the top two on the Zanu PF list for the Senate and the top candidate on the MDC list for the senate will also gain seats.

A similar process would be adopted for the provincial councils and the National Assembly — though in the latter instance, how the candidates are to be determined is left open in the absence of any mention of party lists.

At first glance this seems straight-forward and for this reason no further links seem to have been added to the chain of thought of the persons who negotiated the proposed constitution.

However, unknown to the political parties, the devil lies in the unknown details of proportional representation which can lead to fiendish complexities — and sophisticated and subtle voter decision making.

This is exacerbated by the fact that the way proportional representation has been structured under the proposed constitution means that a single vote for a National Assembly candidate will also determine the elector’s choice of reserved seats for women in the National Assembly, senatorial candidates and provincial councillors.

What if the voter adores the first candidate on the party senate list and abhors the candidate standing for the National Assembly? And what if a candidate beloved of the party, but less so of the electorate is put up as a party candidate for the National Assembly and placed top of the senate list?

Now have it, ought to be on roughly the same date.

The aleatory fates will never be so kind that when the ratio of constituency votes in each province per party is determined, the result will be integers — such as 3:2:1. The calculation will, to be certain, produce ratios of the following ilk — 2,7 seats to MDC-T; 2,3 seats to Zanu PF; 0,67 seats to MDC and 0,33 seats to Zapu.

What is to be done with the fractions? One could round MDC-T’s ratio up to the nearest integer and give it three seats and round Zanu PF’s down and give them two.

What of the minor party votes? One could put their votes together and let them have a combined seat, or one could say they didn’t get enough votes separately for a seat, so the calculation must be done afresh ignoring their votes and dividing the seats among those who did get enough votes for a seat.

The choice of system will affect the number of seats allocated to each party. One system might yield a 3:3 split of seats between MDC-T and Zanu PF, a 4:2 split, and yet another a 3:2 split, all based on the same number of votes cast for each party.

If this scenario is replicated for each of the 10 provinces, the difference could be an extra 10 seats in the National Assembly for one party if a system favourable to that party is deployed.

If the balance of power in parliament pertaining currently is replicated in the next election, these 10 seats could determine whether a party holds a majority of seats in parliament or not.

The proposed constitution blithely provides that an Act of Parliament shall provide “a system of proportional representation”. Which system is selected may be of vital importance. Some systems favour small parties; some favour large parties with a slight numerical advantage; and some favour large parties with a slight numerical disadvantage.

Once the parties have crunched the numbers from the last elections and made their prognoses for the next, there may be some hard bargaining and negotiating to be done over the choice of proportional representation system.

Parliament will stand automatically dissolved on June 29 2013 by which time this issue will need to be resolved so that the legislation is in place in time for the election, which has legalities.

A court order, which already reflects two applications for a postponement, requires the president to gazette the dates for three of several outstanding by-elections before the March 31. Although, the order only pertains to by-elections in three constituencies, the postponement of the by-elections has been granted on the basis of the President’s assertions that he intends to set the dates for a general election before March 31.

His proclamation in this regard must set the date for the sitting of the nomination courts no later than 21 days from the date of the proclamation and the elections must take place no later than 63 days.
Matyszak is a senior researcher with RAU, an independent non-governmental organisation.

Recent Posts

Stories you will enjoy

Recommended reading