More significant than John Steenhuisen’s comfortable election as DA Leader this weekend was his announcement of a ‘moonshot pact’ against an ANC-EFF ‘doomsday coalition’. Steenhuisen’s proposal – to form a pre-election alliance with like-minded opposition parties – is an idea whose time has come.

The DA has long stated that its key objective is to bring the ANC below 50% of the vote. Since the DA does not have the numbers to obtain a clear majority, its intention to govern South Africa in a coalition has always been implicit in this goal. So, the question for the DA is not whether it should form a national coalition government, but whether it should form such a coalition before or after the election.

Now that Steenhuisen has answered this question, it is worth looking deeper into the modalities of pre-election coalitions and what this move could mean for our politics.

Pre-election coalitions are common across the world. In around half of the elections in OECD countries since World War II, at least one of the coalition partners announced their intention to form a government together before the campaign commenced.[i] Such pre-election coalitions tend to form when there is a belief amongst party leaders that such a coalition can influence the probability of electoral victory. [ii]

The ‘rainbow coalition’ in Kenya – although short-lived – is a good example of how the formation of a pre-election coalition can influence the outcome of an election.

In 1997, the second Kenyan election since the re-introduction of multiparty democracy in 1991, the governing Kenya African National Union (KANU) won 107 seats (51%). According to scholarly observers of the election, Kanu’s victory was attributed to the inability of opposition parties to unite against it:

“The opposition parties approached the 1992 and 1997 elections thoroughly fragmented…the parties were preoccupied with the desire to win and believed they could so on their own. As a result of the fragmentation, KANU and President Moi were victorious and were able to rule the country with a slim majority.”[iii]

Importantly, opposition parties learned the lessons of the 1997 election:

“It was obvious to all that their chances of winning would be slim if they did not form a coalition in the face of the formidable electoral machinery of the incumbent KANU. Citizens’ expectations and their message to the opposition parties were that they must unite in order to win the elections.”[iv]

The National Rainbow Coalition (NARC) was formed in September 2002. It was a pre-election coalition between the Liberal Democratic Party and the National Alliance Party of Kenya (itself an amalgamation of the Democratic Party of Kenya with 13 small parties). The formation of NARC was followed by “a nationwide tour to popularize the party, holding rallies which attracted unprecedented crowds.”[v] The NARC went on to win the 2002 parliamentary election with 125 out of 210 parliamentary seats (59,5%) to KANU’s 64 (30,5%).[vi]

In the Kenyan case, the first-past-the-post electoral system provided a powerful incentive for a pre-election coalition. All coalition partners stood to gain by uniting behind the strongest coalition party candidate in a voting district because it avoided splitting the vote, thus ensuring that more seats were won by the coalition.

Indeed, some of the literature on pre-election coalitions suggests that the electoral system is the biggest determinant of whether a pre-election coalition will form and whether it will be successful. For example, writing in the British Journal of Political Science, Strøm, Budge and Laver state:

“Systems not based on PR lists tend to force parties to coalesce before elections in order to exploit electoral economies of scale. The more disproportional the electoral system, the greater the incentives for pre-electoral alliances.”[vii]

In proportional systems there is a weaker incentive to form pre-election coalitions because nearly every vote counts towards a party’s final tally, and vote splitting is therefore less of a factor. However, while it may be true that the incentive to form pre-election coalitions is stronger for parties in first-past-the-post systems, this does not mean that the incentive to form pre-election coalitions is absent in proportional systems such as ours.

In practice, pre-election coalitions do form in proportional voting systems. For example, pre-election coalitions formed in six of the sixteen Dutch elections and in ten of the fifteen Austrian elections that occurred between 1946 and 1998. Of these, five Dutch and nine Austrian pre-election coalitions formed a government following the election.[viii]

There are at least six good reasons to explore a pre-election ‘rainbow coalition’ in South Africa:

1. A pre-election coalition may encourage more opposition voters to turn out than ANC voters. Evidence from around the world suggests that pre-election coalitions boost turnout by offering meaningful choices and clear potential outcomes.[ix]

2. A pre-election coalition would necessarily build in non-aggression/cooperation agreements with coalition partners. Instead of expending resources fighting each other, opposition party resources could be pooled to weaken the ANC.

3. A pre-election coalition would avoid the media chatter that is likely to dominate the election campaign around who will form a coalition with whom. This frees up opposition parties to focus on one simple message: the need to defeat the ANC to save South Africa.

4. A pre-election coalition agreement would avoid the post-election horse-trading that gives small parties disproportionate power as ‘kingmakers’. Evidence suggests that cabinets formed by pre-electoral coalitions tend to distribute posts more proportionally than other coalition governments.[x]

5.  A pre-election coalition helps prevent the ANC from co-opting opposition parties after the election in return for cabinet posts and other concessions.

6. In the future, a pre-election coalition could serve as a catalyst for a split in the ANC, as ANC ‘moderates’ realise there is another viable electoral vehicle available to them.

Of course, as John Steenhuisen will be acutely aware, there are also dangers that the DA must find ways to attenuate. For example, some loyal DA voters may withhold their vote for the DA if they know in advance that it plans to go into a coalition with parties they strongly dislike.

There may be other DA voters who have hitherto only voted for the DA because it is the biggest opposition party. With all opposition votes being put into the same pot, such voters may be tempted to put their X next to another member of the coalition with whom they have a closer ideological, ethnic or religious affinity.

Whether the benefits of a pre-election coalition outweigh the costs is a calculation that each political party must make. However, in seizing the initiative, Steenhuisen has boldly positioned himself as the driver of change. How other parties respond will depend on the lessons learned from coalition experiments in municipalities and, ultimately, their leaders’ emotional and political maturity.

Notes