As the world confronts the biggest year of elections in history – or as some have termed it, the biggest test of the health of democracy – these questions have never been more important. More than four billion people – more than half of the world’s population across more than 40 countries – will go to the polls in 2024. What will drive them to first vote in the first place, and then to choose which box to cross in the voting booth?

As Valentina Bali and Lindon Robison of the University of Michigan have noted, this puzzle has been tackled by scholars at length yet remains unsettled.

In short, the academic theories range from more selfishly driven motivations to vote – for example, benefits potentially forthcoming from an elected representative – to those that are more altruistic and relate to a sense of civic duty. The truth is surely some mix of both.

These varying factors are most compelling when large-scale shifts in voting behaviour occur. For example, in a survey released last week by the New York Times, it is clear that exactly such a secular shift is happening in the US among non-white voters. The survey shows that Biden is currently leading Trump by only 56% to 44% among voters of colour. In 2020, the last time these two men ran for the White House, the gap between them among the same electorate was almost 50 percentage points.

Interestingly this is not a recent phenomenon. Data from previous election surveys show that the Democrats’ advantage over the Republicans among black, Latino and Asian Americans peaked in the 1960s and has been steadily declining.

There are three explanations offered. 

First, it is due to fading memories and weakening ties. Americans of colour who lived through the civil rights era still support the Democrats but younger generations are wavering as those years become more distant.

Second is the weakening correlation between income and voter choice in US politics. As John Burn-Murdoch points out in the Financial Times, the image of the GOP as the party of “wealthy country club elites is dimming, opening the door to working- and middle-class voters of all ethnicities”. A similar trend was noticed in the UK in the last election, as the “red wall” of traditionally safe northern Labour seats voted overwhelmingly for the Tories.

Finally, many of America’s non-white voters have long held much more conservative views than their voting patterns would suggest. The shift seen over the past 50 years therefore is not so much natural Democrats becoming disillusioned but natural Republicans realising they have been voting for the wrong party.

Read more in Daily Maverick: 2024 elections

This is shown by US political scientists Ismail K White and Chryl N Laird whose book Steadfast Democrats demonstrates the mismatch between many black Americans’ policy preferences and votes. This is particularly evident on issues such as gay rights or gun control, where much larger numbers of black, Latino and Asian conservatives have tended to vote Democrat. History, culture and community have long overridden this misalignment between non-white conservatives and choice of vote.

The shift among South African voters

As South Africa faces its most important election since 1994 the latest polls, carried out by the Brenthurst Foundation and covered by Daily Maverick, indicate that a shift may be happening for those who have tended to vote for the ANC. Evidence points to sharply lower support for the ANC and substantially higher support for the new uMkhonto Wesizwe (MK) party. These explanations from the US, of why people change voting allegiances, offer some insights in the South African context.

First, 30 years on from democracy, memories of the struggle years have faded and ties to the parties of the fight for human rights and equality have weakened. The so-called born free generation, such as the mythical girl Tintswalo invoked by President Cyril Ramaphosa in his State of the Nation Address, simply do not have the same allegiances to the ANC as those who came before.

Second is the correlation between voting choice and income. Traditionally, poor and rural communities have tended to vote ANC. It is unclear whether this trend will hold, or if ANC support in its traditional heartlands will be eroded by other parties, such as MK.

Finally, outside of societal or legacy pressures to vote ANC, it is unclear what the actual voter preferences of South Africans are, particularly as they pertain to social issues. How conservative are South Africans on issues such as gay rights and abortion, and how far is the median voter away from the quasi-extreme liberalism enshrined in the Constitution and largely espoused by the ANC since 1994? In the future this will raise interesting tensions and sources of instability. How will these strains between electorate and Constitution be resolved, or perhaps how will they be instrumentalised by smart and potentially disreputable political entrepreneurs?

Jonny Steinberg, in an insightful piece for Business Day, makes one further and more concerning point. South Africa’s electoral system of proportional representation does the opposite of providing a countervailing structure to this trend of fragmentation. Instead, by giving voice to minority parties, it “enables the crazies to bloom”.

While these trends show a deepening and maturing of South African democracy, they also portend that the first 30 years of South Africa’s democracy will almost certainly be viewed as a period of quite remarkable stability and calm compared with whatever might come next. The next chapter is almost certainly going to be more complicated, less predictable and more prone to voter manipulation in pursuit of nefarious, self-motivated ends than this one. The honeymoon is over. DM