A note on the limits of social commentary

In the analysis that I am going to be publishing over the next few months in advance of the 2024 elections I will suggest that there are quite a few plans and strategies afoot that are not currently, publicly recognised. I would not make these assertions without being fairly confident that they are accurate.

And I do have a track record of sorts in this area. Back in 2017, for instance, I wrote two articles on the dynamics playing out in the lead-up to the ANC’s elective conference (see here and here). In the second piece I made two key predictions that went contrary to the conventional wisdom at the time. First, I argued that the likelihood a split in the ANC was asymmetrical: if Ramaphosa lost the elective conference he and other key leaders would likely leave because they had outside options, whereas if he won his opponents would likely stay because they relied heavily on party and state patronage. Second, I argued that the concerns about too many presidential candidates ‘diluting’ Ramaphosa’s vote was unlikely. Rather, I suggested that was part of a deliberate strategy by Ramaphosa to bring in support from a range of constituencies and that the other candidates (besides his main rival Dlamini-Zuma) would ultimately throw their support behind his campaign. Both these key predictions turned out to be accurate.

However, it is important to remember that analysis and social commentary is ‘endogenous’: it can influence the same systems and dynamics it seeks to describe or predict. As an obvious example: if a journalist discovers that a major bank robbery is planned for Saturday 9 March 2024 and publishes an expose on Friday the 8th of March, there is a significant probability that the bank robbery will not happen. That depends on how likely the robbers are to find out someone has been tipped off about their plans, how likely the authorities are to act on such information, and whether or not the robbery could be pulled off anyway.

Similarly, when one publishes analysis of behind-the-scenes scheming a few months or weeks in advance of an election there is a possibility it could influence those dynamics in such a way that it renders the predictions false. If, for example, someone were to argue in Parliament that removing a social grant for unemployed voting-age adults immediately before the election amounted to inappropriate electoral influence, that policy decision may well be reversed. Alternatively, if one argues that a certain ‘election app’ is being used to influence voter decisions or voter turnout, the use of that app may change.And so forth.

This is important for at least two reasons. First, assessing the accuracy or inaccuracy of any of the analysis published here after the actual events have taken place (or not). Second, if you have the good fortune of having money to invest, in making investment decisions based on this analysis and the associated predictions. I will be writing at least one article on how investment decisions could be informed by my analysis - and I will be taking some such decisions myself - but that should not be treated as making any kind of recommendations. Social and economic dynamics are complex. And they are influenced by analysis of them.

It is perhaps not a coincidence that one well-known character who understood this problem of ‘reflexivity’ is also the founder of an organisation that is involved in influencing some of the very political dynamics I will be describing.

With these cautions in mind, look out for for the next of my pre-election analyses in the coming week.