2020 Mortality Rate Puzzle

I am reposting this this morning because, on reflection, it is a significant graph. I had opportunity to drop more statistics into the graph of South Africa's 2020 mortality rate, now with the forecast mortality rate included. All 2020 mortality figures are from the South African Medical Research Council (SAMRC), which see. I assume that facts from the CIA Fact Book and macrotrends are more or less correct. The forecast mortality rate was 486,262 or 8.15. That looks like 60,000 deaths too few. It would thus be reported as 60,000 deaths too many. The actual mortality rate was 542,715 or 9.10, which is as if COVID-19 never happened. Stats SA put the population at 59,620,000. You may click on the graph to enlarge.

POSTSCRIPT: I'm offering this graph as a question rather than a statement. If any of the figures are incorrect (I don't think they are), do let me know. In feedback I received so far, the suggestion was that COVID-19 may have been (hugely) offset by pre-emptive measures, such as the alcohol ban. I did a quick, independent calculation of US statistics, as a cross check. There my own calculations were close to published figures. Only one thing didn't make sense there. I estimated that (very roughly) 50% of Americans had had COVID-19. Published figures are generally around 10%. In March 2020, my co-editor and I may have been the first, in an editorial, to (almost) correctly estimate the infection fatality rate of COVID-19, at 0.3%.