As we have long argued in this blog, South African policymakers should monitor inflation risk alongside measures of the inflation outlook. This is because inflation risk affects the extent to which inflation harms welfare. Inflation dispersion bears on the optimal inflation target and assessment of the appropriate stance of policy at any particular point in time.
In today’s post, we compare three different inflation risk measures for South Africa. Codera’s measure of inflation dispersion draws on hundreds of goods and services categories to summarise inflation pressure. As we showed in our recent policy brief, our inflation pressure measure describes divergences in inflation from the inflation target, with inflation tending to rise when the inflation distribution widens. To measure inflation dispersion, we calculate a summary measure as the difference between the mean and the median of the inflation distribution. When the mean exceeds the median of the inflation distribution, it is positively (i.e. rightwardly) skewed, with inflation pressure picking up.
We also plot Ha, Mumtaz and Ruch (2026)‘s measure of the probability that inflation exceeds 5 percent over a twelve-month horizon to capture the right-tail risk of the inflation distribution. Their estimate has a high average value over the sample, suggesting South Africa has relatively high inflation risk by global standards.
Lastly, our measure from Christensen and Steenkamp (2026) is a bond market-based long-term (5yr5yr) inflation risk measure estimated using the term structure of bond prices. Although the Ha et al. measure comoves with our long-term, market-based inflation risk measure during several periods of above target inflation, their correlation is low and they diverge after the post-pandemic cost of living crisis. Market-based inflation risk fell dramatically ahead of the the adoption of a lower inflation target, even if market-based inflation expectations did not.
At present, these measures suggest inflation risk is still low by historical standards. We continue to monitor these measures for evidence of broadening inflation pass-through and inflation pressures.
