Just how low can inflation go in an inefficient and highly regulated economy?
The government’s recent move to cap fuel price increases comes at an interesting moment in SA’s inflation fortunes. The government has cut the fuel levy temporarily and opened the way for deregulated fuel pricing for the first time. It’s a move that highlights how much of SA’s inflation basket reflects policy and regulatory choices.
And it comes as debate about whether SA should lower its inflation target could be about to reopen. The Treasury has completed a long-promised macroeconomic policy review and is preparing to release this. An independent study of monetary policy, one of the inputs to the review, has recommended that SA cut its inflation target from the 3%-6% range towards a point target of 3%.
This is a cause Reserve Bank governor Lesetja Kganyago began to espouse publicly in 2021. He called for the target to be cut to levels closer to those of SA’s emerging market peers to make us more competitive. During his tenure he has in effect reset the target to the 4.5% midpoint of the range. He has managed to get inflation and inflation expectations down to this level without too much economic pain, thanks to low global and food and fuel price inflation. That seemed to be a platform on which to push towards an even lower target — in line with the government’s original objective when it introduced inflation targeting in 2000.
But it now feels as if Kganyago might have missed the boat. War in Ukraine and post-pandemic supply chain disruptions have sent inflation soaring globally. SA’s inflation rate is heading up too, and will soon breach the top of the target range. Amazingly, even at 5.9% it’s now lower than US inflation (7.9%) for the first time anyone can remember. But pushing SA down towards 3% would be a stretch, one that would surely come at high cost to an economy just recovering from Covid-19 in the short term, even if lower inflation would be good for the economy in the long term.
The study for the Treasury by international experts Patrick Honohan and Athanasios Orphanides is puzzlingly silent on the challenges involved in reducing the target, if that is indeed what SA should be doing.
The academics urge a more “ambitious” approach that would take the target down to first 4% then 3%, which they say would provide a stable platform for monetary policy. They don’t provide any modelling on why 3% — or how high interest rates might have to go, at what cost to economic growth, to get it there. And they baldly state that: “A decision to opt for 3% as the point target would need to be accompanied by a commitment from the fiscal authorities to play their part and make budgetary policy and policy on administered prices consistent with the chosen inflation rate.”
If it were that easy. Inflation in administered prices — the water, electricity, fuel, transport, rates and other charges set or regulated by the government or state-owned enterprises — has run well ahead of the average inflation rate for a long time. Some speculated that one of Kganyago’s main aspirations in lowering the target was to put the pressure on to curb administered price rises. But doing so would require policy shifts, which in many cases would be contested. It would also require a much more capable and efficient state.
And if administered prices could not be curbed, harsher action would be needed to force down other prices in the economy. Nedbank corporate and investment bank analyst Walter de Wet modelled this in a report in 2021, which concluded that while a 4% target could be feasible — and in line with the average 3.9% for emerging market economies — 3% would potentially require more than two percentage points of interest rate hikes that the Reserve Bank would otherwise not have had to implement.
Admin prices are a big part of that story, as De Wet tells it. Excluding administered prices, the inflation rate since 2017 has been just 4%. But inflation in admin prices has been 6%. And given infrastructure deficits and high state-owned enterprise debt levels, it’s not that easy to get it down — hence stronger action would be needed to cut inflation in the rest of the basket. De Wet calculated that if SA were to target a 3% inflation rate the effective target would be just 2.4% for the basket excluding admin prices.
The government’s moves on fuel at least open the way to review other admin prices, and ironically the extent to which the fuel price reflects taxes and regulated margins rather than global oil or the rand is helping cushion SA’s inflation rate from the full extent of $120 per barrel oil prices.
But the problem is not just admin prices strictly defined, but also those inflated by tariffs that keep SA’s inflation rate high. A study by Trade & Industrial Policy Strategies researcher Neva Makgetla for the Reserve Bank in 2021 found that the tariffs on food had increased steadily since 2013 and were now one percentage point higher than tariffs on all other goods. The higher tariffs are mainly targeted at staple foods for low income groups, ranging from 50% tariffs on poultry and sugar to 10% on cooking oil.
High staple foods inflation has a multiplier effect on the inflation rate more generally because these foods make up a large part of the inflation basket for low-income workers and so put upward pressure on wages, and therefore cause broader price pressures in a way that tariffs on luxury goods would not, Makgetla’s study found.
All of which goes to suggest that SA’s inflation problem in the longer term is as much about trade, industry & competition minister Ebrahim Patel or mineral resources & energy minister Gwede Mantashe or public enterprises minister Pravin Gordhan, as it is about global oil or wheat prices in the shorter term. The Reserve Bank has managed to cut SA’s double-digit inflation rates to the mid single digits over the past couple of decades. Taking it down further will depend on the government.
• Joffe is editor-at-large.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.