Figure 1: SA Consumer Inflation: history and  forecasts
  • The SARB surprised market expectations for the second consecutive MPC meeting, this time, by leaving the repo rate unchanged at 6.75%, versus consensus of a 25bp reduction.  The MPC vote was equally split.
  • Underpinning the decision was the “deteriorating assessment of the balance of the risks”, according to the MPC statement.
  • In particular, the SARB revised its assessment of the risks to its inflation forecasts to “be somewhat on the upside” instead of being “broadly balanced”.
  • An upside risk pertains to an electricity tariff increase in July 2018 that exceeds the SARB’s assumption of 8%. NERSA’s determination of Eskom’s application for a 20% tariff increase is scheduled in December 2017.
  • Although inflation expectations are anchored below the upper end of the target, the MPC “remains concerned that inflation expectations of business people and trade unions remains above or close to 6% for the next two years”. This is a potential source of upside risk as these are “key price setters”.
  • The rand remains a key upside risk on continued vulnerability to political developments, sovereign credit rating downgrades and a shift in global sentiment towards emerging markets. The rand also remains vulnerable to a faster than expected monetary policy normalisation in the major advanced economies however with inflation still broadly benign, markets remain positioned for only a gradual pace of policy normalisation. In the event, portfolio flows to higher-yielding emerging markets could remain resilient. 
  • The SARB marginally lifted its CPI inflation forecast in the outer years of the forecast horizon when taking account of “a lower repurchase rate, a less appreciated exchange rate assumption, a slightly narrower output gap and a marginal adjustment to the food price forecast” (see figure 2). 
  • Although the SARB remains data dependent it would appear that the possibility for the next interest rate cut to occur at the November MPC has diminished given the SARB’s concerns outlined in the September MPC statement. Specifically, the SARB noted that some of the event risks, in relation to the rand, “particularly those of a political nature, are now more imminent but with no greater degree of clarity regarding the outcome”. 
  • Event risk is expected to rise into year-end in view of the MTBPS in October, the rating agency reviews at the end of November and the ANC elective conference in December. Rand volatility could increase over this period, rendering it more challenging to ease interest rates at the November MPC.
  • As such, we could see a resumption in interest rate cuts in Q1.18. Moreover, we project CPI inflation to moderate to 4.2% y/y in Q1.18, comfortably in the vicinity of the mid-point of the target range. 
  • Overall, we expect a shallow easing cycle as SA needs to maintain a positive real interest rate differential versus developed economies, if substantial and sustained rand depreciation is to be averted.
SARB CPI inflation forecasts (% y/y)
Figure 3: Foreign sector assumptions
Foreign Sector Assumptions:
  1. Trading partner GDP growth: is determined broadly via the IMF’s Global Projection Model “GPM” which is then adjusted to aggregate the GDP growth rates of South Africa’s major trading partners on a trade weighted basis. Individual projections are done for the four largest trading partners (euro area, China, United States and Japan), while the remaining trading partners are grouped into three regions: Emerging Asia (excluding China), Latin America and the Rest of Countries bloc. The assumption takes account of country specific “consensus” forecasts as well as IMF regional growth prospects.
  2. Commodity price index: is a weighted aggregate price index of the major South African export commodities based on 2010 prices. The composite index represents the total of the individual commodity prices multiplied by their smoothed export weights. Commodity price prospects generally remain commensurate with global liquidity as well as commodity demand/supply pressures as reflected by the pace of growth in the trading partner countries.
  3. Brent crude oil price: is expressed in US dollars per barrel. The assumption incorporates the analysis of factors of supply, demand (using global growth expectations) and inventories of oil (of all grades), as well as the expectations of the US Energy Information Administration (EIA), OPEC and Reuters.
  4. World food prices: is the composite food price index of the Food and Agriculture Organization of the United Nations (FAO) in US dollars. It is weighted via average export shares, and represents the monthly change in the international prices of a basket of five food commodity price indices (cereals, vegetable oil, dairy, meat and sugar). World food price prospects incorporate selected global institution forecasts for food prices as well as imbalances from the anticipated trend in international food supplies relative to expected food demand pressures.
  5. International wholesale prices: is a weighted aggregate of the producer price indices of South Africa’s major trading partners, as per the SARB’s official real effective exchange rate calculation. Although individual country CPI forecasts provide a good indication for international wholesale price pressures, the key drivers for the assumed trend in global wholesale inflation are oil and food prices, as well as expected demand pressures emanating from the trends in the output gaps of the major trading partner countries. Other institutional forecasts for international wholesale prices are also considered.
  6. Real effective exchange rate: is the nominal effective exchange rate of the rand deflated by the producer price differential between South Africa and an aggregate of its trading partner counties (as reflected in the SARB Quarterly Bulletin). Although the nominal rate is a weighted average of South Africa’s 20 largest trading partners, particular focus is placed on the rand outlook against the US dollar, euro, Chinese yuan, UK pound and the Japanese yen. The assumed trend in the real effective exchange rate remains constant from the latest available quarterly average over the projection period. However, due to the time delay for the calculation of the real effective exchange rate, the most recent trend in the nominal effective exchange rate is adjusted with the assumed trend for the domestic and foreign price differential for the current quarter. This may result in a technical annual adjustment over the current and next forecast year that differs from zero.
Figure 4: Domestic sector assumptions
Domestic Sector Assumptions
  1. Government consumption expenditure (real): is broadly based on the most recent National Treasury budget projections. However, since these projections take place twice yearly, the most recent actual data points also play a significant role in the assumptions process.
  2. Administered prices: represents the total of the regulated and non-regulated administered prices as reflected by Statistics South Africa (StatsSA). Its weight in the CPI basket is 18.48 per cent and the assumed trend over the forecast period is largely determined by the expected pace of growth in petrol prices, electricity tariffs, school fees, water and other municipal assessment rates. Petrol price: is an administered price and comprises 5.68 per cent of the CPI basket. The basic fuel price (which currently accounts for roughly half the petrol price), is determined by the exchange rate and the price of petrol quoted in US dollars at refined petroleum centres in the Mediterranean area, the Arab Gulf and Singapore. The remainder of the petrol price is made up of wholesale and retail margins as well as the fuel levy and contributions to the road accident fund (RAF). Since most taxes and retail margins are changed once a year, the assumed trajectory of the petrol price largely reflects the anticipated trend in oil prices and the exchange rate. Electricity price: is an administered price measured at the municipal level with a weight of 4.13 per cent in the headline CPI basket. Electricity price adjustments generally take place in the months of July and August of each year, and the assumed pace of increase over the forecast period reflects the multi-year price determination (MYPD) agreement between ESKOM and NERSA with a slight adjustment for measurement at the municipal level.
  3. Potential growth: the pace of potential growth is derived from the SARB’s semi-structural potential output model. The measurement accounts for the impact of the financial cycle on real economic activity and introduces economic structure via the relationship between potential output and capacity utilisation in the manufacturing sector (South African Reserve Bank, Working Paper Series, WP/14/08).
  4. Repurchase rate: The “repo” rate is the official monetary policy instrument and represents the interest rate at which banks borrow money from the SARB. Although the rate is held constant over the forecast period, this assumption is relaxed in alternative scenarios where for instance the policy rate responds to deviations of output from its potential and the gap between future inflation and the inflation target, i.e. via a stylised “Taylor rule”; one that is based on market expectations of the future path of the policy rate; and other paths as requested.