My article below was published in the Business Day on Monday, below I repost the article.
More monetary policy guidance please
Over the last several years, market expectations of the South African Reserve Bank (SARB) policy rate has regularly diverged from the SARB’s policy guidance. Had market expectations been more aligned with the SARB’s there would have been a lot less market volatility. SARB can better anchor expectations to its policy path by reducing the uncertainty around how it would react to shocks and data outcomes.
In recent Monetary Policy Committee (MPC) news conferences, SARB MPC members have frequently used the truism ‘decisions are data-dependent’. This phrase has become increasingly commonly used by policymakers at major central banks. But policymakers’ decisions do not just depend on how the economy is evolving. They must be part of a coherent monetary policy strategy. Central bank best practice is to communicate the central bank’s assessment of the drivers of the economic outlook and its monetary policy strategy, providing enough information for the market to evaluate the reasonableness of the MPC’s assessment and its policy decisions.
Around the world, analysts have been asking central bankers the same question: ‘When will you begin to cut interest rates?’ Yet many central banks have been reluctant to answer this question.
There are two reasons for this. The outlook for the economy is always uncertain, and therefore its difficult to know when inflation will fall persistently. The other is that most central banks do not explicitly explain how careful they want be in driving inflation lower. A central bank may, for example, prefer to only adjust interest rates slowly as inflation falls to avoid having to raise rates again or it may wish to cut more quickly if it is worried about the impact that high interest rates might be having on growth.
Best practice among inflation targeting central banks is to be clear about the bank’s inflation target and how the bank is likely to react to specific shocks if they to occur. Private sector analysts may expect the economy to evolve differently to projections provided by the central bank or may expect different policy decisions that what are observed. If a central bank does not communicate policymakers’ policy preferences and the information underlying its forecasts, it is hard for market analysts to distinguish whether a central bank has a different tolerance for inflation than assumed by analysts or whether central banks’ forecasts might reflect confidential information policymakers are privy to. Ambiguity about MPC’s policy preferences or secrecy around central bank projections shield central banks from scrutiny about the quality of their economic analysis and decision-making.
Central banks do differ in how much information they make public. International best practice is to publish forecasts and explain how policymakers expect to have to set policy to achieve their target. While the SARB publishes forecasts and a projection for the policy rate, it does not provide quarterly profiles for many key variables, including GDP and the policy rate. The SARB also provides a short written MPC statement each time it makes a policy decision. By international standards, the SARB allows a lot of analyst interaction with the MPC, with question-and-answer sessions after MPC decisions, Monetary Policy Forum events, speeches by MPC members and analyst conference calls used to explain policy thinking.
In line with other central banks, the SARB has increased its public disclosures over time. An analysis of SARB communication by du Rand et al (2021) finds that communication has improved over the last two decades. But the research notes that there has been an absolute decline in speeches over recent years and that more policy communication through MPC statements has not implied greater clarity of communication, with the readability of its communication being at a level that is typically associated with a graduate-level degree of difficulty.
Another important aspect of best practice among major central banks that use models to communicate forecasts is to publish forecasts that are consistent with the currently decided policy stance, or at least, to publish scenarios that explain any deviations in monetary policy decisions from their published forecasts. The SARB does not provide detailed analytical content explaining its balance of risk assessment or explicit scenario analysis, choosing instead to publish analytical content related to its forecasts in its bi-annual Monetary Policy Statement.
The SARB stands out from best practice in publishing forecasts that are not necessarily consistent with their policy decisions. Central banks where the MPC do not ‘own’ the published forecasts typically label such forecasts as ‘staff forecasts’ to clarify that they may be based on different assumptions and judgements about the outlook for the economy than underlying the MPC decision. While the SARB uses a model to produce its projections, it sometimes publishes forecasts that are not based on the actual decisions made by the MPC. This creates uncertainty around the reaction function of the SARB (i.e. how the Bank will react to specific economic shocks) and makes it difficult for market participants to anticipate how interest rates might change over time. Commentators and analysts also have a hard time interrogating the reasonableness of the SARB projections since the SARB does not communicate the economic narrative underlying monetary policy decisions through the lens of its main forecasting model.
Another problem with providing limited analytical content alongside policy decisions, as the SARB does, is that there is pressure on the MPC to provide information about its views and preferences in forums outside its public disclosures.
This is also why well-known Harvard economist, Greg Mankiw, has called for the US Federal Reserve’s (Fed) news conference to be cancelled. He asserts that the Fed chair ‘seems to be conveying the least possible information in the most words possible’. He argues that policy statements should provide all the required information to understand policy decisions and the central bank’s assessment of the balance of risk around its projections and how it might respond to shocks.
Others argue that too much transparency could reduce central bank credibility if projections turn out to be incorrect. Like other central banks, the SARB’s post-pandemic projections for inflation proved inaccurate as the SARB initially assumed that the pandemic and economic lockdowns would have predominantly demand-side impacts on the economy. This meant that the recent spike in global inflation caught SARB by surprise, as they have been slow to revise their judgements about the underlying trends in the economy, for a long time holding on to a view that inflation pressures would be transitory despite growing evidence to the contrary. SARB, like other central banks, eventually attributed more of the 2020Q2 fall in GDP to supply shocks. This gradual change in interpretation of economic developments amounted to an acknowledgment that the inflation impulse being observed was not just a temporary phenomenon and that risks to trend inflation were intensifying.
Central banks should not be vague when they are uncertain. Central banks should communicate what they are uncertain about and how they make decisions in the context of uncertainty. Transparency about the assumptions underlying projections help the market assess the reasonableness of these assumptions and the implications for central bank policy. Leading central banks even publish the code to their models so that the public can learn how central banks might react if certain shocks occur and assess whether policymakers are learning from their mistakes.
Greater analytical detail and more clarity around the conditionalities associated with SARB’s projections would reduce the risk that the SARB surprises the market. This would reduce market uncertainty and unnecessary volatility in interest rates and other asset prices. It is time for SARB to provide more guidance about its monetary policy decisions.
Dr Daan Steenkamp
Daan is CEO of Codera Analytics and a research fellow with the Economics Department at Stellenbosch University.
