20 November 2009 Review the target By ANNABEL BISHOP SA has missed its inflation target for more months than it has achieved it, indicating that a too-narrow band was initially imposed for a country that has a history of double-digit inflation.

The targeted measure of inflation, whether CPIX or CPI, has averaged 6,8% year on year - well above the 6% upper limit of the target band. Though it is not optimal to change the inflation target range when inflation is above the target and has been for well over two years, the narrowness of SA's range and the high level of misses do warrant some review. In the absence of changing the target measure from plain CPI to CPI excluding energy costs, the inflation-targeting band should be temporarily (and only temporarily) widened to 2%-2,5% around the midpoint, instead of the current 1,5%, with the midpoint itself at 5% or even 5,5%. Indeed, the IMF points out that keeping inflation within a fairly narrow range has proven harder, or perhaps less sensible, than was believed in the early days of inflation targeting. Even a target range of 7%-2%, or 2,5% around the current midpoint of 4,5%, would mean SA achieved the inflation target more often than it missed it, making this a more appropriate band to attempt to first achieve stable inflation (inflation that generally falls within the target band), while not ignoring the growth imperatives of the still economically marginalised majority of South Africans. The benefit of achieving well-anchored inflation expectations (within the target range) is that salary and wage increases will be unlikely to deviate too significantly from the range, meaning that even when inflation misses the target it will be drawn back into target without too substantial a change to monetary policy. This will allow a lower long-term path for interest rates. Ideally, the target band should be narrowed and/or reduced over a few years, eventually bringing it down to the current 6%-3%. A good introductory midpoint would have been 5,5%, with 7,5%-3,5% for the target range, given that 3% is close to the average inflation rates of SA's key trading partners - the ultimate aim for the midpoint of SA's inflation target. Once inflation is stable in a higher band, the midpoint and/or band can be successively reduced every two to three years, depending on domestic (administered prices) and global inflationary forces, as stable inflation is repeatedly achieved. This would make inflation targeting a more flexible process rather than a rule that has proved for SA to be difficult to achieve consistently. Since the late 1990s (the average time of the adoption of inflation targeting), evidence from 25 inflation-targeting countries is that it appears not to have improved a country's economic performance, due to a fairly benign global environment. If inflation targeting does not promote stronger growth in the long term, nor necessarily lower inflation, is even the short-term trade-off justified? Too strict a focus on inflation causes growth to suffer, due to high interest rates, and too flexible an approach undermines the credibility of the whole exercise. Given that wage settlements are yet to fall to the midpoint of 4,5%, or even 3%-6%, SA's inflation targeting has arguably been too flexible, even lacking in credibility. The time is fast approaching when we will be back in the target. The costs of not reviewing the target may outweigh the benefits to growth and employment and it should be re-looked at to determine if it indeed serves the best interests of the majority of SA's citizens. The dual aim of greater employment (by means of significantly higher real economic growth rates), together with low inflation may be more appropriate for SA. The latter would also relieve the social welfare burden placed on the state, which is set only to escalate in the future. Bishop is an economist at Investec