Less growth for no less inflation: a trade off that could have been avoided with lower, not higher, interest rates
Since the Monetary Policy Committee (MPC) of the Reserve Bank decided to raise its repo rate by 0.25 percentage points on Thursday, the outlook for SA growth has deteriorated, because of the likely impact of higher interest rates on spending; while the outlook for inflation has deteriorated, because the rand has weakened. Less growth for more inflation hardly seems like a useful tradeoff, but that is what the SA economy has to confront.
Can we however blame the Reserve Bank for the weaker rand? We can, if the prospect of slower growth is expected to reduce the case for investing in SA and therefore is associated with the weaker rand. But the rand may have weakened for other reasons unrelated to the Reserve Bank decision. Emerging market risks may have simultaneously increased, thus discouraging capital inflows, or something the MPC also worries about – interest rates in the US may increase, so driving capital away from emerging markets leading to a weaker rand.
Neither of these forces since Thursday last week can explain the fact that the rand lost a little ground to other emerging market currencies, while interest rates in the US fell rather than rose. Moreover, while long term rates in SA remained little changed, the spread between RSA and US rates widened since the interest rate increase, indicating an increase in the SA risk premium (a wider spread is usually associated with rand weakness).
Furthermore short term rates – up to one year duration – all rose in tandem after the MPC meeting, indicating that the outlook for interest rates to come has not changed in response to Reserve Bank action or explanation. The interest rate carry in favour of the rand, AKA the SA risk premium, remains as it was. The forward looking stance of monetary policy, in the collective view of the money market, has therefore not softened – a softening that might have served to explain the weaker rand, but does not.
All of this indicates another point we have made repeatedly. The impact of any move in policy-determined SA interest rates on the exchange value of the rand is essentially unpredictable. This makes the relationship between interest rate changes and inflation also highly unpredictable, so undermining the logic of inflation targets. Inflation targets, if they are to be met with interest rate settings, demand a predictable relationship between interest rate movements and inflation itself. This predictability does not exist.
The unpredictable reactions in the currency markets help vitiate the presumption that interest rates can be a useful instrument for realising inflation targets. Upredictable increases in administered prices, the price of electricity, water, municipal services etc. that may drive inflation temporarily higher (as might a weaker maize harvest) are supply side forces that do not respond to higher interest rates. The notion that interest rates should rise in response to an economically damaging drought is surely risible. Higher interest rates in SA do have one highly predictable effect and that is to reduce spending.
The latest surprise for the inflation forecasting model of the Reserve Bank from which interest rate settings take their cue, is that the so called pass through effect from a weaker rand to higher prices is about half as strong as predicted by the model. Both the rand prices of imports (helpfully) and exports (unhelpfully for the domestic economy) are lower than they were a year ago, reducing rather than adding to the pressure on the CPI. This time round it is not the weaker rand that can be blamed for higher inflation to date- but a still weaker rand clearly imposes the risk of more inflation to come.
The MPC, by increasing short term interest rates, willingly added to the risks of still lower growth rates. Our view is that this represents a distinct error of judgment.
To quote the MPC statement:
“The MPC has indicated for some time that it is in a hiking cycle in response to rising inflation risks, and a normalisation of the policy rate over time. The MPC is cognisant of the fact that domestic inflation is not driven by demand factors, and the outlook for household consumption expenditure remains subdued. Economic growth remains subdued, constrained by electricity supply disruptions and low business and consumer confidence and the risks to the outlook remain on the downside. However, as emphasised previously, we have to be mindful of the risk of second-round effects on inflation, and the committee is concerned that failure to act against these heightened pressures and risks will cause inflation expectations to become entrenched at higher levels.”
We would take issue with the relevance of the so defined second round effects that is so important to the Reserve Bank view of how inflation comes about. That is the notion that more inflation expected can lead to more inflation as a kind of self fulfilling process and that it takes, if necessary, painfully higher interest rates to control such expectations. The reality, in our view, is that these inflation expectations held in SA are particularly well entrenched and highly stable at around the 6% level, the upper end of the inflation target band. That is, if we infer inflation expected from the actions of investors in the bond market, being the difference in yields offered by vanilla bonds and their inflation protected alternatives of similar duration. These differences are shown below for 10 year bond yields.
Thus, the remarkable fact about the extra rewards for taking on inflation risk – the difference between a coupon exposed to unexpected inflation and one completely protected against actual inflation – I is how stable it has been, around about 6%, the period of the global financial crisis in 2008 excluded. The daily average spread since 2009 has been 5.95%, with a maximum yield spread of 6.92% and a minimum of 4.55%, with a Standard Deviation of 0.41%. It would seem to us that the inflation leads inflation expected – not the other way round- and that it would take an extended period of inflation well below 6% p.a to reduce inflation expected. These second round effects are a theory without empirical support that is preventing monetary policy from acting in a usefully counter cyclical way. A cycle that calls for lower not higher interest rates to encourage not discourage growth that attracts capital and might support not weaken the rand.
Furthermore, the MPC should recognise that price setters, that is most firms, set prices according to what the market will bear, that prices are not simply cost or wages plus sum pre-determined profit margin. Higher costs will lead to lower margins if demands from the market restrict pricing power, and higher wages can lead to fewer people employed in the presence of weak demand and the absence of pricing power.
In the distinct presence of weak demand, fully recognised by the MPC, neither expected inflation nor higher wages explain higher inflation in SA. Nor does money or credit growth help explain why inflation in SA is currently as high as it is. Households are borrowing very little more than they did a year ago and firms are borrowing more, but to invest abroad not locally. Money supply growth remains subdued.
Higher taxes, in the form of higher administered prices, explain much of inflation to date and help explain much of the inflation forecast by the Reserve Bank Model. Administered prices, petrol and electricity for example, are assumed to increase by 11.7% and 12.5% respectively in 2016. Yet these price increases add further to the pressure on household and business budgets and further inhibit spending. Yet the MPC seems convinced their monetary policy settings remain supportive of the economy rather than a threat to it. To quote the PMC statement further:
“The expected inflation trajectory implies that the real repurchase rate remains low and possibly still slightly negative at times, and below its longer term average. The monetary policy stance therefore remains supportive of the domestic economy. The continuing challenge is for monetary policy to achieve a fine balance between achieving our core mandate of price stability and not undermining short term growth unduly. Monetary policy actions will continue to be sensitive, to the extent possible, to the fragile state of the economy. As before, any future moves will therefore be highly data dependent.”
We must beg to differ about the measured stance of monetary policy. A prime rate of 9.25% is well ahead of the price increases most private businesses are able to charge their customers. They do not have the monopoly powers of an Eskom or a municipality to charge more regardless of the state of demand. Keeping prices rising in line with headline inflation (not of their making) is becoming much more difficult, so making monetary policy ever less supportive of the domestic economy.
The economy is fragile and higher interest rates have made it still more so. Had the Reserve Bank been more sensitive to the state of the economy and more data dependent (and not embarked on a premature path to higher interest rates) the economy would have better prospects and a stronger not weaker rand might well have reflected this.