Enquiry into underlying inflation
The Riksbank has contacted the government appointed commission on CPI in order to receive help in determining an indicator of underlying inflation that has the characteristics to win broad acceptance and can be published by Statistics Sweden on a regular basis. Such an indicator could in the future play a more direct role in monetary policy, either by superseding CPI as a target variable or by being used as an operative target in the regular formation of policy. When choosing a target variable it is also important to consider what inflation measures that will be used within the European Union.
"It is important to point out that any future change of the target variable do not change monetary policy's basic goal of maintaining price stability," emphasise Lars Heikensten, the deputy governor of the Riksbank, during a speech at "Arbetarrörelsens ekonomiska råd" earlier today.
Sveriges Riksbank has also requested that Statistics Sweden, the Swedish central statistical bureau, should begin to publish one or more indicators of underlying inflation. It is intended that these measures should be published monthly at the same time as the consumer price index CPI.
Temporary effects on CPI are normally seen as being of subordinate importance for monetary policy and should fall within Riksbank's tolerance interval of 1-3 per cent. As long as deviations in the inflation rate derive from supply side changes or other price changes that cause small, temporary fluctuations in the rate of inflation around a level that is stable in the long run, there is no reason to try to counteract them with monetary policy. On the contrary, this could destabilise real economic development. The central question is how the long run, or trend inflation rate develops. It is this inflation rate which more directly affects aggregate demand in the economy and in turn monetary policy.
It is difficult to estimate the core inflation rate in the economy. Nor is it clear what developments have a temporary effect on inflation and thus should be excluded. This can even vary over time. Moreover, it is difficult to estimate how great an effect temporary shocks have on inflation. Some price shocks can be casused by supply changes of a permanent nature and affect the inflation expectations of households and firms. These are important for the formation of monetary policy.
During 1998, inflation has been lower than the Riksbank forecasted in the last Inflation Report. An analysis of inflation, divided into three central components, (domestic inflation, imported inflation and interest costs for homeowners) suggests that domestic inflation has thus far developed in accordance with the Riksbank's expectations. The forecast error can be explained instead by the fact that the imported inflation developed more slowly than expected. Unexpectedly low oil prises are responsible for slightly more than half of the forecast error. The remaining deviation was caused primarily by the fact that the long interest rates were lower than expected. Thus the household sector has had lower interest costs than expected.
Changes in indirect taxes, subsidies and interest costs are usually considered as having a temporary effect on inflation. In Sweden and in many other countries, changes in prices due to changes in oil prices are considered to be temporary.
Since the Riksbank’s Inflation Report was published in March the Government has presented its spring budget. The budget included a proposal to change several indirect taxes and subsidies, amounting to approximately –0.5 per cent of inflation during 1998 and –0.1 per cent during 1999 compared to what was previously known. If today’s interest levels and low oil prices remain during the rest of 1998, inflation will be dampened even more.
All together this means that temporary effects partly explain the divergence between inflation figures and the forecasts of inflation, and partly the probability that inflation during 1998 will be considerable lower than expected in the previous Inflation Report and might reach levels below the Riksbank's tolerance level of 1 per cent.
All known inflation measures have defects. This means that advantages and disadvantages must be weighed against each other when choosing a target indicator for monetary policy. An advantage of CPI is that it is well known and an established inflation measure amongst economists and the public and is used in many different types of agreements. This was one of the central arguments for the choice of CPI as a target indicator in the beginning of 1993.
CPI has the disadvantage that it contains information about both the core inflation developments and the temporary price changes. Since the inflation target regime has been in effect, different temporary changes have on several occasions affected CPI noticeably. This has been the case for e.g. repeated changes in the level of indirect taxes, swings in interest rates that affect the mortgage costs in the CPI and other types of supply shocks. This has meant an information problem for the Riksbank.
It has been necessary to discuss inflation developments and explain monetary policy on the basis not only of developments in the CPI but even in the core inflation rate. Since December 1997, the Riksbank has also included a forecast path for a measure of the core inflation rate in the Inflation Report.