Rosenberg: The Riksbank's forecasts and current monetary policy

Deputy Governor Irma Rosenberg gave a speech on Thursday at a meeting organised by Danske Bank. One subject discussed was the low inflation rate in recent years and the Riksbank's forecasts, and another subject was current monetary policy. ”Over the past two years, inflation has been significantly lower than the Riksbank’s inflation target of 2 per cent. This has led to a discussion of the quality of our forecasts. We have also been criticised for conducting what was claimed to be an overly contractionary monetary policy. To try and further nuance this debate, I shall begin with a look back at the forecasts we made for 2004 and 2005 and how they compare with other analysts’ forecasts and with the actual outcomes. I will also give my view of how it might have affected monetary policy in recent years if our predictions of the path of inflation had been more accurate. In conclusion, I shall look ahead and say a few words about the current monetary policy situation,” began Ms Rosenberg. ”Let me begin by noting that inflation has on average been very close to the target during the period we have had an inflation-targeting regime and until a year or so ago. In 1995-2004 inflation according to the UND1X measure of underlying inflation was on average 1.9 per cent (Table 1). Given that the Riksbank cut the repo rate by almost 7 percentage points during this period, I consider it reasonable to regard target fulfilment from the perspective of the UND1X measure, as this does not include the direct effects of these interest rate cuts, i.e. the fact that mortgage interest expenditure fell,” said Ms Rosenberg. ”However, during 2004 and 2005 inflation was significantly below the target. Inflation fell substantially at the beginning of 2004. One explanation was that energy prices, which had been pushed up sharply at the beginning of 2003, had fallen again slightly. We had already foreseen this and therefore assumed in the first Inflation Report published in 2003 that inflation would be below target in 2004. Since the fluctuations in energy prices were assumed to be temporary, we did not consider that the rise in inflation in early 2003 or the predicted decline in 2004 gave cause for any monetary policy measures. We also said this clearly in connection with the publication of the monetary policy decision. Thus, the forecasts the Riksbank made with regard to UND1X inflation in 2004 were already at the beginning of 2003 in line with the final outcome (Figure 1),” continued Ms Rosenberg. ”The forecasting errors regarding inflation in 2004 were thus relatively small, but domestic cost pressures were much lower than expected. Important reasons for this were the fact that productivity developed more strongly than we had assumed, while imported inflation was substantially lower than predicted. These are factors that have been shown to hold back inflation in a more lasting way than expected. The forecasts we made in 2003, in particular, but also those made in 2004 for UND1X inflation in 2005 therefore entailed a clear overestimate. The assessment was that growth would increase and that inflationary pressures would at the same time rise, in accordance with traditional cyclical patterns. However, while growth strengthened much more substantially than we had anticipated, inflation remained low and our view of both international and domestic cost pressures had to be gradually revised. It was not until the beginning of 2005 that we made a relatively large downward revision in the inflation forecast, but on the other hand we then went too far, which can be seen in Figure 1,” said Ms Rosenberg. ”But why was inflation so low in 2005? As I have mentioned already, inflation has been kept down by high productivity growth and weak growth in imports. This appears to be connected with changes of a structural nature. One such change is unquestionably increased global competition. It is also probable that part of the explanation can be found in the fairly comprehensive investments in new technology in recent years. The low inflation is therefore in many ways a consequence of a number of changes in the supply side of the economy. A natural question following on from this is whether the Riksbank should have been able to predict this development and thereby the low inflation rate. Of course, it is not easy to answer this type of question afterwards, but there is the possibility of comparing our forecasts with those of other forecasters to see how we rate,” Ms Rosenberg pointed out. ”The Riksbank publishes information to enable an assessment of the monetary policy conducted over the past 2-3 years in a special appendix to the first Inflation Report of every year. It contains, for instance, comparisons of our forecasts with other forecasts and with outcomes. The information provided in the February Inflation Report shows that most forecasters making predictions in 2004 substantially overestimated CPI inflation for 2005 and that our forecast compared favourable with the others (Figure 2). With regard to GDP growth, all forecasters underestimated the strength of developments in 2004 and other analysts therefore did not differ significantly from the Riksbank (Figure 3). This probably reflects the fact that changes of a structural nature are difficult to capture at an early stage of the forecast work,” said Ms Rosenberg. ”A reasonable interpretation is that the Riksbank’s forecasts have been no worse than those of any other analysts and in some respects they have been better. The Riksdag (Swedish Parliament) Committee on Finance draws similar conclusions in its assessment of monetary policy published recently. From the end of 2002, when the most recent interest rate cut period began, and up to summer 2005, the repo rate was cut by a total of 2.75 percentage points to 1.5 per cent. This entailed a substantial stimulation of the economy. Of course, with the benefit of hindsight one can say that it might have been possible to make the interest rate cuts slightly more quickly. However, one can question whether it would have been appropriate to cut the rate further, given that demand was developing strongly back in 2004. The rapid increases in household sector indebtedness and in house prices are evidence that monetary policy was very expansionary. To further push up demand when the economy was already showing strong growth could have entailed risks for the stability of the real economy further ahead,” said Ms Rosenberg. ”I would now like to move on to current inflation prospects and monetary policy. The development we envisaged in the February Inflation Report was that international growth would continue to be good. The new data received since then indicates that the world economy has possibly developed even more strongly than anticipated. New national accounts data point to high growth in the United States during the first quarter of this year. The indicators received during the spring have given slightly mixed signals, but on the whole imply continued good development. In the euro area, developments during the fourth quarter were roughly in line with expectations, but the indicators received since then point to more slightly favourable developments this year than we envisaged in February. In Asia the picture of a stronger development is even more evident. There has been a clear upswing in Japan, while growth in China and the rest of Asia has remained high,” said Ms Rosenberg. ”In February, we saw ahead of us continued strong growth in the Swedish economy. According to the preliminary national accounts just published, GDP growth was slightly higher than expected in the first quarter. However, household consumption was weaker than we had assumed, which is partly balanced by slightly stronger investment. Other information over the past month also indicates that there has been continued good development. According to the National Institute of Economic Research’s business tendency survey for May, firms in both manufacturing and construction are reporting increased orders and larger production volumes. Moreover, construction firms are beginning to experience major problems in finding suitable labour. The retail trade is showing strong sales and all parts of the private services sector report increased demand. In addition, households are optimistic with regard to both their own finances and the Swedish economy,” said Ms Rosenberg. “There has been turbulence in the financial markets recently, with a relatively large fall in share prices. Changes in asset prices can reflect or lead to increased uncertainty regarding future developments, which can subdue companies’ willingness to invest and lead to households increasing their precautionary saving. However, share prices can rise sharply over a long period of time and then recoil, without this being a sign of a more lasting decline. At present, it is not possible to draw any far-reaching conclusions from the recent stock market developments,” noted Ms Rosenberg. ”With regard to the situation in the labour market, there were numerous indications in February that employment would continue to increase. New information received since then regarding, for instance, new vacancies and redundancies points in the same direction. According to the National Institute of Economic Research’s business tendency survey, the upswing in the number of employed will continue during the second quarter. All in all, this would rather indicate that the Swedish economy has developed slightly more strongly than we assessed in the February Inflation Report,” pointed out Ms Rosenberg. “With regard to inflation, the assessment in the Inflation Report was that inflation would rise and approach 2 per cent a couple of years ahead, on the assumption that gradual increases were made in the repo rate during the forecast period. Inflation has recently been higher than we estimated in the Inflation Report, mainly due to unexpected price increases on a limited number of goods, such as petrol, heating oil and electricity. There is reason to assume that oil and petrol prices will remain high for some time to come, which is reflected in the fact that forward prices have risen. However, on the assumption that energy prices do not continue to rise, it appears that the effects on inflation will be temporary,” considered Ms Rosenberg. ”GDP growth thus looks to be slightly stronger than we assumed in February, but productivity growth remains good and cost pressures are moderate. The basic assessment of inflation developments in coming years thus remained largely unchanged, even taking into account the new information received since the Inflation Report. The assessment is that inflation will be close to two per cent a couple of years ahead. At the same time, monetary policy is still very expansionary, which is expressed for instance in the continued rapid growth in credit and the rapid increase in house prices – although we can now possibly see, for the first time in a long period of time, some signs that the rate of increase in lending is slowing down. All in all, this indicates, in my opinion, that now may be the time to take a further step towards less expansionary monetary policy to ensure that inflation remains on target and that we have balanced growth in the real economy. This also agrees well with the prevailing market expectations,” concluded Ms Rosenberg.