Separate minutes of the Executive Board meeting in 19 October
Lars Heikensten, Chairman
§ 1. The current inflation assessment
It was noted that Victoria Ericsson and Annika Svensson would prepare draft minutes of paragraphs 1 and 2 on the agenda for the meeting.
The discussion was based on the new data and analyses presented by the Monetary Policy Department. These analyses are based on the assumption that the repo rate will develop in line with the financial markets' expectations, as reflected in implied forward rates. This is a technical assumption for the purposes of calculation and should not be interpreted as reflecting the interest rate path the Riksbank considers to be most desirable.
1. Recent data regarding economic developments
The Monetary Policy Department reported on new data received since the Executive Board meeting on 12 October.
Inflation had risen sharply in both the United States and the euro area in September, which was in large part due to the high oil price. Underlying inflation rose moderately. Statements by both the Federal Reserve and the ECB had contributed to increasing expectations of tighter monetary policy in the longer term. As a result of this and of rising inflation expectations, long-term interest rates had risen in the United States, the euro area and Sweden over the past week. Other statistics received from the United States had shown poor outcomes; industrial production fell by 1.3 per cent in September compared with August, retail trade showed only marginal growth and preliminary statistics from the University of Michigan indicated a continued fall in consumer confidence in October.
The stock markets in the United States, Germany and Sweden had fallen, primarily due to rising interest rates and the continued high oil price. Statistics on capital flow indicated a large inflow to the United States in August. This was primarily due to private investors buying US corporate bonds. The capital flow information had contributed to a strong appreciation in the dollar recently.
In Sweden, CPI rose by 0.7 per cent between August and September, which gave an annual rate of 0.6 per cent. The measure of underlying inflation, UND1X, rose by 0.7 per cent, which gave an annual rate of 1.0 per cent. Both outcomes were in line with the forecast in the October Inflation Report. Underlying imported inflation was slightly higher than forecast, which was mainly due to clothing and shoe prices rising more than anticipated and, to some extent, to the unexpectedly high oil price. However, this was balanced by the fact that domestic underlying inflation was slightly lower than expected.
Pricing in the financial markets indicated that monetary policy expectations in Sweden had been dampened marginally in the short term, while expectations of tighter policy had increased in the longer term, in line with global developments. Pricing in the financial markets indicated that the repo rate was expected to remain unchanged today.
The krona had weakened by 1.5 per cent in terms of the TCW index over the past week. This was partly due to the dollar appreciation. However, the krona had also weakened against the euro, which could be linked to the fact that market agents appeared to be focusing on the interest rate differential, which was expected to increase according to monetary policy expectations.
2. The economic outlook and inflation prospects
Deputy Governor Irma Rosenberg presented the draft Inflation Report 2005:3, Annex A to the minutes. This Inflation Report reproduced the main features of the presentations and discussions at the Executive Board meetings on 4 and 12 October 2005.
It was observed in the Inflation Report that international GDP growth was strong last year. The assessment made in the June Inflation Report and at the monetary policy meeting in August was that international growth would be good this year and over the coming years, but would decline somewhat. This picture remained largely unchanged. However, the oil price had been higher than expected and the assessment was that it would continue to be higher than was previously assumed. This was expected to increase international price pressures and subdue growth somewhat.
The slowdown in the Swedish economy at the beginning of the year was temporary. Growth in private consumption had started to increase and was expected to strengthen further in the coming years. Public sector consumption was also expected to increase, particularly in 2006. At the same time the investment upturn that began last year would continue, although it would gradually enter a calmer phase.
Compared with the assessment made in June, GDP growth and capacity utilisation were now expected to be slightly higher this year and over the coming years. The forecast for this year was affected by the slight upward revision to the GDP outcome for Q1 2005. Thereafter, economic growth was assumed to be stronger, due in large measure to more expansionary economic policy; for instance, the repo rate was now lower than it had been in June. The Budget Bill presented in September indicated that fiscal policy during 2006 in particular would stimulate demand more than the Riksbank had assumed in June and August.
There were signs of improvement in the labour market, such as an increase in the number of hours worked during the second quarter of this year. Relatively strong growth in domestic demand in future and the measures recently announced by the Government meant that demand for labour was expected to increase.
Inflation had recently risen somewhat but remained low. The assessment was, as before, that inflation would rise, but at a fairly modest rate. At the beginning of the forecast period the oil price in particular was expected to push up the rate of price increase. After that, rising capacity utilisation in Sweden and abroad, together with diminishing effects of a number of price-dampening factors, was expected to lead to inflation rising. These factors included international price pressures and productivity growth. The inflation forecast had been revised upwards slightly, compared with the June figures.
The risks of lower inflation were considered to be offset by the risks of higher inflation. There was still uncertainty regarding international economic activity, which was linked to the oil price, global imbalances and the depressed long-term interest rates. At the same time, the high oil price risked leading to higher inflationary pressures. There was also considerable uncertainty as to how the factors that had recently dampened inflation would develop in future.
One member observed that the high inflation outcomes in the United States and the euro area, which had been received after the cut-off date for publication in the Inflation Report, were primarily caused by the high oil price. Inflation excluding energy had not risen, which indicated that the oil price upturn had not yet had any contagion effects. The fact that such effects could arise was partly included in the risk scenario depicted in the Inflation Report, although no such effects had yet been noted. At present there was no reason to question the assessment in the Inflation Report.
Another member considered that the inflation figures in the United States were more worrying than the previous member had stated. The member noted that interest rate market agents were already pricing higher inflation and expecting a more pronounced upturn in interest rates in the United States. This development could trigger an adjustment in the property market, too, and influence households’ choice between saving and consumption. However, the member agreed that the changes in the picture were not large enough to change the basic assumptions for the interest rate decision.
Another member noted that higher than expected inflation in the United States also meant that the key rate could be expected to reach a higher peak than was previously expected. The member shared the previous member’s view that this could encourage households to increase their saving - which was desirable from another point of view, as the saving imbalance was a serious problem. However, if the process was a rapid one, it could have greater effects on economic activity than anticipated and thereby affect the risk outlook.
One member noted that the inflation outcome in the euro area had also been higher and that the interest rate market had interpreted the ECB’s
statements and priced an increase in interest rates.
The Executive Board made the overall assessment that the new information received since its meeting on 12 October did not give reason for a change in the view of inflation developments, compared with that presented in the Inflation Report.
The Executive Board decided
- to adopt the Inflation Report as presented and
- to publish the Inflation Report at 8 a.m. on 20 October.
§ 2. Monetary policy discussion
Deputy Governor Irma Rosenberg presented a proposal for the monetary policy decision and at the same time clarified the new assumptions that form a basis for the forecasts.
The forecasts in the main scenario of the Inflation Report were now based for the first time on the technical assumption that the repo rate would develop in line with expectations in the financial markets as they are expressed in implied forward rates. That meant in this report that the repo rate was assumed to be unchanged until around the second quarter of next year, when it would gradually begin to be raised.
It was important to emphasise that the new interest rate assumption was a technical assumption for calculation purposes, in the same way as the assumption of a constant repo rate had been used earlier. It should therefore not be interpreted as a stance regarding which repo rate path the Executive Board considered to be most desirable. There were many different interest rate paths that could lead to roughly the same inflation forecasts over the coming two-year period. As before, the Executive Board would make a decision solely regarding the current repo rate. The fact that the forecast was based on the assumption that the repo rate would follow market expectations thus did not mean that the Executive Board was committing itself to any particular future interest rate decisions.
It was also important to point out that the new interest rate assumption did not entail any change in monetary policy strategy. The ambition would normally be to aim monetary policy at meeting the inflation target within two years. Sometimes there might be reason to allow this to take longer time, as explained in the monetary policy clarification published in 1999. This type of consideration could still be justified from time to time for the reasons reported in the clarification. However, one change was that the simple policy rule which had previously been used to describe the strategy no longer applied. According to this policy rule, the repo rate would normally be cut/raised if the inflation forecast, under the assumption of a constant repo rate, was below/above the target one to two years ahead. Now that the assumption of a constant repo rate no longer formed the basis for the forecasts made, this policy rule was no longer applicable, as it was based on this assumption. Instead, the decisive factor for monetary policy would be whether an interest rate development in line with market expectations was considered to lead to the inflation target being met within two years.
Growth was expected to be good over the coming two years and resource utilisation was expected to increase. Inflation was now low, and was expected to rise at a fairly moderate rate. If the repo rate developed in line with market expectations, which meant an unchanged repo rate at today's meeting, UND1X inflation, which was currently judged to provide the best picture of underlying cyclical inflationary pressures, would be 1.8 per cent in two years time. This would put inflation close to the target. This indicated that market expectations of monetary policy meant an interest rate path that on this occasion gave a reasonable development of inflation in the years ahead. However, there were many paths for the future repo rate that could lead to roughly the same inflation forecasts. If the repo rate instead was assumed to remain constant, inflation would rise slightly more rapidly and was expected to be 2.1 per cent in two years time, that is, roughly in line with the target. As before, there was also reason to observe that household borrowing and house prices were continuing to rise rapidly. The proposal was that the repo rate would be held unchanged.
One member supported the proposal for an unchanged repo rate, but felt some concern over the sustainability of the current economic developments. There were already some signs of a slowdown in the United States, which were probably connected with concern that inflation was beginning to pick up and with the effects of the Federal Reserve’s interest rate increases. It had long been clear that there must be a slowdown in domestic demand in the United States in order to correct global imbalances. However, there was a risk that the correction of the imbalances would be too abrupt and would be combined with a rapid fall in property prices. This could lead to a rapid deceleration in consumption. If this were to occur without an upturn in the euro area, it could be particularly problematic for the Swedish economy, as such a large part of Sweden's exports go there. Given that there were as yet few signs of an upturn in Europe, while economic growth in the United Kingdom had been dampened, there was uncertainty over economic activity. This situation could require much more than just a single monetary policy in Europe. It was difficult to use monetary policy as a means of further stimulation while there were inflation tendencies. In addition, there was a risk that the stimulation would come too late. Weaker growth in the United States could reduce the interest from China in investing in US government securities, which could further push up long-term market rates. This would occur during a downturn, while the effects of the upturn in oil prices would continue to push up inflation. All in all, continued the member, there was an evident risk that international economic activity would begin to slow down as early as next year or the following year. The dilemma was that in this case it would occur in a situation where interest rates were already low in many countries, which limited the scope for monetary policy to counteract this trend.
The economic situation for Sweden looked positive and conditions were good for relatively strong economic growth in future, continued the member. Companies were showing good profitability, investment was beginning to pick up and there were signs of an improvement in the labour market. It was the international outlook and its effects on Sweden that were worrying. A possible international slowdown over the next two years could lead to growth in Sweden not reaching the level indicated in our main scenario. As it would take some time before higher capacity utilisation affected prices, despite an increase in employment, there was a risk that Sweden would enter a more negative phase of the economic cycle with a very low key rate to begin with. The krona might very well be strengthened by the international scenario, which could further dampen exports and hold back price rises. On the other hand, the risk of weaker demand from abroad could be balanced by a continued upturn in domestic consumption and investment, noted the member. According to the member there were thus downside risks for economic activity that could lead to a need to postpone possible interest rate increases.
One member felt that the argument that an excessively low interest rate could be problematic was often put forward. However, what conclusions could really be drawn from this regarding future monetary policy? Given the gloomy scenario depicted by the earlier member, it should reasonably be better to have a low interest rate now as this increased the chances for the economy to accelerate. The member did not share the opinion that the krona would appreciate in a situation with a tangible international slowdown. Historically, the krona had tended to weaken in these situations.
One member pointed out that the low interest rate was one of the reasons why the effects on the Swedish economy of potentially weaker international growth would be balanced. The expansionary domestic monetary policy functioned as a counterbalance to the international downside risk. With regard to the krona, the member maintained that it could strengthen as it was generally considered to be weak at present, and the member also asked why it could not strengthen if there was weaker economic activity in both the United States and the euro area.
A further member supported the proposal to hold the repo rate unchanged. The member pointed out that it could be troublesome if an economy suffered an exogenous shock via foreign trade in a situation with a very low interest rate. There would be less scope to parry disruptions. However, the member claimed to be more optimistic with regard to international economic activity than another member. The turnaround in Japan appeared to have been more stable this time, which meant that the world’s second largest economy was beginning to accelerate. There had even been signs of improvement in Germany. At the same time, there was some slowdown in China, which was good given the overheating tendencies that could be discerned in certain sectors. It thus appeared as though developments were moving in a direction that could contribute to fewer imbalances between the United States and South East Asia.
The member agreed with the picture of a slightly brighter Swedish economy. Private consumption was beginning to accelerate, while disposable incomes and the labour market situation were improving. House building was an important growth engine and the infrastructure investments decided last year still stood. Productivity had been high, which had improved Sweden’s competitiveness. Together with the weak krona and wage developments, this contributed to a favourable situation for the export industry, and could mean that exports would develop more strongly than expected. However, Swedish relative wages were a problem. Low-productive work tasks were not being completed because the cost was too high. This meant that demand for education was increasing, while conditions must be created for new jobs in the services sector. An adaptation of relative wages in Sweden and changes in labour market legislation would facilitate developments. The idea of Sweden, as some people were advocating, returning to the policy conducted in the 1960s and 1980s, that is, a policy of strong stimulation of demand, was no real alternative. This policy had proved unsustainable in the long term and was not compatible with free movement of capital. The member expressed concern for future developments in the labour market and maintained that there was a risk that unemployment would remain high if the wage structure were not altered. The member expressed support for the assessment in the Inflation Report that unemployment would decline, but considered that structural improvements in the economy were necessary to enable a really low rate of unemployment.
One member pointed out, in the light of an earlier member’s comments that there were already signs of a slowdown in the United States, that it was important to note that incoming data were difficult to interpret at present because of the recent hurricanes. The member then expressed an opinion on the monetary policy conclusions of the two inflation forecasts reported, with a constant repo rate and with the repo rate following implied forward rates. Under the assumption of a constant repo rate, the forecast was for inflation to be slightly above target, but this was only at the end of the forecast period, after remaining below target for a fairly long period. This could indicate that an interest rate increase should be made now or in the near future. The assumption of a repo rate path in line with financial market expectations as reflected in implied forward rates would put inflation close to the target, given the successive increases entailed in this interest rate path. Thus, the inflation forecasts using both assumptions could be interpreted to mean that it was not necessary to raise the repo rate now; that it was possible to wait. Another indication that this was the case was that the forecasts using both assumptions put inflation well below target level over the coming period and the fact that forecasts further ahead were more uncertain. The member thus advocated an unchanged repo rate on this occasion. However, the member pointed out that market expectations could not be read with precision from implied forward rates because of the different premiums.
With regard to the uncertainty connected with the forecast, the member noted that productivity growth in Sweden has been surprisingly positive on several occasions. The current forecast assumed that underlying productivity growth would be slightly higher than had been forecast in June. At the same time, it was anticipated, as before, that economic activity would lead to lower productivity growth in future. All in all, the effects meant that unit labour costs and inflation would be reduced by this more favourable assumption of productivity growth. However, the assumption was uncertain and it was important to closely follow future developments. Moreover, the effects of the continued high oil price were an uncertainty factor. Developments in an increasing number of countries had led to high CPI outcomes. The fact that expectations had turned towards oil prices remaining high over the coming years while resource utilisation in more countries had increased or was increasing, meant that the risk of secondary effects was greater. They could already be seen in some countries. The risk for Sweden was also growing as resource utilisation rose.
A further uncertainty factor was the rising house prices and household indebtedness, to which low interest rates and the prevailing macroeconomic stability had contributed. The member considered that this development led to a risk that households would be lulled into a false sense of security, expecting the low interest rates to prevail and would thus take on too much risk. Whether property wealth, household borrowing and capitalisation of real assets affected demand in expansionary phases of the economic cycle had not been closely studied using Swedish data. It was difficult to capture this in forecasting models. In the United Kingdom, the Bank of England was lulled into a feeling of security by the fact that consumption was not increasing at the same rate as borrowing towards the end of the expansionary phase. However, now that house prices had stabilised or fallen in some regions, consumption was also falling. It would appear that the connection was stronger on the downside, at least in the United Kingdom. The member considered that growth in credit and wealth could mean that the interest rate would need to be raised earlier than was indicated by inflation forecasts a couple of years ahead.
Another member agreed with the other members’ assessment that the repo rate did not need to be raised now. However, it was clear that the time for an increase was coming closer, if the outlook remained the same. The member pointed out that it was important to continue thinking about why inflation was so much lower in Sweden than in other countries. The fact that international prices had shown weak growth was not sufficient explanation, as these prices also affected other countries. One explanation was probably productivity growth, but it was at the same time fairly unclear why it had surprised as it had. Another question was how much of the low inflation could be attributed to an adjustment in Swedish prices to the level prevailing in Europe.
It was difficult to determine the relative significance of these different factors and also to evaluate how they could conceivably develop once resource utilisation began to increase. It was worth emphasising that two or three years ago, no one had predicted the current low inflation rate. Could it be that we were now underestimating the force of the upturn in inflation if several factors at once began to affect prices in the opposite direction? This was one of the key issues for the future. Given that the interest rate was very low, there was no reason to wait too long before beginning a normalisation of interest rates once the labour market had turned. Another factor pointing in this direction was house prices and household indebtedness.
One member pointed out that the low inflation in Sweden in relation to other countries was to all accounts due to several factors making an impact at the same time; the increased competition that contributed to putting pressure on, for instance, food prices, productivity growth, dampened international prices, the weak labour market and increases in efficiency as a result of information technology. In addition, it might be the case that the Swedish labour market functioned better than the labour market in the euro area.
One member agreed with this, but maintained that the problem was that there was considerably uncertainty over how long these factors would subdue inflation and what the effects would be when the economy eventually rebounded. It was therefore important to be vigilant with regard to how the factors subduing inflation would develop in future.
Another member noted that there were not many people being offered permanent jobs now and that many were uncertain over developments in the labour market. With regard to the scenario one member had depicted earlier with regard to Swedish inflation, the member considered that a stronger krona in line with the forecast could be a counterbalancing factor that contributed to inflation remaining low for a further period.
One member pointed out the generally positive main scenario in the Inflation Report, at least compared with the spring, and maintained that there was less uncertainty now. Developments in Japan were now more positive than they had been for many years and the situation looked good in other Asian countries. Although the euro area was a little sluggish, it was possible to discern some brighter signs, for instance in the German economy. Sweden was also showing improved figures and there were signs that the labour market was rebounding. At the same time, Swedish companies were increasing their borrowing and their investment. It appeared that the Swedish economy was in a relatively normal stage of the economic cycle, where tighter policy would be necessary sooner or later. The member thus shared earlier members’ views that a situation where the repo rate needed to be raised was probably coming closer. Moreover, the current economic upturn meant that it was quite reasonable to expect continued rises in house prices and indebtedness, which would also call for an interest rate increase. The upturn in economic activity had contributed to prices rising in many parts of the country, not merely metropolitan areas. An increase in the interest rate in a situation where employment and incomes were growing would not lead to any rapid fall in demand, which could be seen in the United Kingdom and Australia. The interest rate increases in these countries were carried out during a fairly long period, without any drastic fall in private consumption or house prices. It would be more dangerous when economic activity turned around and income growth evened out. Then there would be a risk of a negative spiral with falls in demand and house prices. There was a possibility that the United Kingdom was now entering such a period. However, the member did not think that there was imminent danger of this for the Swedish economy; developments were currently bright in general.
The member considered a continued high oil price to be a risk factor, if it made an impact on private consumption in the United States, at the same time as the central bank would be forced to raise its rate more than intended. This could reduce consumption further while saving increased. Such a development could be relatively orderly and have the effect of reducing imbalances in the world economy. However, there was also a risk of an abrupt development and then the negative effects would be greater. However, this was not something included in the main scenario, said the member, and the most probable scenario was an orderly development.
One member also expressed some concern over the continued high oil price and its effects on growth and inflation, if the oil price increase were to spread and lead to higher wage demands and price increases in other areas. However, the member agreed with several other members that there were positive signs with regard to growth in the world, for instance in Japan and large parts of the rest of Asia. This would indicate more sustainable economic activity than seen for a long time. The oil price was high, primarily as a result of the high international demand, although supply shocks had contributed to this during the autumn, particularly in the United States. If growth in the world economy were to slow down and thereby demand for oil were to decline, the oil price would probably be rapidly corrected downwards.
Another member maintained that if we were to see such a gloomy scenario as had been painted at some points of the discussion, with a rapid slowdown in economic activity resulting from rising oil prices and a fall in private consumption in the United States, the krona would probably weaken rather than strengthen. This should promote Swedish exports and thereby counteract the negative effects on demand.
One member concluded the discussion by observing that none of the members had found reason to adjust the interest rate at present. Neither the forecast based on a constant repo rate nor that based on implied forward rates indicated this at present. Inflation was currently low and would rise only gradually during the forecast period. If developments continued in accordance with the inflation forecast, the repo rate would need to be raised in future. The interest rate path indicated by implied forward rates showed an inflation rate in line with the target. However, there were many different interest rate paths that could lead to roughly the same inflation outcome over the coming two-year period.
§ 3. Monetary policy decision (1)
The Chairman observed that the members of the Executive Board were agreed that UND1X inflation at present provided the best picture of underlying inflationary pressures.
The Chairman found that there was only one proposal: to hold the repo rate unchanged at 1.5 per cent.
The Executive Board decided
- that the repo rate would be left unchanged at 1.5 per cent and that this decision would apply from Wednesday, 26 October 2005,
- that the lending rate would remain unchanged at 2.25 per cent and that the deposit rate would remain unchanged at 0.75 per cent with effect from Wednesday, 26 October 2005,
- to announce the decision at 8.00 a.m. on Thursday 20 October 2005 with the motivation and wording contained in Press Release no. 63 2005 (Annex B to the minutes) and
- to publish the minutes of today’s meeting on Monday, 31 October at 9.30 a.m.
This paragraph was confirmed immediately.
(1) Board members who are present and do not enter a reservation have participated in and agreed to the Board's decision.