Heikensten: Economic recovery with low inflationary pressure
The whole speech
Allow me to begin by thanking you for your invitation to today's meeting.
In order to get an idea of economic developments in different areas of Sweden and to explain how the Riksbank works, I pay regular visits to various parts of the country. The items on the agenda for today's visit include a talk at Mid Sweden University on how the Riksbank works, a visit to SCA and a discussion with county representatives about developments in business and the labour market in Västernorrland. At this lunch, I intend to give my views on the economic situation and monetary policy.
Before going into the economic assessment itself, however, I would like to say a few words about the Riksbank's work methods. One of the Riksbank's objectives is to maintain price stability. We have decided to define this as 2 per cent inflation as measured by the consumer price index, with a tolerance range of plus/minus 1 percentage point. Monetary policy is conducted on the basis of a forecast of inflation over primarily the coming two years. If the forecast in this perspective indicates that inflation will be lower than 2 per cent, the repo rate should normally be cut and vice versa. The reason we focus on inflation 1-2 years ahead is because monetary policy affects the economy and inflation with a certain time lag.
This approach has served us well over the decade during which we have conducted monetary policy with an inflation target. Since the target was adopted, inflation has by and large averaged 2 per cent. At the same time, we have experienced better general economic progress than during the two preceding decades. The clear target coupled with our mode of action has also helped to make the Riksbank's policy accessible to both the public and the financial markets. Inflation expectations have been anchored fairly stably around our target and it has been possible to predict our actions rather well.
Nevertheless, there is cause to underline that we do not run on autopilot and mechanically change the repo rate in relation to our inflation forecasts. There may be reasons for us to wait before taking a decision in order to be able to perform more in-depth analysis or obtain more information. There may also be reason to take into account additional aspects in our decision rather than just inflation 1-2 years ahead. Such aspects could include how the economy in general would be affected, inflation in both a longer- and shorter-term perspective and the impact on financial stability. But it is important to have a framework that is as explicit as possible, with clearly defined and stated objectives. In this way, everyone can follow our line of reasoning, question it and comment on it. Monetary policy is most effective when it is firmly established.
International upswing is continuing
Both real economic activity and inflation in Sweden are highly dependent on developments abroad. Broadly speaking, demand and economic growth since the end of last summer have followed the paths expected by the Riksbank at the time and subsequently. However, when seen over the entire period, growth in the US has been somewhat faster than anticipated. This also applies to the situation in Asia and in Eastern and Central Europe. Conditions are different in Western Europe, particularly in the core countries of the euro area - Germany, France and Italy. Here, demand has not risen as expected and the upswing has been delayed.
What has happened since the Riksbank's last decision on the repo rate in February has followed the same pattern. Recent developments in the US economy have also been unexpectedly positive. This also applies to large parts of Asia, led by China and Japan. In addition, a number of Sweden's other important export markets such as the UK and Norway appear to have performed somewhat better than anticipated. As regards the euro area, the most recent data does indeed suggest that the recovery in industry has taken root but at the same time domestic demand continues to be weak.
The situation in China deserves a special mention. Since 1990, output in China has grown at an average rate of 9 per cent per year. Over the past three years, it has been 8 per cent. This means that China's share of global output has more than doubled from 6 per cent to 13 per cent. This situation has great significance for the world economy today. China has emerged as one of the world's major importers of commodities, especially regarding crude oil, copper and agricultural products. For example, China's imports of copper totalled no less than 15 per cent of world trade in copper in 2002, compared with only a couple of per cent in 1990. At the same time, Chinese exports of manufactured goods are increasing rapidly, currently accounting for 6 per cent of world exports compared with 2 per cent in 1990. When we today see an unusually big gap between international price inflation for manufactured goods, which is presently exceptionally slow, and for commodities, which instead is very fast, it is probably somewhat due to what is happening in China.
My overall view is that international economic activity will continue to strengthen in the years ahead largely as we expected before. Growth in the OECD area could reach about 3 per cent this year and average around the same in 2005 and 2006. This is marginally higher than we expected and implies slightly better export conditions. International prices of manufactured goods are likely to rise in tandem with the economic recovery. But by how much is uncertain. There are strong arguments to suggest that the increase in prices could be slower than usual against the background of globalisation and the weak price inflation for manufactured goods. Pointing in the opposite direction is increases in commodities prices, which could have a greater impact were demand to accelerate further. However, futures prices of commodities are, if anything, indicating future price declines.
Demand in Sweden in line with previous forecasts
The Swedish economy has come through the recent economic slowdown comparatively well, with growth in the past three years averaging around 1.5 per cent per year. This is important to remember in order to get some perspective on the outcome of the conduct of economic policy now that employment has fallen somewhat. An important reason that economic activity has not been weaker is that household consumption has been relatively high. This has been bolstered by loose fiscal policy and low interest rates. The relatively weak krona has also boosted Swedish exports and resulted in a considerable trade surplus.
Swedish economic developments have largely followed the paths described by the Riksbank in the late summer and autumn 2003. Growth has gradually strengthened in line with international demand. At the same time, private consumption has continued to be relatively high. Investment has remained weak, however. This picture is confirmed in the latest National Accounts. Growth for the full year 2003 is now judged to have been 1.6 per cent. However, there was a marked change over the course of 2003; between the first and second halves of the year growth rose from 1.4 to 1.8 per cent.
The situation at present is that the upswing in industry has progressed somewhat. Export demand has gathered pace, and firms are fairly optimistic about the future. It also appears that capacity utilisation in industry is continuing to rise. Exports were unexpectedly strong during the final two quarters last year, and business tendency data point to continued optimism in the period ahead. Households' expectations regarding both their own finances and the Swedish economy have also begun to rise again after having fallen somewhat at the end of last year. Retail sales figures indicate a continuation of robust consumption. Low interest rates, rising equity prices and low inflation will boost consumption in the period ahead as well.
While on the subject of household consumption allow me to say something about the recent discussion on household indebtedness. Debt as a proportion of households' disposable income continued to grow last year, and the debt ratio is now approaching the levels seen during the sharp expansion in credit at the end of the 1980s. At the same time, low interest rates mean that households' interest costs as a percentage of income is at a historically low level. The latter means that it is not a great problem today for the household sector to bear its debt even if this debt of course limits households' scope for consumption in the near future.
So this situation hardly constitutes a macroeconomic problem at present. In actual fact, one of the aims of the low interest rates has been exactly to create conditions for households to reduce their saving and maintain their consumption in the weaker economic climate. In spite of this, there is reason for the Riksbank to monitor developments carefully. There is also good reason for individual households to reflect on their debt situation. Interest rates are unusually low at present and will most probably change in the future.
On the whole, there is no reason to make any marked adjustment to the overall outlook for demand in the Swedish economy in the coming years compared with in December or February. Most factors still indicate that growth in household consumption will continue to be stable around 2.5 per cent in the years ahead. The contribution to growth from exports will probably be somewhat higher than previously expected, but a large part of the effect on demand will be offset by the fact that this could also be the case for imports. Local government consumption appears to be somewhat weaker than we thought in December. Investment is a source of uncertainty and could probably pick up somewhat later than forecast. On the whole, therefore, GDP growth may be roughly as we expected in the December Inflation Report and when we took our decision in February, reaching around 2.5 per cent per year in the years ahead.
Low domestic cost pressure
In other words, the outlook from the autumn has so far remained largely intact, and economic activity is strengthening as expected. The situation is different for domestic cost pressure. In this respect, since the middle of autumn 2003, new data has gradually been received that has warranted changes in our view of the inflation situation. These relate to the labour market, which has been weaker than expected with consequences for our assessments of wage developments, to productivity, which has risen more than anticipated, with lower cost pressure as a result, and finally to actual inflation, which has been lower than expected, with effects also on inflation expectations, which have declined somewhat.
Employment growth has therefore been weaker than expected, particularly at the end of last year. The same applies to the number of hours worked. The weak employment growth coupled with a lower number of labour market policy measures has resulted in higher unemployment. This rose from 4.5 per cent in February last year to 6.0 per cent in February this year. Business tendency data and other indicators currently suggest that employment growth in the private sector will be relatively weak in 2004. There are some signs that the labour market may be about to turn, but these are weak. Very recently, for example, the number of new job vacancies has increased slightly and redundancy notices have declined. At the same time, the strained financial situation in the public sector points to weak employment growth in local governments in the period ahead as well. Thus, employment growth is likely to be somewhat weaker during the year compared with our December forecast.
As early as February, we changed somewhat our outlook for the rate of wage increases as a result of the situation in the labour market. There is reason to expect total rises in wage costs to average around 3.5 per cent in 2004 and 2005, which in terms of average collectively agreed levels leaves little room for more than about 2 per cent on average in the years ahead. The wage bargaining round appears so far to have gone relatively well, but the historical experiences of Swedish wage formation suggest that one should be cautious about assuming that we are home and dry. The parties appear to be far from agreement on a number of issues, and there are risks of conflicts arising. As long as the agreements are reached at moderate levels from an inflation perspective, longer agreement periods are preferable. This would alleviate some future uncertainty and should thereby contribute to a better economic situation.
So while employment, measured by the number of hours worked, in the business sector was weaker than expected last year, GDP growth was roughly in line with the Riksbank's expectations. This means that productivity was higher than anticipated. This picture has been confirmed by the latest National Accounts data and was underlined as early as in our December assessment. Since then it has become clearer, and the somewhat better productivity was an important argument for the rate cut in February.
It is likely that the more robust productivity has partly been due to the developments in economic activity; at the beginning of a recovery, existing production resources are utilised more efficiently and only when the revival has been established will a rise in investment and new jobs begin in earnest. But this explanation is not enough since productivity appears to have risen more than normal given the economic situation. This pattern is very clear in the US but is also somewhat discernible in Sweden. One possible interpretation is that it is due to unusually large-scale investment in, for instance, IT at the end of the 1990s. It is also conceivable that it reflects more permanent changes that could be related, for example, to technological advances in IT and telecommunications. At present, it is not possible to determine exactly what has caused the improvement in productivity. However, when we look ahead to the coming years, we see reason to expect a better productivity outlook than that we had in December.
The new inflation data that have been received also point to somewhat more subdued inflationary pressure. The decline in inflation seen over the last six months has to a large extent been in line with the Riksbank's previous forecasts. The reason for this is that electricity prices have adjusted downwards as expected. The fact that the expected decline has been confirmed is positive for the inflation outlook since the risk of contagion to other parts of the economy diminishes. These risks have also decreased in that rent agreements at least so far have been concluded at moderate levels. In addition, the inflation outcomes in recent months have been lower than expected for other reasons as well. This is partly due to lower import prices, although domestic underlying inflation is also somewhat lower than previous forecasts.
Inflation and monetary policy
Permit me now to summarise my view of demand and inflation in the period ahead. It appears that international economic activity is progressing largely as expected. Compared with December, there is perhaps reason to expect marginally stronger activity due mainly to the rapid growth in Asia and the US. Economic developments in Sweden are also essentially following the paths we expected, even if there has been a certain shift between different demand components in that private sector consumption and particularly exports are growing slightly faster, while growth in local government consumption will probably be somewhat weaker. In all, there is still reason to expect GDP growth in the years ahead to be somewhat higher than what the Riksbank believes to be sustainable in the long term. For inflation in the years ahead, it is of course important that there are some unutilised resources to begin with.
So what can we expect in terms of inflation developments? In this respect, the outlook has gradually undergone a number of more marked changes since the end of the autumn. Principally, as a result of lower expected increases in wage costs and better productivity growth, unit labour costs are expected today compared with in December to grow a number of tenths slower in the years ahead. Normally, there are few factors that are more significant for inflation a few years ahead than unit labour costs. In addition, international price pressure is muted. Against this background, most factors suggest that inflation in the next two years will be below the Riksbank's target.
Also with regard to the various risks to economic activity and inflation, I believe there is cause to reason somewhat differently than in December. As the economic upswing is gradually confirmed, the risk of a setback in the coming two years will diminish. Correspondingly, the probability that economic activity could prove unexpectedly positive and that the upswing will be quicker than anticipated is also likely to increase. However, this does not mean that there are no downside risks whatsoever in the short term. One such risk is that the weak labour market in the US or in Sweden - confirmed by the latest data - will worry households and dampen their consumption. It seems unlikely, however, that the upswing could be derailed entirely for this reason.
As regards inflationary pressure, there are a number of question marks as I mentioned earlier. We don't know how long the recent rise in productivity will last. In the inflation outlook that I have presented, some consideration has been given to the new data. At the same time, there are good arguments pointing to both higher and lower productivity growth. The same applies to our view of international inflation. As regards wages, there are also risks on the upside, which are essentially related to the way in which wage formation in Sweden is conducted. It can still not be ruled out entirely that the wage bargaining round will be derailed. Should the outcomes prove higher than expected, it is likely to have an impact on the future level of interest rates.
So far I have not said anything about the exchange rate. There may be reason to touch upon this. For a long time, we have expected a krona appreciation that over the course of the forecast period would bring us to levels approximately in line with what we believe to be appropriate in the long term, given mainly the competitiveness of the Swedish economy. During the period 2002-2003 as a whole, the krona strengthened roughly as expected. Recently, however, this strengthening has been curbed. One reason for this appears to be expected relative interest rates, where there has been a certain shift since the autumn. There may be reason to take this into account and expect a somewhat lesser probability than before of a krona appreciation.
On the whole, this discussion of different uncertainty factors and risks does not lead me to see reason to make any appreciable changes to the assessment I have just presented. The risks I have discussed could perhaps push up inflation somewhat. Nevertheless, with what we know today, the conclusion remains that inflation will be below the Riksbank's target one to two years ahead.
By way of conclusion, allow me to stress that much work remains to be done before the next Inflation Report is published and it is this that will form the basis of our decision about the repo rate at the end of March. So it is not just today's assessment that could be affected by new data. In the light of the analyses currently being carried out, the Executive Board may assess some of the factors that are significant for inflation in a different way than I have today.
Thank you.
