Heikensten:The economic situation and monetary policy
Thank you for the invitation to the Swedish embassy and for the opportunity to speak about the economic situation and Swedish monetary policy. It is important for the Riksbank to come to London occasionally to meet the participants in the financial markets that regularly follow the Swedish economy and our work.
My intention today is to give a broad picture of my views concerning the inflation outlook in the years ahead. The starting point will be the most recent Inflation Report in May and the assessments in the press releases that the Riksbank has published thereafter. I will also discuss particularly some issues that have recently been touched upon in the monetary policy debate: oil price developments and household indebtedness as well as developments in residential property prices. Following my introduction I look forward to a lively discussion.
International developments
The Riksbank’s assessment in the latest Inflation Report in May was that international economic activity would continue to strengthen this year and during the coming two years. During the summer we concluded that there was reason to be somewhat more optimistic about economic growth this year. The high oil price was indeed judged to stifle growth somewhat, but not to such an extent that would derail the upswing.
Developments have differed somewhat between various regions. Our assessment in May was that economic growth in the United States would slow at the end of this year, chiefly due to the assumption that fiscal policy would be less expansionary. Coupled with gradually rising interest rates, it was judged that this would dampen private consumption and housing investment somewhat. Despite the slowdown, however, the economic outlook was expected to be mainly bright over the coming years. Data for the second quarter indicates that private con-sumption has slowed somewhat earlier than anticipated. It is possible that this is partly due to the high oil price, which has lowered households’ real incomes. Even though more recent data from the United States has given a mixed impres-sion, I think there is reason to be somewhat less optimistic about developments in the United States in the period ahead than indicated in the assessment in May. Growth this year of almost 4½ per cent now appears likely. In 2005 and 2006 there is reason today to expect about 3½ per cent annual GDP growth.
Growth in the euro area was judged in the latest Inflation Report to increase gradually this year following a very weak development in 2003. New information during the summer broadly confirms this outlook. In Germany it is mainly net exports that are contributing positively to GDP growth, while in other countries, notably in France and Spain, it seems to be domestic demand that is growing re-latively quickly. For the euro area, it now appears reasonable to expect growth of 2 per cent this year and a slightly higher figure in the next two years.
For the Nordic region, the UK and the rest of Europe as well as for Latin America and Asia, many analysts have revised up their growth forecasts since the end of the spring. This is partly because of a number of new statistics releases from these areas which have been more robust than many had expected. Russia and some Latin American countries are also benefiting from the high prices of commodities. As regards Asia it is primarily the outlook for the Japanese economy that has be-come increasingly optimistic. At the end of last year economic growth there was at its highest in fifteen years. There are many indications that there has been a slowdown in the Japanese growth rate very recently but in relation to the devel-opments of recent years it is still relatively strong. China has experienced a rapid industrial expansion with average annual growth of almost 10 per cent in the past ten years. This year, too, growth has been high. One contributory factor to the strong growth is that investment has increased sharply. Nevertheless, most analysts expect growth in China to be somewhat lower in the years ahead, owing to the measures now being taken to dampen demand.
Taken together it is possible that there now, as at the time of the monetary policy meeting in August, is reason to be somewhat more optimistic about economic growth abroad this year compared with the Riksbank’s assessment in May. The high oil price could indeed dampen the upswing somewhat and there have been some less positive signals over the summer from the United States, among other countries. But it appears that this will be countered by growth in other parts of the world, which seemingly will be more robust than forecast in May.
In this context it may be worth noting that Sweden’s trade with countries such as China and India is still relatively small, but that the countries outside the OECD area (OECD-19) together account for about 20 per cent of Swedish export markets. This can be compared, for example, with the United States, whose share of Swedish exports is just over 10 per cent. If the performance of these countries is strong, it is of course of great significance for Sweden.
Financial market developments during the summer have not entirely been in line with expectations. Equity price developments have been relatively weak. Both short-term and long-term interest rates have accordingly been weaker than we anticipated. The cause of this weaker development in interest rates has been uncertainty, primarily about the strength of the US economic upswing and the ef-fect of the high oil price on global growth. However, the uncertainty seems again to have diminished somewhat recently, partly as a result of more robust labour market data in the United States. This has led to a recovery in the stock market.
When the Riksbank in its Inflation Reports looks ahead a couple of years, it is important not to be too influenced by the recent weeks’ movements in the financial markets, but rather to stick to our fundamental, longer-term assessment of economic developments. As long as we don’t markedly change our view of future economic prospects and the inflation outlook, there is, in my opinion, neither any reason to change the view from the early summer of future financial conditions. Interest rates today are low in an historical perspective and they can be expected to rise somewhat as global resource utilisation picks up. At the same time, the krona should appreciate somewhat, partly due to the continued substantial sur-pluses in Sweden’s trade with other countries.
The oil price and monetary policy
In May, the Riksbank forecast international inflation to increase during the year, primarily as a result of rising global resource utilisation. Developments since then point to a somewhat stronger rise in prices than expected. To some extent this is due to the unexpectedly high price of oil. A couple of weeks ago the oil price rea-ched record-high levels and is today roughly as high as during the oil crisis in1979-80.
An unexpectedly steep increase in global demand for oil, not least from the rapidly growing Chinese economy, seems to have been a significant driving force behind the rise in the oil price in 2004. This has led to a marked decrease in the idle capacity in the oil sector. Other relevant factors include unease over the situ-ation in the Middle East and the various circumstances surrounding the Russian company Yukos. The rapid, short-term changes that have characterised the price formation in the oil market also point to a not insignificant element of specula-tion. Exactly what weight should be attributed to different factors from time to time is difficult to establish. But it is clear that the overall price rise in 2004 has been demand-driven to a larger degree compared with other occasions when the oil price has risen sharply. The increases in the oil price in 1973-74, 1979-80 and 1990 were as we know essentially a result of reductions in the oil supply.
The demand for oil will of course play an important part for price developments in the period ahead as well. A key aspect is that more and more countries that previously were economically undeveloped will take the step towards more energy-intensive production. Even if new innovations can be put to use in a couple of years, the fast growth in many developing countries points to a relatively high oil price in the period ahead, too. Nevertheless, most analysts expect the supply to be so large that the oil price will fall back fairly sharply from current levels in the next few years, as geopolitical concerns diminish. Futures prices indicate that the oil price will fall by between USD 5 and 10 per barrel in the coming two years. In the long run it can also be expected that the oil price will be held back as more energy-efficient production methods are gradually introduced also in the recently industrialised countries, in the same way that happened in the West after the oil crises in the 1970s. But this is likely to be a process that takes a fairly long time.
An increase in the oil price is often problematic from a stabilisation policy per-spective. This is because it tends to push up inflation at the same time as it has a dampening effect on the real economy. This makes it difficult to simultaneously stabilise inflation and output. The appropriate policy must quite simply be determined from case to case and is due in part to the persistence of the rise in the oil price, how other prices in the economy are affected, and to what factors that otherwise drive the inflation process.
There are some aspects to the current situation that may be worth highlighting in this context. As I said earlier, the rise in the oil price in the past year has, more than previously sharp oil price increases, been due to a pickup in the demand for oil owing to world economic expansion. A rise in the price of oil that is largely a result of high international growth is likely to have less negative consequences for the real economy than a more outright supply shock.
In addition to this, of course, the economies in the West – and particularly the Swedish one – are far less dependent on oil than they were during the oil crises in the 1970s. This should mean that the impact of oil price rises on both the real economy and inflation would be considerably more limited than what proved to be the case at that time. The assessment of most analysts today seems also to be that the increase in the oil price in 2004 could indeed curtail somewhat the inter-national economic upswing but that the recovery will not be derailed.
The impact of the oil price on inflation, too, should be less than before. Since the oil price shocks in the 1970s there has been a shift in the economic policy regime in many countries which has meant that low, stable inflation has become one of the central targets of economic policy. Today, low-inflation policies seem to enjoy high credibility in most countries, which is indicated in part by the fact that the rise in the oil price at least so far has had small effects on inflation expectations. The risk of an oil price rise spreading to wages and other prices in the economy is therefore also likely to be low. Furthermore, in the case of Sweden it should be added that inflation is rising from a low level and that agreements in the labour market have been settled for 2005 and 2006, further reducing the risk of contagion effects via the jobs market.
To sum up: the fact that the increase in the oil price is partly a demand-driven phenomenon, that the dependency on oil has decreased and that inflation-targeting regimes enjoy high credibility suggests that the higher oil price should mainly be reflected in a temporary, limited rise in inflation and that the effects on the real economy should be fairly small. If this is correct, the general economic and inflationary developments are more significant for monetary policy than developments in the oil price. This brings me to a discussion of the broad outlook for the Swedish economy in the years ahead.
Economic activity in Sweden
Despite the Riksbank’s relatively optimistic outlook for the Swedish economy compared with other forecasters, we have gradually revised up our forecasts in the past year. Our assessment in May was that economic activity in Sweden would continue to strengthen in the period ahead owing to both relatively loose monetary and fiscal policy and an expected economic upswing abroad. GDP growth was forecast to average almost 3 per cent from 2004 to 2006.
One reason that there now is cause to revise up the growth forecast is that exports once again have increased unexpectedly fast. Export growth has been robust because international demand in general has grown more rapidly than an-ticipated and because the relative price of Swedish exports has been unexpectedly low. As I pointed out earlier, some 20 per cent of Swedish exports go to countries outside the OECD area where economic growth has been vigorous. During the first half of this year exports of goods, measured in current prices, rose 9 per cent compared with the same period in 2003. Swedish exports have also benefited in part from the revival in the IT and telecommunications sectors. However, robust growth in exports is not unique to Sweden; average export growth in the EU has been roughly as strong during the first half of this year.
Another positive signal is that a pickup in investment, which the Riksbank and other forecasters have been waiting a long time for, seems to have begun. This was a major contributory factor to our more positive outlook for the Swedish economy at the latest monetary policy meeting in August.
Since the monetary policy meeting in August, the government and its collabora-tion parties have announced proposals, intended to be presented in more detail in the budget bill, which may entail a more expansionary fiscal stance in the period ahead. These proposals include cutting income tax for households, abolishing in-heritance and gift tax, lowering wealth tax and increasing support to local governments. In total, this is estimated to provide a stimulus of just over SEK 20 billion in 2005 and 2006. How large the impact of this will be on economic growth and inflation is not so easy to forecast. This is partly dependent on the extent to which households save their increased incomes. Using traditional rules of thumb it could entail a few tenths of one per cent higher GDP growth in both 2005 and 2006. As things stand, therefore, there is reason to expect growth to average above 3 per cent in the coming two years.
In spite of the fact that the economy has performed fairly well for almost two years, developments in the labour market so far have been weak. Even though the demand for labour is still low, the situation is nonetheless judged to have stabilised somewhat. The historical pattern is for employment to lag behind a change in output somewhat. It is common for economic upswings to begin in manufacturing, which then results in a rise in manufacturing employment. Gradually, there is also an increase in output in the services sector, which boosts employment in that sector. Our judgement is that this will also be the case this time. At the same time, it is important to remember that a process of structural change has been underway in the Swedish economy for some time, where jobs in the manufacturing sector have been replaced by jobs in the private services sector. This development, too, is likely to continue in the period ahead. Consequently, there is no reason to expect robust growth in manufacturing employment in the coming years. Furthermore, in the current economic upswing, it appears to be taking somewhat longer than usual for jobs to pick up. The number of hours worked, on the other hand, has risen more than normal for the current economic climate. Firms thus appear to have boosted production by making greater use of already employed staff instead of recruiting new employees. Although there is still considerable uncertainty about developments in the labour market, it is nev-ertheless likely that the number of employed will increase some time around the turn of the year. At roughly the same time unemployment should be able to begin to fall.
The other side of the coin of robust growth and a subdued jobs market is continued high productivity growth and low cost pressures in the Swedish economy. The extent to which this is temporary and due to economic activity, or is a result of longer-term factors, is hard to determine. It is therefore also difficult to judge how productivity will develop in the period ahead. Our assessment has been that the robust productivity partly reflects longer-term changes but that it is also due to the economic situation, and is thereby partly of a more transitory nature. We examine this assessment regularly, of course.
In order to judge how demand in the economy will affect future inflation it is important to have a good picture of resource utilisation. There is no exact measure of this, however. In the latest Inflation Report, the assessment was that total resource utilisation in the economy was still relatively low but that it would rise gradually and in 2006 reach levels at which it no longer would restrain price and wage increases. New data from the National Institute of Economic Research, for example, suggest that resource utilisation is still relatively low, even though all indicators are not completely in accord. The indicators based on labour market conditions point consistently to low resource utilisation. The somewhat stronger GDP growth that there is now reason to expect suggests, however, that the idle resources will be employed at a somewhat faster rate than forecast both in May and August.
House prices and debt levels
One issue that is of importance to the Riksbank for several reasons is household indebtedness and developments in residential property prices. The Riksbank is required to maintain price stability. In addition the Bank is required to promote an efficient, stable payment system. Furthermore, if these objectives are met, the Bank should also attempt to take account of the other targets of economic policy, e.g. economic growth and high employment. Developments in household indebtedness and house prices can be significant in all these respects.
When forecasting inflation, the Riksbank tries to take account of the connection between indebtedness and developments in house prices, on the one hand, and inflation on the other. For example, it may be a question of judging how large an impact rising house prices have on household wealth and, at the next stage, on demand in the economy. These relationships are in no way simple or straightforward, and even though statistical studies show a clear correlation between these different variables, it is entirely possible that there are underlying explanations, e.g. sentiment in the economy, that could affect the relationships. In any case, when producing our forecasts, we do our best to estimate the effects of house prices and indebtedness.
The Riksbank also illustrates regularly the effects on the payment system of deve-lopments in house prices and household debt, mainly in the Bank’s reports on financial stability. Our assessment so far is that there are no signs that property prices are markedly out of balance. High growth in households’ disposable incomes, low interest rates and a low level of housing construction, coupled with inc-reased demand for housing, may to a large extent justify the price rises seen in the housing market at national level, as well as current debt levels. It is also pos-sible that the increase in debt is partly an adjustment to an economy with – compared with before – better functioning financial markets, low inflation and expec-tations of relatively low interest rates in the future. Most indications also suggest that the majority of households would be able to service their debt even if interest rates were to rise sharply. Nor does it appear likely that banks and mortgage institutions would risk incurring losses that would appreciably affect their capital bases. This wasn’t the case either during the financial crisis at the beginning of the 1990s. While the banks’ losses during the crisis were largely propertyrelated, they did not stem from lending against collateral in housing but from loans secu-red by commercial properties. And prices of commercial properties have this time not at all shown the same development as prices of residential property. On the contrary, they have fallen during the economic decline and have still not recovered.
What remains then are the effects on the economy in general. In this respect it is perhaps first a question of whether the build-up in debt and developments in house prices should prove unsustainable, that the prices therefore should fall and that this on the whole should result in a decline in the real economy in general. The risk of this happening is of course related to your view of residential property prices. If they seem to be highly elevated in relation to before, or are difficult to justify through established economic arguments, the risk is greater. The risk of problems is also connected to other economic developments. If these are stable, there is a smaller probability that those who have borrowed to buy a house will have made their decision on what later proves to be the wrong grounds.
An example can illustrate the latter form of reasoning. Interest rates are currently at an all-time low. This indicates that they will rise, which is an assumption that most households ought to have taken into account. However, many of them may also have been counting on rising income, as this is the normal situation; interest rates usually rise at the point when economic growth and thereby income in society increases. There is of course a risk here. A couple of decades ago we experi-enced a long period of both low growth and high inflation. If this should be repeated – although it does not at present appear likely – it is quite possible that many households would encounter difficulties.
To sum up, one can say that the Riksbank in its normal work process does take into account, as far as is possible, the effects of indebtedness and house prices on inflation via effects on, for instance, consumption and general demand. These effects are taken into consideration in our inflation forecasts. In addition, we might choose to include effects on the payment system and on general economic performance if, for instance, we saw a significant risk of a future financial crisis. The clarification of our monetary policy that we published in 1999 states that we would explain clearly the reasoning behind our decision if this situation were to arise. Naturally, this question – of whether the risks are sufficiently large to requi-re special consideration in the monetary policy decision – is not a simple one. It concerns weighing up difficult issues about which opinions may differ from time to time. So far, however, considerations with regard to the effects of house prices and household indebtedness on factors other than the inflation rate have only played a marginal role in interest rate policy.
Inflation and monetary policy
Let me conclude with a summary of my current views on inflation and monetary policy. Since the early summer, inflation in Sweden has developed roughly in line with the Riksbank’s assessment in May. Inflation is low both by Swedish and international standards, at around 1 per cent. At the same time, it is important to emphasise that the decisive factor for monetary policy is not the present inflation rate, but our views regarding future inflationary pressures.
At our monetary policy meeting in May, the Riksbank judged that inflation would remain relatively stable over the coming year and then increase gradually as a result of rising international and domestic resource utilisation. Given the risks we saw before us – primarily a slightly stronger economic upturn – inflation was ex-pected to be in line with the Riksbank's inflation target a couple of years ahead. This was why the repo rate was left unchanged. The outlook from May had not changed significantly at the meetings in June and August. However, we observed in August that economic growth appeared slightly stronger both in Sweden and abroad. This meant there was also reason to expect slightly higher resource utili-sation. Moreover, the economic upturn had continued for a further couple of months.
What is the present outlook for inflation over the coming years? My own assessment is that the conclusions drawn by the Riksbank at the previous meeting in August remain valid on the whole. It is likely, however, that some upward revi-sion of the inflation forecast may be warranted compared with the May assessment, and perhaps even compared with the August forecast. In the short term the high oil price points to such a scenario. It also appears reasonable to expect somewhat higher resource utilisation over the coming years compared with in May and August. Factors that support this are the strong data releases during the summer and the more expansionary fiscal policy that now appears likely. Another factor is international developments, which appear to be strong. Slightly higher resource utilisation should lead to a slight increase in inflationary pressures over the coming years, compared with previous assessments.
There are, of course, uncertainty factors. I have already mentioned the uncertainty surrounding international developments. Another uncertainty factor concerns developments in the labour market, which in turn are related to productivity growth.
On the whole it is likely that there will be reason to revise up somewhat the infla-tion forecast compared with the May Inflation Report. Inflation a couple of years ahead can nevertheless be expected to be roughly in line with the Riksbank’s target. But as economy activity continues to strengthen, which is likely to imply less idle resources in the economy, there is reason to expect gradually rising inflationary pressures.
Thank you!
