Bäckström: Swedish monetary policy outside EMU
I am pleased to note that the Centre for Business Policy Studies (SNS) is maintaining the tradition, together with the Riksbank, of arranging the Monetary Policy Forum. This is our sixth annual opportunity of discussing monetary policy issues in the Forum and what could be more natural on this occasion than to consider the monetary policy future in Europe in connection with the launching of what is perhaps history's most momentous project in this field. Besides the question of how monetary policy will be conducted in the euro area, an equally important issue for us in Sweden is how our monetary policy will be affected when Stage Three of EMU starts and Sweden does not participate initially.
The discussion of “Monetary Policy in EMU” is to be opened by my Finnish colleague, Dr. Sirkka Hämälainen, and I want to take this opportunity of wishing her heartily welcome to Sweden. In this context I should also like to express my appreciation of the close cooperation between Suomen Pankki and Sveriges Riksbank, which we reckon to continue even though Finland participates in the monetary union.
The topic assigned to me is “Swedish Monetary Policy Outside EMU”. First I shall paint my picture of what an outsider position will entail for Sweden's economic policy. Then I will look at how monetary policy in Sweden is constructed in principle and conclude with some remarks on the current situation.
ECONOMIC POLICY OUTSIDE EMU
In that Sweden will be outside the euro area initially, our economic policy will be constructed to fulfil the goals that are set up in Sweden. These goals are based in turn on our domestic priorities together with the commitments we have as a member of the European Union. To a large extent, Sweden's priorities are the same as the EU's: implement an economic policy constructed for stability and low inflation, thereby creating conditions for a good development of prosperity. Participation or otherwise in the monetary union does not affect the need for a balanced economic policy.
Monetary policy aimed at price stability
A political majority has stated that membership of the European exchange rate mechanism, ERM, is to be seen as a preliminary step towards participation in the monetary union. Thus there are reasons for supposing that the question of ERM membership will not be topical until the issue of future EMU participation has been resolved. Meanwhile, the Riksbank will continue to operate with a flexible exchange rate regime and an explicit price stability target for monetary policy. Here it should be underscored that experience to date indicates that this regime is working well.
From the beginning of 1999 the Riksbank's present independent position will be formally embedded in law. The price stability objective will have statutory force and the function of the Riksbank will be to ensure that this objective is fulfilled. The amendments to the legislation on the Riksbank have already been approved once by the Riksdag (Sweden's Parliament) and a second vote on this constitutional change is scheduled for after the general election in September. The reform is supported by five of the political parties and thus by a clear parliamentary majority.
These changes meet EU requirements concerning monetary policy and the role of the central bank. Whether or not we adopt the single currency, Sweden has undertaken to have an independent central bank and this entails a far-reaching delegation of monetary policy decisions on the basis of an operational objective.
Budget policy aimed at surplus
Similarly, Swedish fiscal policy will continue to be based both on our national objectives and on the requirements that have been adopted for the European Union.
An objective for Sweden's public finances has been set up by the Government and approved by the Riksdag. In the medium term the government budget is to show a surplus, measured as the cyclical average, of 2 per cent of GDP. In that a surplus is generated, government debt can be reduced so that future challenges which Sweden's economy may face can be tackled without creating problems for government finances.
Considering the government debt that has accumulated in the 1990s and the fact that Sweden's public finances are relatively sensitive to cyclical activity, this objective is reasonable. Current assessments, based on existing decisions and official economic policy, suggest that the budget policy target will be met in the coming years.
Besides the greater awareness of the importance of stable finances, the conditions for fulfilling the budget objective have been improved by institutional changes to the budget process. Sweden's new budget process is inherently more disciplined than before and this can contribute to the maintenance of sound finances.
The Swedish objective of government financial consolidation is in line with the EU budgetary aim of positions close to balance or in surplus over the medium term. So Sweden's fiscal policy is also subject to EU agreements even though we have deferred participation in the monetary union. Sweden has undertaken to aim for the objectives in the Stability and Growth Pact.
More generally, as a member state, Sweden has undertaken to regard its economic policy as a matter of common concern to all EU member states. We will accordingly be participating in the economic policy discussions and be subjected to assessments and evaluations. We are also taking part in the drafting of the broad guidelines of economic policies; these guidelines, which can be likened to fiscal plans for the Community, are to be incorporated in the construction of EU countries' economic policies.
Sweden moving towards stability
A broad consensus has been established—in the world in general as well as in Sweden—about the objectives and institutional arrangements that are appropriate for economic policy, as well as the importance of international economic cooperation. The inflation economy that prevailed earlier caused major problems. An expansionary economic policy with subsequent devaluations does not generate any lasting positive effects in the real economy. It is liable to result in high inflation that erodes the short-run gains in terms of costs. The longer the period of inflation, moreover, the more painful will be the readjustment to a more sustainable economic development based on low inflation.
Developments in Sweden have been moving in the right direction for a number of years.
Inflation is low and growth favourable. Earlier imbalances, in the government budget as well as in foreign transactions, have been turned into surpluses. Whether the adjustments in the labour market are sufficient to permit a sustained and more extensive increase in employment in the coming years remains to be seen.
FLEXIBLE EXCHANGE RATE WITH PRICE STABILITY TARGET
Sweden is in its sixth year with a flexible exchange rate regime and the fourth year with a specific price stability target. Insights into the new regime have been gained step by step, both in the Riksbank and in the Swedish community in general. The construction of the regime and its underlying ideas have been discussed over these years, not least in these Monetary Policy Forums.
The present monetary policy regime provides good conditions for long-term price stability and thereby for an economy that functions better. It also contributes to an appropriate short-run balance between acting as a clear anchor for price stability and promoting real economic stabilisation. Basically, it is a matter of minimising both the fluctuations around the price stability target and those around the Swedish economy's growth trend. Cyclical fluctuations cannot be eliminated but we can hope that sequences of intense boom-and-bust are a thing of the past.
Low inflation and good growth go hand in hand
When inflation converges internationally and long-term inflation expectations are permanently lowered to the level of the target for price stability, the core of the inflation process lies in the relationship between prevailing demand and the sustainable level of output. If resource utilisation is so high that it becomes strained, inflation will tend to rise. Conversely, when resource utilisation is low, inflation tends to fall. To a large extent it is the interaction of demand and supply that determines whether inflation is rising, falling or constant in relation to the 2 per cent target.
Monetary policy is capable of exerting a direct effect—through the repo rate—on only one of these two sides of the economy, namely demand. The supply side is essentially influenced by structural factors that the Riksbank cannot control directly. But permanently high inflation can indirectly impair the workings of the economy, leading to weaker growth. By the same token, today's low inflation should help the supply side to function more efficiently.
Seen from this angle, it is easy to appreciate that a policy for safeguarding price stability should be able to go hand in hand with a development of demand that matches the supply side's sustainable capacity. In the event of a demand shock, even in the short run the Riksbank's objective therefore coincides with that of general economic policy—a favourable real economic trend. In view of the discussion that arises from time to time about objectives for the Riksbank, this should be born in mind.
Consider these two examples:
- Suppose, first, that a shock causes demand to fall; resource utilisation will then decline, with a risk of inflation dropping below the 2 per cent target. The monetary stance must then be adjusted so that the rate of inflation returns to the target. This involves lowering the repo rate so that demand recovers.
- Suppose, instead, that a shock causes demand to rise faster than the supply side can cope with. Inflation will then be liable to rise above the targeted rate. The monetary stance must therefore be tightened so that inflation falls back to the target, which involves subduing demand to output's sustainable level.
In the various situations, the impact of monetary policy can be contractive, expansive or neutral, depending on the degree of resource utilisation and the risk of inflation deviating from the target in either direction. The relevant rate of inflation here and the one that is affected by total demand, is the underlying trend. In that this trend is in line with the target, conditions are created for sustainable growth. In the event of a demand shock there is thus no conflict between combating inflation and promoting growth.
Transient price effects of secondary importance
Besides effects from demand shocks, price movements are influenced by disturbances in supply. Supply-side shocks vary both in kind and in the duration of their effects on the economy. The monetary policy response accordingly depends on how suddenly such a shock occurs and how soon it fades. A supply shock with a direct effect on prices, for example a VAT increase, should be distinguished from an increase in productivity, which affects demand via resource utilisation. Let me start with the former case.
A VAT increase will tend to subdue demand at the same time as it adds to the 12-month rate of CPI inflation over a period of one year. In this case, in the first year after the tax increase, demand and inflation move in opposite directions. If the monetary stance were to be tightened in order to achieve a prompt return to inflation's targeted rate, it would cause demand to fall further. Temporary price stabilisation would be accompanied by a destablisation of real economic activity. A similar dilemma is posed by VAT reduction, which stimulates the economy and lowers registered inflation. A monetary policy response in the form of a repo rate cut—to return inflation to the target—would act as a further economic stimulus. Here, too, temporary price stabilisation would be accompanied by a destablisation of real economic activity.
I have used VAT adjustments here as an example of shocks from the supply side. Effects of the same type are associated with changes in other indirect taxes, for instance the current proposal to lower the tax on tobacco. The same applies to international oil price movements, as well as to the Riksbank's own interest rate adjustments, which have a direct impact on house mortgage interest costs.
How, then, should monetary policy deal with a situation in which prices are exposed to supply shocks of this kind? Part of the answer lies in the circumstance that—provided people's medium- or long-term inflation expectations remain unaffected—most shocks of this type are transitory. The Riksbank has concluded that in the normal course of events, transitory price effects of this type ought to be a secondary consideration in the construction of monetary policy in so far as they do not affect inflation expectations and thereby have a more permanent impact on inflation.
Effects of supply-side shocks may, however, be more permanent. An improvement in trend productivity, for example, implies lower resource utilisation and thereby lower underlying inflation, and vice versa. Such supply shocks should therefore influence the construction of monetary policy.
Advantages and drawbacks of the CPI
A drawback with the CPI is that it combines information about both the underlying trend and temporary price movements. So why has the Riksbank chosen the CPI as its target variable instead of some indicator of underlying inflation? The reason lies in the CPI's advantages as a familiar index that is used in many contexts, whereas the opposite applies to the concept of underlying inflation.
Statistics Sweden could, in fact, supplement the CPI figures with a series on underlying inflation that would ultimately become more familiar. This is a matter that the CPI Enquiry will be considering. But it is still not clear just how an index of underlying inflation should be constructed.
What is clear, however, is that this function cannot be performed by the Net Price Index because that excludes only one type of transitory effect, namely indirect taxes. In that it includes other transitory effects—for example mortgage interest costs and international oil price movements—the Net Price Index gives a picture of inflation's underlying trend that is incomplete and distorted.
Another alternative would be to use the EU's harmonised index of consumer prices (HICP), which is constructed to exclude house mortgage interest costs. One disadvantage of this index, however, is that some of its components have not yet been finalised.
As an aid in the analysis of price trends, the Riksbank's inflation reports include various indicators of underlying inflation.
The tolerance interval of ±1 percentage point serves, among other things, to accommodate the various transitory price movements that normally occur. Experience has shown, however, that in periods when the transitory price effects are extensive, inflation may move outside the interval without this by itself having consequences that should be considered in monetary policy. Thus, the tolerance interval represents the range that is normally expected for fluctuations in CPI inflation. It serves to indicate that the target variable, the CPI, cannot be controlled exactly.
Still, I am convinced that people's awareness and understanding of these factors will gradually grow as time passes. This will happen not least as a result of the efforts of all those knowledgeable analysts and business journalists who follow monetary policy closely and mostly do their utmost to explain these complex details to a broader public.
Exchange rate of indirect importance for monetary policy
Now I should like to say something about the role of the exchange rate. Our monetary policy strategy is premised on a flexible exchange rate. In other words, the Riksbank does not explicitly target the exchange rate and, given Sweden's position on ERM, it seems today that the rate will continue to be flexible at least until a new political decision is taken on the issue of EMU. This complies with EU agreements. ERM membership is voluntary. But even in a flexible regime, exchange rate movements constitute an important policy indicator, albeit more indirectly.
- If the exchange rate weakens and remains at the weaker level, demand in the Swedish economy is stimulated via increased exports and decreased imports. In this way, a permanent depreciation can lead to higher inflationary pressure. The Riksbank then has to keep its instrumental rate at a higher level than otherwise in order to fulfil the price stability target.
- A permanent appreciation of the krona tends to dampen demand, leading to lower inflationary pressure. All else equal, the Riksbank can then keep the instrumental rate at a lower level.
With reference to the exchange rate, the Riksbank is wont to talk of “permanent” or “lasting” developments and that is applicable here. We do not react to occasional market movements because they are unlikely to influence inflation prospects.
The Maastricht Treaty requires all member states to treat exchange rate policy as a matter of common interest; this applies even if the exchange rate is flexible. An EU country is expected to focus economic policy—both monetary and budget policy—on stability to the same degree as the other EU countries. This, basically, is the absolutely crucial condition for sustained exchange rate stability. An exchange rate that deviates permanently from the long-term real equilibrium level may be an indication that economic policy lacks credibility. In such a situation the EU member states have a “common interest” in discussing how confidence in economic policy can be enhanced.
During the spring there has been a discussion in Sweden about the exchange rate tendency in connection with the move to Stage Three of EMU. What will happen if, for instance, companies adopt the euro as their unit of account? The Riksbank has underscored that it does not see any strong reasons for large currency outflows on account of the monetary union. And if such outflows were to occur and the krona suddenly weakened, we consider that countervailing forces would act in financial markets.
The value of a currency is determined by numerous factors, of which the most important is confidence in the country's economic policy. The general trend towards internationalisation in recent years has entailed relatively large currency outflows due to portfolio diversification among resident investors. But at the same time the krona has displayed an appreciating trend as growing confidence in economic policy generated a compensatory currency inflow.
All in all, conditions for a stable exchange rate between the krona and the single currency should be good. There are many indications that the European Central Bank will define a price stability target that resembles Sweden's. Similar inflation trends, together with confidence in economic policy as a whole, should result not only in stable nominal and real exchange rates between the euro and the krona but also in conditions for a reasonable valuation of the krona.
MONETARY POLICY’S CURRENT CONSTRUCTION
A long target horizon permits a gradual monetary adjustment
When assessing monetary policy's construction, we look ahead to assess the path of inflation in the coming twelve to twenty-four months. A relatively long target horizon does have the drawback that earlier assessments may have to be revised. One reason for choosing a long horizon—with the attendant difficulties in making reliable forecasts—is the time lag before effects of monetary policy materialise. By acting in good time, it is possible to adjust interest policy more gradually and smoothly. In this way, any consequences that altered inflation prospects may have for the monetary stance will usually be gradual and small.
With a shorter target horizon, the inflation assessments might be more accurate but the monetary policy responses would have to be more pronounced in order to fulfil the target. There would also be wider fluctuations in output and employment. Real factors like production and employment are accordingly taken into account when making this choice.
In view of the difficulties in assessing and forecasting economic tendencies and monetary policy effects exactly, the Riksbank has opted for one to two years as its normal target horizon.
Lower CPI figures
The forecasting problem can be illustrated with assessments from the National Institute of Economic Research; I use them here simply as an example, not to single out the National Institute from other forecasters. But the National Institute is one of the forecasters with the largest resources and it usually “sets the trend” for other forecasters. Last November the Institute counted on annual rates of CPI inflation of 2.0 per cent for 1998 and 2.5 per cent for 1999. In other words, inflation would move gradually upwards from the price stability target. In keeping with this, the Institute also counted on an monetary policy realignment, with an expected increase in the repo rate from 4.10 to 5.5 per cent, that is, by 140 basis points.
This March—only four months later—the National Institute's prediction of inflation had changed considerably, to annual rates of 1.1 and 1.5 per cent for 1998 and 1999, respectively. The Institute's most recent assessment is that during 1998 and 1999 there is no need for any change from the present level of the repo rate.
What has happened to forecasters' perception of inflation?
Increased element of transitory price effects
Basically, nothing has happened in the appraisal of the upward cyclical phase. The National Institute, like most other observers, counts on good growth in the coming years. The strength of the upswing may vary but the overall picture is that demand growth will be somewhat above the long-term trend, which implies rising resource utilisation.
But besides expectations of lower wage increases, there have been a number of favourable price shocks on the supply side. Falling international crude oil prices have already led to price reductions for energy-related goods. Moreover, falling interest rates have lowered house mortgage interest costs. Then there is the Government's proposed reduction of tobacco tax. However, the price shocks from the supply side are mainly transitory and tend to lower the CPI temporarily but not necessarily inflation's underlying trend. It follows that these effects probably should not have much influence on monetary policy. This holds even though the CPI change figures are likely to move outside the lower tolerance limit for a time during 1998, mainly on account of the lower tobacco tax.
How strong is the upward trend?
As I mentioned earlier, the primary concern for monetary policy is inflation's underlying rate and trend. In that the monetary conditions have been expansionary for some time and both economic activity and resource utilisation are rising, Sweden should be moving towards a situation with some increase in underlying inflation. This phase may not be imminent or involve anything dramatic but rising resource utilisation will ultimately lead to it. But there are some factors which may defer it. They have to do with the momentum in the Swedish economy and the impact that may come from the crisis in Asia. Another central factor is the exchange rate's underlying trend. In addition there is the increasingly discussed uncertainty about the sustainability of international share prices.
To sum up, most of the evidence suggests that economic growth in Sweden will become stronger in the coming two years. To date, however, the upswing has been accompanied by subdued price movements. This mainly reflects a number of favourable transitory effects but there may also be signs of a somewhat lower inflation propensity.
The strength displayed by productivity may perhaps enable the Swedish economy to expand somewhat faster without immediately generating inflationary pressure. But it is still too early to form a more definite opinion on whether the productivity trend has changed. It is now important that incoming statistics are still followed closely and assessed. As I just said, there is no reason at present to alter the repo rate.
