Bäckström: The current situation for monetary policy
First let me express my thanks to the Committee for the invitation to discuss Sweden's economy and monetary policy. In my opening remarks I shall begin by saying something about the Riksbank's recent evaluation of the price stability target and commenting briefly on last week's initial decision by the Riksdag (Sweden's Parliament) on amendments to the Riksdag Act. I shall conclude with our recent implementation of monetary policy, our current assessments of future tendencies and their implications for the monetary stance.
Good experience with price stability target
It is now five years since the Governing Board of the Riksbank decided to adopt a price stability target in January 1993. The decision had to be made at rather short notice, with little time for detailed analyses. At the Riksbank in recent months we have been taking a closer look at aspects connected with the formulation of the target and how monetary policy should be conducted in order to meet it. Experience has been gained and there is a period on which to look back. The conclusion we have reached in discussions in the Governing Board is that the target has functioned properly to date and there is no reason to reformulate it in any respect.
In the period since the target came into force, annual CPI inflation has averaged 1.5 per cent. Thus, the average outcome is below but relatively close to the 2 per cent target and well inside the tolerance interval. In the same period, underlying inflation has been nearer to 2 per cent. In other words, transitory effects that lower inflation have been stronger on average than transitory upward effects. There have been periods when the transitory effects were particularly marked. From July 1996 to July 1997, for example, the 12-month CPI change figures were outside the target's lower tolerance limit. The main reason for this was that falling interest rates held back the CPI increase though decreased interest costs on housing.
Transitory effects are price movements of an occasional nature, generated for example by changes in indirect taxes and subsidies, international price movements for primary products, repo rate adjustments, etcetera. In that such effects raise or lower the general price level, they have an impact on the registered rate of CPI inflation that lasts for a period of twelve months. Provided such transitory effects do not alter inflation expectations, they do not play a part in the inflation process's long-term trend.
When the 12-month CPI change figures dropped below the target's tolerance interval for a period, a discussion arose as to why the Riksbank had ‘missed the target’. In this context it is important to note that the cause lay in transitory effects on inflation and that in the period in question these effects were unusually large—not least on account of the Riksbank's own repo rate cuts. Monetary policy should not be expected to counter its instrumental rate's direct impact on consumer prices; that would be tantamount to monetary policy chasing its own tail, as it were. Over a business cycle, moreover, the interest rate effects normally cancel out.
This leads on to the broader issue of the Riksbank's approach to transitory price effects in general. Is it reasonable for monetary policy to try to counter effects from, for example, altered indirect taxes and subsidies, or from sizeable international primary product price shocks? Rather drastic monetary policy measures might then be needed, with the attendant risk of destabilising production and employment, which is obviously undesirable. The Riksbank has therefore concluded that, provided they do not alter inflation expectations and thereby generate a more permanent impact on inflation, transitory effects should be a subordinate consideration in the construction of monetary policy.
Would it not then be reasonable for this conclusion to be reflected explicitly in the choice of the target variable? A drawback with the CPI is that it contains information about both the inflation process's long-term trend and the transitory price movements. If an indicator of underlying inflation had been chosen instead as the target variable, there would have been no need for the recent discussion. One problem, however, is that such an indicator would not be familiar to people in general. We have therefore concluded that the CPI is preferable on the whole and should continue to be the target variable. But it is indeed important that various indicators of underlying inflation are analysed in the inflation reports, as well as in other contexts, in order to benefit from the information they provide about inflation's long-term trend as well as transitory price movements. In time, moreover, these factors will become more familiar and better understood.
Independence a support
The commitment to a price stability target originated from the Riksbank but the Government and the Riksdag have expressed support for price stability as the monetary policy objective as well as for the price stability target that the Riksbank specified five years ago.
Last week the Riksdag took the initial decision on the new legislation for the Riksbank that is scheduled to come into force from 1 January 1999. Monetary policy's commitment to low inflation has thus been confirmed still more strongly.
In the five years since the price stability target was first adopted, monetary policy's primary ambition has been to establish the credibility of the target and the new regime.
This has been a long and arduous process. The credibility of the price stability target is a function of public confidence that inflation will stay permanently close to the targeted rate.
A lack of credibility means that the monetary stance has to be more restrictive than would otherwise be necessary—leading to lower resource utilisation—in order to counter inflationary impulses from too high inflation expectations. In that inadequate credibility entails real economic costs, it is essential that inflation expectations are brought into line with the target as quickly as possible.
The amended Riksbank Act will provide valuable support for the work of making people feel sure that inflation will be permanently in the vicinity of the 2 per cent target. This in turn implies better conditions for wage settlements, for example, as well as a situation where market rates can centre on a lower level. A stable exchange rate trend is also promoted.
Another important factor for confidence in the low-inflation policy and thereby for monetary policy's environment is, as we all know, the situation for government finances. This was evident when the deficit was growing, just as it is now that the deficit is disappearing and even turning into a surplus. It underscores the importance of economic policy in general being sustainable in the long run and focused on low inflation. The commitment must also be nurtured, not least now that Sweden will not be adopting the single currency from the start. The demands on economic policy are particularly great in this situation, not least in the light of the discussion about the krona's future trend.
In recent years, confidence in the new low-inflation regime and the 2 per cent target for price stability has been notably enhanced. This made the marked reduction of the instrumental rate feasible in 1996. It also gave better conditions for the current round of wage negotiations, for instance. Another example is the small rent increases.
Interest rates, moreover, are historically low and the bond rate differential with other countries is notably small, though there is still room for a further improvement as confidence in the policy is rooted even more firmly.
Increased need of transparency
The Riksdag's initial decision on new legislation for the Riksbank enables us to start on the work of preparing for the new organisation of the governing board and the directorate. This work includes taking further steps for more transparency. The statutory independence of the Riksbank naturally calls for increased insight into and control of the considerations behind monetary policy decisions.
Transparency makes it plain how policy functions, which helps the efforts to gain public support for the price stability target. Market players can obtain a clear picture of what is influencing the Riksbank. The price stability target is a means for stabilising expectations about monetary policy responses; transparency accordingly helps to make policy effective. Insufficient transparency and clarity make it more difficult for inflation expectations to sight the target and this in turn can obstruct the attainment of high, stable growth.
It is encouraging that the efforts for increased transparency at the Riksbank have produced results. In recent years the implementation of monetary policy has functioned smoothly. People also have a deeper understanding of what monetary policy can and cannot do. It is gratifying that regular surveys commissioned by the Riksbank show very satisfactory popular support for the price stability target; the 2 per cent inflation target is supported by as much as 62 per cent, while only 13 per cent are against it. Broad support is important for monetary policy's perceived legitimacy.
The new legislation accentuates the need of transparency. The process should therefore be carried as far as possible. The Governing Board has, for example, already stated that the voluntary publication of the minutes of the Directorate's meetings should not conflict with the Official Secrets Act and that the matter should be considered once the new Executive Board is functioning smoothly.
Repo rate increased
Now let me turn to the current implementation of monetary policy.
Since October, when I last attended your Committee to discuss monetary policy, the repo rate has been raised from 4.10 to 4.35 per cent, that is, by 0.25 percentage points. This decision last December represents a consistent application of the following rule for monetary policy:
If the overall assessment of inflation—assuming an unchanged instrumental rate—suggests that in a time horizon of twelve to twenty-four months the rate of inflation will be in line with the price stability target of 2 per cent, then the monetary stance is appropriate. If the assessment indicates instead that inflation is liable to exceed the target, then the monetary stance is too expansionary and the repo rate should be increased. Conversely, a risk of inflation being below the target indicates that the stance is too restrictive and the repo rate should be lowered.
The decision was criticised in some quarters. In my opinion, the discussion of monetary policy would benefit if the arguments against the Riksbank's action were spelt out more clearly. Is a different target, above 2 per cent, advocated or does the criticism stem from an inflation assessment that differs from the Riksbank's? These two standpoints are not always clearly distinguished in the public debate.
The Riksbank holds that monetary policy must focus consistently on the target. Otherwise there is liable to be uncertainty, which tends to push inflation expectations up. That, as I mentioned earlier, is something that has to be avoided. Expectations of inflation above the targeted rate lead to unnecessary real economic costs.
The reason for choosing such a long target horizon as twelve to twenty-four months, with the attendant difficulties in constructing reliable forecasts for monetary policy, is that interest rate policy can then be adjusted more gradually than otherwise. With a shorter horizon, which would certainly favour more accurate assessments, meeting the target would call for a sharper impact from monetary measures. Production and employment would then fluctuate more markedly. This risk needs to be born in mind when deciding the target horizon; in doing so, consideration is paid to real factors like production and employment. Partly for these reasons, the Riksbank has decided that the target horizon should normally be twelve to twenty-four months.
In my previous discussion with this Committee I pointed out that the monetary stance must not remain too expansive for too long in an upward phase of activity. In such a situation it should be adjusted in good time so that actual growth moves into line with the economy's long-term trend. Otherwise there is a risk of the economy expanding too rapidly in the short run, giving rise to various bottleneck problems in the labour market or the supply of real capital. It is in such a situation that inflation is liable to accelerate. That means that, sooner or later, the upward growth phase is broken, production slackens and employment falls.
A timely monetary policy realignment accordingly serves, not to break economic growth, but to render it more sustainable and balanced. This is what is meant by a policy for price stability that is perceived as credible. Thus, it is not a question, at least initially, of actually making the monetary stance restrictive but of moving in a less expansionary direction.
The assessment in our Inflation Report last December was that GDP growth would be in the vicinity of 3 per cent in both 1998 and 1999. Increasing demand would lead to rising capacity utilisation and gradually create pressure on the available resources for production. This was expected to generate some inflationary pressure. Given an unchanged monetary stance, CPI inflation was expected to move up by degrees to 2.5 per cent, with a somewhat higher level for underlying inflation. In order to avoid more pronounced interest rate hikes at a later stage, it was considered that the time had come for a realignment in a less expansionary direction. The repo rate increase of 25 basis points in December was a first step.
The latest Inflation Report
Since the repo rate increase in December, the GDP outcome for the fourth quarter has been published. GDP growth between the fourth quarters of 1996 and 1997 was 3.3 per cent, which implies that activity in the Swedish economy rose appreciably towards the end of last year. At the same time, the crisis in Asia has worsened. Although we still envisage a continued economic upswing in Europe in general as well as in Sweden, the current assessment is that the Asian crisis will probably subdue both activity and inflation to a greater extent than expected earlier. To this should be added an outcome of the early wage negotiations that is somewhat more favourable than seemed likely earlier, as well as lower rent increases than expected in the current rent agreements.
In the latest Inflation Report, which the Riksbank presented at the beginning of March, the assessment applies as usual to the coming twenty-four months. It indicates that, with an unchanged monetary stance, annual CPI inflation will be broadly in line with the targeted rate in 1998 and 1999. The underlying rate is expected to be somewhat higher towards the end of the period, which mirrors a continued increase in activity, notwithstanding the crisis in Asia, and gradually rising capacity utilisation. Thus it is considered that the Asian crisis will tend to curb the upward phase but will not break it.
The inflation prospects are based on GDP growth of just over 2.5 per cent this year and about 3 per cent in 1999. With the prevailing expansionary monetary stance on which the conditions for the forecast are based and provided international effects of the Asian crisis are not too protracted, good growth is also foreseen in 2000.
Employment is expected to rise as demand becomes stronger. Signs of this are already discernible in the statistics. Together with various educational programmes, this indicates some further reduction of unemployment.
If this main scenario materialises, economic development in Sweden in the coming two years should be fairly favourable, with good growth, some increase in employment and moderate inflationary pressure. The crisis in Asia, as assessed for this main scenario, can be said to result in some deferment of the ongoing monetary policy realignment.
Future monetary policy
The information that has become available since the end of 1997 points on the whole to the prospect of somewhat more subdued inflation. The overall assessment of inflation normally incorporates not just one scenario but a set of assessments involving different types of risks.
The altered spectrum of risks that we now face is therefore also important for the construction of monetary policy. There is still a risk of the Asian crisis becoming more profound, so that inflation will be lower than now seems likely. The repercussions on economic activity, internationally as well as in Sweden, could then be more extensive and protracted. At the same time, the risk has receded of increased inflation generated by high wage increases and weak productivity growth.
Another factor in the Riksbank's assessments is the exchange rate trend. In recent months the krona has tended to weaken. If this situation were to continue, it could affect the inflation prospects and consequently the construction of monetary policy.
Briefly, then, the inflation prospects as assessed by the Riksbank in the latest Inflation Report point to a tendency that is somewhat more subdued than was expected at the end of 1997. But the main scenario does involve rising activity, with a gradual increase in capacity utilisation and thereby some accentuation of inflationary pressure. This would call for a cautious tightening of the monetary stance in the year ahead. At the same time, the crisis in Asia means that inflation may be lower. Moreover, the exchange rate tendency has added to the uncertainty about future inflation.
All in all, in the Inflation Report the Riksbank concluded that the repo rate should not be altered at present. With the uncertainties in the assessment, however, there are strong reasons for appraising the construction of monetary policy continuously as new information become available.
Since the presentation of the Inflation Report, some new statistics have been produced. They do not motivate an altered conclusion about the monetary stance.
