The current situation for monetary policy
First a word of thanks to the Committee for the invitation to discuss the current situation for monetary policy.
A good deal has happened in the global economy since our previous discussion with the Committee last March. The crisis in the emerging markets in Asia, regarded initially as a regional concern, has continued to spread, first to emerging markets in other parts of the world and then, mainly via financial markets, to the industrialised countries. This has altered the economic climate.
1. The Swedish economy in perspective
Five years with good growth and low inflation
Let us begin, however, by shifting our attention from recent events and noting that since the crisis in the early 1990s, economic development in Sweden has been rather good. Relatively high growth has been combined with a rate of inflation that has remained low.
Since the summer of 1993, total output (GDP) has been rising at an average annual rate of almost 3 per cent. That more or less matches the growth rate during the upswing in the 1980s (Fig. 1). It is markedly higher, however, than the average annual rate of 2 per cent which was achieved in the two decades from 1970 and 1990; it also exceeds what can be regarded as the sustainable long-term rate.
One important reason why inflation could still be kept at a low level in recent years is that in this period the economy has had a reserve of unutilised capacity. That, in turn, is a consequence of the deep recession that preceded the upswing, plus the fact that the growth of capacity has been high. Moreover, inflation expectations have gradually stabilised around the 2 per cent inflation target (Fig. 2). Recently, however, and in certain earlier periods, transitory price movements have brought the average annual change in the consumer price index (CPI) just outside the inflation target's lower tolerance interval.
Large transitory price effects
The nature of transitory price effects and the appropriate monetary policy response were matters that I discussed in detail with the Committee last March. So now I shall simply reiterate the Riksbank's view that as long as more long-term inflation expectations are unaffected, transitory price movements should be a subordinate consideration in monetary policy. A contrary opinion would necessitate drastic monetary policy measures that might have destabilising effects on output and employment, which is clearly undesirable.
In theory, a CPI change figure that is higher than the current 0.4 per cent could of course be engineered simply by raising the interest rate. That would fuel CPI inflation via higher house mortgage rates. It need hardly be said that that is not the way to conduct monetary policy.
It should be borne in mind that a monetary policy target formulated in terms of the change in the CPI has major advantages. The CPI is a broad and well-known measure of inflation. But the construction of the CPI does pose problems, not least for this Committee, when it comes to assessing monetary policy's past performance. This will be still more important if the Riksbank is accorded the independent status which the Riksdag (Sweden's parliament) will shortly be voting on. The use of a target variable, such as the CPI, which necessitates frequent reference to factors that cause temporary deviations, is not an ideal solution.
There are grounds for supposing, however, that sizeable transient price movements of the type we have experienced recently are not commonplace. They derive predominantly from the marked improvement in recent years in economic policy's credibility. There have been few periods in the past when, for example, house mortgage rates have dropped from about 14 per cent to the present level around 5 per cent. So the problems with targeting CPI inflation should not be exaggerated. In order to facilitate the assessment of monetary policy, the Riksbank has arranged for Statistics Sweden to supplement its CPI statistics with figures for various indicators of underlying inflation.
Another matter to bear in mind is the lag before monetary policy measures affect the economy. If the rate of inflation happens to move outside the tolerance interval in one month, the central bank is not capable of bringing inflation directly back on target in the following month. That is why the central bank needs to present an inflation forecast, which has two main advantages. One is that the central bank can be explicit about the time horizon at which inflation is to be returned to the target. The other is that people in general can judge for themselves whether monetary policy can be considered to be properly balanced. If the forecast deviates from the inflation target at the time horizon at which monetary policy exerts its main effect—twelve to twenty-four months—then the instrumental rate has to be adjusted.
Jobless growth
Although Sweden has experienced five years of good economic growth, there has been no corresponding reduction of unemployment. The picture since 1993 speaks for itself: a clear upturn in the output curve, while the low level of employment is still virtually unchanged (Fig. 3). This suggests that neither low demand nor a poor growth potential are primarily responsible for the weak development of employment.
Part of the problem with employment may be that much of the growth in the past five years seems to have occurred in sectors that are capital-intensive. More balanced growth in the future would presumably result in stronger demand for labour. But the weak employment trend probably also stems from more fundamental factors. For several decades now, the general structure of business activities in Sweden has tended to move towards a growing input of capital at the expense of labour.
The fact that certain countries have achieved a sustained improvement in employment is often explained not so much by high growth as by a production process that has been generally more labour-intensive. If an economy is characterised by the combination of high, stable growth and weak employment, this points to the existence of structural problems. It should be added that Sweden shares the problem of jobless growth with other countries in Europe, though not with all.
2. Monetary policy to date
Warning lights a year ago
A year ago there were many indications that economic growth in Sweden would continue at a high rate. It also looked as though capacity utilisation would rise in the coming years and lead to some increase in inflation. The crisis in Asia was expected to ease inflationary pressure to some extent but not sufficiently.
As most of the warning lights were thus blinking red, the Riksbank adjusted its policy stance; in December 1997 the repo rate was raised from 4.10 to 4.35 per cent.
In my statement to this Committee in October 1997, I described the planned policy realignment as a movement from an expansionary to a more neutral monetary stance. It amounted to keeping the economy moving at the rate which is structurally feasible, not to arresting its movement. A gradual monetary policy adjustment would bring the path of growth into line with the sustainable long-term trend, thereby protecting the economy from capacity shortages and inflation. A small interest rate increase at an early stage was intended to prevent problems from arising later on and requiring larger interest hikes to bring inflation back to the targeted rate. In this way, both the price level and growth would be more stable.
In the December 1997 Inflation Report, however, we included a risk scenario that pointed to lower inflation compared with the main scenario. We noted: “As regards inflation, however, a substantial impact from the economic crisis in Asia does presuppose that the situation deteriorates so much that international activity is seriously weakened.”
As things turned out, the effects of the Asian crisis did exceed our assumption for the main scenario. During the first half of this year the main assessment was revised successively downwards. As a result, the monetary policy adjustment did not continue as envisaged earlier. In view of the risks of a weaker trend, the repo rate was left unchanged during the spring. In connection with the June Inflation Report, the Riksbank concluded that inflation prospects had changed; our inflation forecast, based on an unchanged repo rate, pointed to a rate of price increases twelve to twenty-four months ahead that was below the 2 per cent target. In accordance with the Riksbank's procedure, at the beginning of the summer the repo rate was therefore lowered from 4.35 to 4.10 per cent.
Increased financial market unrest and lower international growth
In the September Inflation Report the Riksbank noted that the real economic situation was still relatively bright. In our main scenario, growth was expected to remain stable while inflation slowly moved up towards the target in the coming one to two years. However, this path for inflation did start from a lower level.
The risk of slower international growth on account of the Asian crisis had grown but the picture was also complicated by the marked weakening of the krona, which could have a contrary effect on inflation. In the current exchange rate regime, the Riksbank does not target the level of the exchange rate but this is still one of the factors that has to be weighed into the assessment of future inflation.
In the context of monetary policy, the most pertinent aspect in recent years has proved to be the extent to which the exchange rate affects inflation via its impact on production, primarily in the export sector. The effects that a weaker exchange rate may have on import prices should be a secondary consideration, along with other transitory price movements, as long as more long-term inflation expectations are unaffected.
During September and part of October—before and immediately after the publication of the Inflation Report—the monetary policy environment was turbulent and difficult to assess. Neither was the uncertainty confined to the international scene. It also came from the general election in Sweden and the future direction of fiscal policy. All in all, this prompted the Riksbank to keep the repo rate unchanged. The possibility of a lower rate would be considered when inflation prospects were clearer.
During the autumn the financial unrest continued to grow. The Russian Government's suspension of payments in mid August and the problems that emerged in late September with the U.S. hedge fund Long Term Capital Management contributed to the much more nervous atmosphere. Share prices dropped around the world and the financial crisis went on spreading to other emerging markets that had previously been spared. There was a risk that the falling stock markets and the wider extent of the crisis would elicit effects via decreased export demand and more cautious behaviour among consumers and firms.
At the same time, the problems with Long Term Capital Management aroused fears of a considerably more serious risk that had not attracted so much attention earlier—the risk of banks becoming so cautious with credit that this, by itself, would paralyse the global economy. When the banks are bent on reducing exposures and slimming balance sheets in order to avoid loan losses, borrowers have less access to capital. If risk aversion becomes unduly pronounced, even sound borrowers are liable to be hit, with serious effects on the total level of activity in the global economy. One sign of an imminent credit squeeze was the growing spread in market interest rates between the soundest borrowers and others.
In this way it was envisaged that the financial crisis would have negative effects on the real economy. Forecasters altered their assessments and the global prospects for growth and inflation were revised downwards. The Riksbank likewise adjusted its assessment of developments in the rest of the world.
3. The current monetary stance
Repo rate cut on 3rd November
The assessment that growth in the rest of the world will be weaker than we counted on earlier led the Riksbank to conclude that with an unchanged repo rate, in the coming twelve to twenty-four months the rate of inflation would be below the target. Resource utilisation would be lower and inflationary pressure would decrease. Signs of this were discernible in recent statistics: net exports had weakened, stocks were accumulating and corporate borrowing was rising somewhat more slowly. The financial unrest had also had some negative impact on household confidence, which could mean that consumption rises more slowly than expected earlier. The effects on inflation from the slowing tendencies were judged to be greater than the stimulus from the weaker exchange rate. Moreover, the financial market unrest had tended to subside. Under these circumstances, on 3rd November the Riksbank lowered the repo rate 25 basis points to 3.85 per cent.
It should be noted that the Riksbank's downward revision of demand and inflation prospects in the Swedish economy is not dramatic. In the September Inflation Report the path for inflation was already below 2 per cent for much of the forecast period. But the revision is of an order that affects conditions for the Riksbank's operations.
In order to provide more room for manoeuvre in the formation of monetary policy, yesterday the Governing Board decided to lower the corridor surrounding the repo rate: the deposit rate was reduced by 0.5 percentage points to 3.25 per cent and the lending rate likewise 0.5 percentage points to 4.75 per cent.
The concern about an involuntary contraction of credit—and the serious consequences it might have—has also prompted the central banks to heighten surveillance of the financial system. For the Riksbank this concerns contacts with the major banks as well as increased oversight of bank financing in cooperation with the Financial Supervisory Authority.
On the whole, the Swedish banking system appears to be stable. This conclusion is elaborated in the year's second Financial Market Report, which we published yesterday. Lending to the most problem-burdened countries is relatively low. Moreover, the banks have built up liquidity reserves as a precaution against a worsening of the situation in the international markets. The supply of credit in the Swedish market does not appear to have suffered to date. Neither has the Riksbank seen any indications that the Swedish banks intend to curtail their lending to domestic sectors.
Recently the global financial unrest seems to have become somewhat less pronounced. It is difficult to tell whether this could be a sign of the calm associated with being in a hurricane's eye—whether the international financial turbulence will return or whether the cyclone is receding. The Riksbank is therefore continuing to analyse the banking system with heightened readiness until the situation has become more stable.
The course of events in the international arena has warranted measures of a short-term nature in order to safeguard the functioning of the global financial system. Interest rates have been reduced recently in a number of countries, for example. In the longer term, work must focus on the construction of international rules and systems that limit the risk of—and preferably prevent—crises of this type from happening at all. What has happened reveals clear shortcomings in how the global financial system functions. But it should also be underscored that “free trade” in the financial domain can very probably confer large welfare gains, just as free trade does in the case of goods and services.
The conditions for forecasting in general and for inflation-targeting monetary policy in particular have been illustrated by what has happened in recent years. A set of indicators has to be relied on in an endeavour to form a definite opinion about future developments. Central banks are unfortunately not in a position to wait for a complete picture before acting. It takes time for our actions to elicit their full effect; when all the pieces of the puzzle are finally available, the time for action has passed. The continuous fulfilment of the monetary policy target would then probably require more drastic measures; such a near-sighted policy would lead to greater economic instability.
The course of the Asian crisis and its contagious effects in the industrialised countries is having a larger impact than we counted on earlier. I cannot rule out the possibility that in order to fulfil the inflation target, further adjustments of the monetary stance will be called for in the same direction as recently. A weaker development in the rest of the world also affects conditions in Sweden.
