Secretary General Arne Øien

Norwegian Monetary Policy

Presentation at seminar hosted by Pareto Fonds AS, August 31, 1995

Ladies and Gentlemen,

I thank you for the opportunity to address you today on the subject of Norwegian monetary policy.

Before I discuss the present monetary policy framework in Norway, it will probably be helpful to give a brief account of our monetary policy history.

1 Some historical lessons

Although institutional settings have been different, I think it is fair to say that the Norwegian experiences in the field of monetary policy over the last 15-20 years are broadly in line with the experiences of many other European countries. From what has happened in European monetary policy and in the currency markets the last few years, I think we can draw two main conclusions:

The first is that, by and large, the policies of stable exchange rates have been fairly successful in bringing about greater nominal convergence. Within the ERM, both interest rate differentials and inflation differentials are considerably lower than in the mid 1980s. In our own country, we had an interest rate differential of around 8 per cent vis-a-vis the European average after our last major devaluation in 1986. We now have interest rates on or even somewhat below the European average. In terms of reducing inflation, the Norwegian record has been quite good the last few years. The emphasis placed on a stable exchange rate has contributed significantly to this.

The second lesson is that a system of fixed exchange rates with narrow fluctuation margins is vulnerable to speculative attacks, even when the exchange rate is at or very close to its "fundamental" equilibrium rate. In response to the turmoil in the European currency markets in 1992, Norway has tried to counteract this by pursuing a policy of stable exchange rates without a specific "target value" and without intervention limits. The countries of the ERM widened the intervention margins to +/- 15 per cent when their currencies came under speculative attack in 1993. By taking away the possibility of one-sided bets offered by fixed rates and narrow fluctuation margins, one has aimed at reducing the incentives for speculative attacks. But the fundamental policy choice was the same in both cases: The day to day conduct of monetary policy was - and continues to be - built on a commitment to stability in exchange rates.

The emphasis on stable exchange rates in Norwegian monetary policy today partly reflects the unsatisfactory results which followed from Norway's monetary policy in the late 1970s and early 1980s. In that period, monetary policy in practice was directed at maintaining, or improving, the international competitiveness of Norwegian industries. The Norwegian krone was devalued on several occasions, often following periods of overheating and strong wage inflation. Our experience, which I believe is shared by many other countries, suggests that this approach to monetary policy very quickly can lead into a cycle where currency depreciation fuels inflation through its effects on the price level of imported goods which in turn leads to renewed pressure for currency depreciation and through its effects on expectations.

In the 1980s, with increased international capital mobility, another negative effect of recurrent devaluations became apparent. Agents in the financial markets demanded higher interest rates on assets denominated in Norwegian kroner than on assets denominated in other currencies. The resulting high level of interest rates discouraged investments and deepened the recession in Norway in the late 1980s. Only after a prolonged period of stable exchange rates were we able to reduce interest rates to the level seen in other European countries. At that time, of course, many other European countries had already made progress in terms of breaking out of the inflation/devaluation cycle through a policy of stable exchange rates.

There has been some discussion on the stance of monetary policy in Norway in the years 1988-92. In particular, it has been suggested that a more expansionary monetary policy in those years might have dampened the recession and reduced the extent of the banking crisis through lower interest rates. Since interest rates in those years were set at a level consistent with a stable exchange rate, lower interest rates could only have been achieved at the cost of giving up the fixed exchange rate policy. This would have made it very difficult to achieve the moderate wage increases we in fact had in 1988 and 1989. Higher wage growth combined with a depreciating currency would seriously have damaged the confidence in the krone at a time when we had just reduced the interest rate differential vis-a-vis the European currencies to around zero from around 8 per cent after the 1986 devaluation. This would have implied higher interest rates. It is thus not at all clear that a devaluation in 1988 or 1989 would have dampened the recession or reduced the extent of the banking crisis. On the contrary, I believe that such a move would have been counterproductive because it would have led us back into the viscious circle we were just about to break out of.

2 Monetary policy in Norway today

After the turmoil in the European currency markets in late 1992 forced Norway to abandon the fixed rate vis-a-vis the ECU, Norwegian authorities had to draw up new guidelines for monetary policy. On 6 May 1994 the Government issued a regulation by Royal Decree for the Norwegian krone exchange rate system. This regulation represented a more precise formulation and continuation of the practise followed since the krone was floated in 1992. The operative guideline for the conduct of monetary policy by Norges Bank is still to maintain a stable exchange rate against European currencies, at about the current level. Norges Bank will not use monetary policy instruments to the same extent to maintain the exchange rate as under a fixed exchange rate regime. However, if significant changes in the exchange rate occur, the instruments will be oriented with the aim of bringing the exchange rate back to its initial range. No fluctuation margins have been established for the krone.

It is important to realize that the floating of the krone in 1992 did not represent a departure from the considerations that have governed Norwegian monetary policy since 1986. Our fundamental policy choice is to use the nominal exchange rate as an anchor for monetary policy, and the guidelines for monetary policy that we now have, represent a continuation of the policy pursued since 1986. However, the formal setting of monetary policy had to be changed to reduce the incentives for speculative attacks inherent in the previous fixed rate regime.

Of course, monetary policy has to be seen in a broader economic policy context. The main objectives of economic policy, as set out in the Government's Long-Term Programme, are to secure a solid basis for full employment and sustainable economic growth. When the new framework for monetary policy was set out in the Revised National Budget of 1994, the Government emphasised that maintaining low price and wage inflation is a necessary requirement for achieving this overall objective. The Government concurred with Norges Bank that there is no trade-off between higher inflation and higher employment in the long run. To the extent that a correlation exists between these two variables in the long run, there is evidence indicating that the opposite is true - excessive inflation destroys the basis for economic growth and higher employment. In a small, open economy, changes in the exchange rate have a considerable influence on price developments. The operational target for monetary policy is therefore to ensure a stable exchange rate in order to achieve low price and wage inflation. High employment and low price and wage inflation require a successful coordination of monetary policy, fiscal policy and structural policies. Maintaining low price and wage inflation, in line with or below that of our trading partners, is a necessary condition for achieving this.

Thus, we have a hierarchy of goals: Our main goal is to secure a solid basis for economic growth and high employment. Low inflation is a precondition for achieving this in the long run. With low inflation in other European countries, the operational target of a stable exchange rate will contribute to low inflation.

Why did we decide to stick to our policy of emphasizing a stable exchange rate when redesigning the framework for monetary policy? Why didn't we opt for an explicit low-inflation target, as countries like Sweden, Finland, New Zealand and the UK have done in recent years?

One simple - but important - reason is that the policy of a stable exchange rate is a policy that has served us well in the last years. It seems to be widely understood and accepted by the social partners and it seems to have a high degree of credibility in the financial markets. It has also contributed significantly to bringing inflation and interest rates down to levels which are are low both by historical Norwegian and present European standards. It would then seem that we would have little to gain and potentially a lot to lose by making major changes to the monetary policy framework. Some of the countries that have switched to inflation targeting in the last few years may have had greater problems with monetary policy credibility. The trade-off between risks and potential rewards may thus have been different for these countries. While the countries that have switched to an explicit inflation target have received much attention for their changes, one should not forget that the nominal exchange rate is the anchor for monetary policy in all ERM countries except Germany, whose stable currency is that fixed point to which the other countries have tied their currencies. So Norwegian monetary policy is still in line with that of major European countries.

A more elaborate answer to the question above would have to be based on an analysis of the optimal nominal anchor for monetary policy in a small open economy like the Norwegian. Time does not permit a full account of such an analysis, but let me briefly outline some of the considerations we made.

Firstly, it is worth noting that the nominal exchange rate usually has been the recommended nominal anchor for monetary policy in small countries with open economies. This is not surprising given the large impact on the economy exchange rate movements have in such countries. Even if a small country with an open economy were to pursue a discretionary monetary policy aimed more directly at achieving low inflation, exchange rate developments would necessarily be a very important part of the set of indicators used to evaluate the appropriate stance of monetary policy. However, given the tendency for exchange rates to overshoot their fundamental equilibrium values, the influence of monetary policy on aggregate demand could be erratic and difficult to control.

A second line of reasoning takes as a starting point the experiences we have in the Scandinavian countries with cost increases in the non-tradable sector. In our economies, this has been a major source of inflationary pressure. With an exchange rate target, cost increases that spread over to the tradable sector translate one-to-one into loss of competitiveness and reduced employment there. With an inflation target, however, the tradable sector is hit even harder. Not only will it lose competitiveness when costs increase, it will also lose competitiveness as the Central Bank tightens monetary policy to neutralize inflationary pressures and the krone appreciates. In this way, the tradable sector will be subjected to a "double squeeze" which may lead to external imbalances. Of course, it is true that a consistent and credible monetary policy aimed directly at low inflation over time may change behaviour in the non-tradeable sector, so as to dampen the tendency to cost increases there. But since we cannot know in advance how fast and to what extent this will happen, this could turn out to be a very costly policy experiment.

This last example highlights another important point: In the long run, it may be very costly - or even impossible - to use a tight monetary policy to compensate for lack of discipline in other policy areas. Therefore, when the Royal Decree on the exchange rate system was presented last year, the Government stressed that all elements of economic policy must be consistent with low inflation in order to lay a sustainable foundation for high employment.

Let us, as an example, look more closely at the relationship between monetary and fiscal policy. Only in the simplest textbook models are monetary and fiscal policy two separate "handles" with which the authorities can influence - or even determine - aggregate demand. In reality, monetary and fiscal policy are tied together through the government's long-term budget constraint; Public sector deficits can only be covered in two ways - by borrowing money or by printing it. If government debt reaches a certain level, it will be tempting - and in extreme cases necessary - to resort to printing money. The inflation associated with this will lower the real value of outstanding government debt. The implication is that credibility in monetary policy also to some extent depends on the credibility of fiscal policy. Where budgetary developments are out of hand, the credibility of a low-inflation target will be low.

In my view, the prospects for continued low inflation in Norway is significantly strengthened by the fact that there is a strong and growing consensus among politicians and social partners that all parts of economic policy must be consistent with low price and wage growth if we are to lay the foundation for sustained growth and a high level of employment.

I am sometimes asked what we will do if we see signs of increasing inflation. Isn't it possible, people ask, that you will reach a point where you will have to make a choice between low inflation and a stable exchange rate, and that there is an inherent contradiction in your exchange rate regime? There is no inherent contradiction here as long as the countries whose currencies we stabilize the krone against have low inflation. The main point is that a stable exchange rate against low-inflation currencies will function as a norm for domestic inflation expectations. In a system where the nominal exchange rate is the anchor of monetary policy, the scope for using monetary policy actively through the business cycle is very limited. Fiscal policy must bear the main burden of stabilization. This is a standard feature of all systems building on stability in nominal exchange rates, and it is something we took account of when designing the framework for monetary policy we now have.

When emerging from a recession, this requires strict fiscal discipline in order to reverse previous increases in expenditure. It also requires that central government has sound finances when entering a recession. Otherwise, efforts to dampen the business cycle through countercyclical fiscal policy may soon lead to an unacceptable debt level. With a high debt level and/or high structural deficits, the scope for fiscal stimulation is small. In such circumstances, the pressure to use monetary policy instruments more actively may increase to a point where the strain on monetary policy becomes too high to bear. The events in the European currency markets in 1992 and 1993 are a reminder of this. In the case of Norway, there is now a rapid fiscal consolidation following the loosening of fiscal policy in the last recession. Although fiscal policy measures were used actively to stabilize aggregate demand, the general government deficit did never exceed the 3 pct. reference value set out in the Maastricht treaty. The Norwegian central government is one of very few to be in a net asset position, and general government is running a surplus estimated at 1,5 pct. of GDP in 1995. By international standards then, fiscal consolidation is progressing from an advantageous starting position. However, Norway - like most other OECD countries - will over the next decades face considerable challenges connected with the ageing of the population and the financing of future pension liabilities. So although we have a good starting position, there is no reason for complacency.

In the most recent Revised National Budget, presented in May this year, considerable attention was devoted to the subject of the Petroleum Fund. Although the fund formally was established a few years ago, there has not yet been built up any capital in the fund due to deficits in the Government's fiscal budget the last few years. This will hopefully change in 1996. Therefore, the Petroleum Fund regulations were reviewed in the Revised National Budget. Time does not permit a full account of the Petroleum Fund, but I would like to comment very briefly on the consequences of acccumulating assets in the fund for monetary policy.

The allocation of capital to the fund will support the conduct of monetary policy along present lines. The fund's capital will be invested abroad, and this will neutralize the expantionary effects on the economy that could have occured if all of our income from the petroleum sector were to be spent as it is earned. It has been claimed that the establishment of a fund will lead to an upward pressure on the krone. But establishing a fund simply means that assets in the form of petroleum wealth is exchanged for financial assets. The net asset position of the country - or the Government - does not change as a result of this. It is only the composition of assets that changes. This should by itself not lead to upward pressure on the exchange rate. On the contrary, the establishment of a fund signals a commitment to avoid a too rapid infusion of petroleum revenue into the domestic economy which would tend to be accompanied by a real exchange rate appreciation.

Let me say a few words on the issue of credibility. There is a clear link between credibility and observability in the sense that it other things equal is more dificult to establish credibility for a system where the link between monetary policy instruments and monetary policy objectives is complicated and where there is a long time lag from the implementation of policy till results are observed. It may be necessary for the central banks in the countries which have recently switched to a more discretionary monetary policy aimed directly at low inflation to show good results through at least a couple of full business cycles before any solid judgement on their performance can be made. The nominal exchange rate target, on the other hand, is readily observable, and anyone can see almost from minute to minute whether the monetary authorities are sticking to their announced policies. One implication of this may be that a Central Bank in a country which has recently switched to an inflation target may be inclined to pursue an overly tight monetary policy in the first years after the policy change in order to bulid credibility more rapidly. If this is the case, the change in policy may have considerable real economic costs.

3 Conclusion

Only time will tell whether our present monetary policy framework is as robust as we would like it to be. Hopefully, we will have learned sufficiently from our historical experiences to stick to a monetary policy that together with other policy elements will ensure low inflation and thus lay the foundation for sustained economic growth and high employment.

An important lesson from the past is to avoid using competitiveness or other real variables as targets for monetary policy. A competitiveness target for monetary policy will only serve to accomodate price and wage increases and may fuel a viscious circle of inflation and devaluations. It is necessary to have a clear nominal target for monetary policy. For the reasons I have discussed above, my conclusion is that a nominal exchange rate target is the best option for monetary policy in Norway.

---

Lagt inn 4 september 1995 av Statens forvaltningstjeneste, ODIN-redaksjonen