III. The Choice of a Nominal Anchor for Norway9
54. Norway has traditionally used an exchange rate target as a nominal anchor, to help guide inflation expectations. An alternative nominal anchor used in a number of other resource-based industrial countries is an explicit inflation target. Because both nominal anchors have advantages and disadvantages, there continues to be an active policy debate in Norway over the choice of the monetary policy framework. This paper examines the pros and cons of these regimes in a Norwegian context.
A. Exchange Rate Targeting
55. Norway has a long history of exchange rate targeting which goes back to the silver standard in the mid-1800s, when the monetary unit was linked to silver at par value.10 In the 1870s a gold standard was established, under which Norges Bank exchanged krone for gold at a fixed rate. Following a short period with a floating exchange rate, the gold standard was abandoned by Norway in 1931 and the currency was pegged to the U.S. dollar and pound sterling. After World War II, Norway participated in the Bretton Woods agreement and, following its collapse, the European currency “snake.” When the European Exchange Rate Mechanism was established in 1978, Norway chose to remain on the sidelines and link the krone to a trade-weighted basket of currencies (later to the ECU).
56. Since December 1992 Norway has operated a managed float exchange rate regime, in which Norges Bank seeks to maintain a stable krone exchange rate. Although the explicit wording of the monetary policy guidelines relates to currency stability against “European currencies,” Norges Bank has generally behaved as if it was targeting the krone/ECU exchange rate with an implicit target range of 103-105 on the inverted ECU index,11 In May 1998, the government announced that Norges Bank would continue maintaining a stable krone exchange rate against European currencies when the euro is introduced on January 1, 1999. It is expected that the euro will replace the ECU as the implicit target. This will simplify the transition process because these two composite currencies will be equivalent on the final trading day in 1998.12
57. Exchange rate targeting has several advantages, including linking the inflation rate for internationally traded goods and inflation expectations to the inflation rate in the anchor country. It also avoids the time inconsistency problem, which can arise when a monetary authority pursues short-run growth and employment gains at the expense of higher inflation and lower growth in the long run. However, exchange rate targeting is not without its drawbacks. These include the loss of an independent monetary policy and the quick transmission of shocks from other countries (particularly terms of trade shocks), which can adversely affect the domestic economy. Moreover, by providing advance information on the policy reaction function, exchange rate targets can make countries more vulnerable to speculative attacks on their currencies, such as the European exchange rate crisis of September 1992.
58. Between December 1992 and August 1998, interest rate movements in Norway were largely determined by interest rate movements in Germany. With interest rates in Germany set in accordance with its own economic conditions (which often differ considerably from those in Norway), the Norwegian economy until recently has been faced with interest rates which are incompatible with its cyclical position. This became most apparent during 1997, when Norwegian interest rates were lowered significantly even though the economy was experiencing excess demand pressures. This situation changed dramatically in August 1998, when the Norwegian krone came under strong downward pressure in the exchange market and Norwegian short-term interest rates were raised by 425 basis points.
59. The problems raised by the procyclicality of monetary policy in Norway in the recent past led to a number of proposals for altering the monetary framework. One proposal that would have maintained a considerable degree of continuity in the monetary strategy was to adopt a broader exchange rate indicator, based on a weighted average of the currencies of all of Norway’s major trading partners. Historically, the difference between the movements of the ECU and a trade-weighted index has been small and the corresponding short-term interest rates have also moved closely together (Figure 5, Panels 1 and 2). In fact, in recent years the ECU short-term interest rate has been above the trade-weighted interest rate, because the very low interest rates prevailing in Japan have more than offset higher interest rates in the United States. Therefore, it is unclear that shifting to a broader exchange rate index would have much effect on Norwegian monetary policy.
FIGURE 5NORWAY: INTEREST RATE AND EXCHANGE RATE DEVELOPMENTS
Sources: IMF, International Financial Statistics.
60. A case for abandoning the exchange rate target completely has been made by those who believe that the strong commodity base of Norway’s trade (including the dependence on oil exports) makes the exchange rate highly sensitive to movements in the terms of trade, and therefore difficult to control. Among industrial countries, Canada, Norway, and Australia have the highest ratios of raw material exports relative to total exports, at 31, 37, and 58 percent respectively. Substantial falls in world market prices for raw materials, as experienced in 1998, result in large real income declines in these countries. Under these circumstances, insufficient immediate adjustment in the demand for goods and services would lead to a sizeable deterioration in the current account balances of these countries, which eventually results in a fall in demand for their currencies. These currency depreciations ultimately help to restore external balance by lowering the demand for imports.
61. The empirical link between changes in the terms of trade and movements in the real exchange rate is well documented in the literature. For example, Amano and Van Norden (1995) find that the ratio of the price of commodity exports to manufactured imports explains most of the variation in the real exchange rate in Canada and Gruen and Wilkinson (1994) find similar results for the Australian dollar using the deflators for goods and services as the measure of the terms of trade. More recently, Hansen (1997) has found similar results for the New Zealand dollar. Concerning Norway, one of the background papers for last year’s consultation found that the present value of petroleum wealth had significant explanatory power for movements in the Norwegian krone exchange rate. In particular, the paper suggested that a 1 percent increase in the present value of petroleum wealth in relation to GDP (comparable to the effect of a 1 percent increase in oil prices) was associated with a 1½ percent appreciation of the real exchange rate.
62. Owing to the large decline in oil and non-oil commodity prices which began in the second quarter of 1997, the current account balances of the major industrial commodity exporters have deteriorated considerably, with Canada and Australia projected to record deteriorations in their current account positions of ½ and 2 percent of GDP respectively in 1998 (Figure 6, Panels 1 and 2). Although New Zealand’s current account position is projected to improve slightly in 1998, it has deteriorated by 3 percent of GDP since 1996. Norway’s current account surplus declined by an estimated 5 percent of GDP in 1998. In response to the sharp declines in commodity prices and the deterioration in current account positions, the exchange rates of the commodity exporters have also depreciated. In particular, the Australian and New Zealand currencies fell by 15-20 percent between the second quarter of 1997 and the second quarter of 1998, while the Canadian dollar and the Norwegian krone have each depreciated by about 10 percent against the U.S. dollar and ECU respectively since the beginning of 1998 (Figure 6, Panel 3).
FIGURE 6NORWAY: COMMODITY PRICES, CURRENT ACCOUNTS AND EXCHANGE RATES
Sources: IMF, International Financial Statistics; and OECD.
63. The imminent onset of the third stage of EMU, at the beginning of 1999, is widely considered to have become a source of increased currency volatility for Norway, because of its small size and close links to the economies of EMU participants. This event, combined with continued uncertainties regarding the outlook for petroleum and other commodity prices, makes it difficult for Norway to maintain the present exchange rate targeting framework.
64. There has been considerable public debate in recent months about the possibility that Norway could increase the credibility and sustainability of its exchange rate target by adopting an explicit target band in terms of the euro, under a formal arrangement involving the potential for liquidity support from the European Central Bank. For better or worse, such a policy would accentuate the rigidities in interest rate policy experienced under the current regime. In particular, since Norway’s terms of trade and cyclical position frequently diverge from those of the countries participating in the EMU, interest rate policy could be expected to be constrained on many occasions from responding to domestic economic conditions.13 In any event, it appears unlikely that the ECB would be willing to provide liquidity to Norway to assist in defending the krone/euro exchange rate, unless Norway became a member of the European Union. With limited political interest in this proposition at present, this possibility could only become operative over the medium term.
B. Inflation Targeting
65. In contrast to Norway, some other resource-based industrial countries have adopted inflation targeting as their anchor for monetary policy (Canada, Australia, New Zealand; nonresource-based advanced economies with an inflation target include Sweden and the United Kingdom). The policy of inflation targeting involves several elements: (1) an institutional commitment to price stability as the primary long-run goal of monetary policy; (2) official announcements of a quantitative medium-term target for inflation; (3) increased transparency of the monetary policy strategy through communication with the public; and (4) sufficient operational independence for the central bank to give credibility that the inflation target will guide monetary policy even when the short-term costs for real activity are apparent; and (4) accountability of the central bank for attaining the inflation objective.14
66. Inflation targeting, like exchange rate targeting, has the advantages that it is highly transparent and easily understood by the public; and that the increased accountability of the central bank helps to avoid the time inconsistency trap of pursuing an overly expansionary monetary policy at the expense of a deterioration of long-term economic prospects. In contrast with an exchange rate target, inflation targeting enables monetary policy to focus on domestic considerations and to respond to shocks to the domestic economy.15
67. A potential drawback of inflation targeting for Norway is that the initial effects of a tightening of monetary policy may be felt more strongly through a strengthening of the exchange rate, tending to depress non-oil net exports, than through changes in domestic demand. Given the importance of oil in the Norwegian economy and the need to maintain the competitiveness of the non-oil sector, the potential of exacerbating exchange rate movements through a more active use of monetary policy has been described as a serious drawback of inflation targeting. Researchers at Statistics Norway have calculated that an appreciated exchange rate provides the initial contractionary effect of a tighter monetary policy stance by dampening net exports and that the effects of higher interest rates on domestic demand occur more gradually over time. They estimate that a 2 percentage point increase in interest rates and a 4 percentage point appreciation of the exchange rate over two years would result in a cumulative output loss of over 2 percent of GDP, comparable to the output loss in other countries in response to a tightening of monetary policy.16 An appreciated exchange rate would lead to an output contraction of ½ percentage point in the first two years with higher interest rates contracting output by an additional ½ percentage point in the second and third years. It could be argued, however, that a counter-cyclical tightening of monetary conditions would not compromise the competitiveness of the non-oil sector over the long-run, because a relaxation of monetary policy in a downswing would have the opposite effects and cancel out over the cycle. This would be the case if the effects of monetary policy on the economy were symmetrical.
FIGURE 7NORWAY: OUTPUT GAP
Source: Staff estimates.
68. Some analysts argue, however, that the effects of exchange rate movements are not symmetrical because investments in a country incur fixed costs. According to the option argument of Dixit (1989), firms enter and invest in a country when its real exchange rate is undervalued and develop valuable intangible assets specific to the location. If the exchange rate should subsequently appreciate, foreign firms will not exit at the same exchange rate at which they entered because of the presence of large fixed costs. Expectations of future pressures for appreciation which, in the Norwegian context, could be associated with the future build up of net foreign assets, could therefore deter foreign investment for long periods.
69. Another concern in transferring the responsibility of controlling cyclical fluctuations to monetary policy is that it could reduce the incentives of the government to maintain a firm hold on fiscal policy. Since late 1993 the Norwegian authorities have relied upon an economic strategy called the Solidarity Alternative, in an attempt to preserve the competitiveness of the mainland economy in the face of a large surge in oil revenues. Under this strategy, the unions have consented to moderate wage settlements, in return for the government’s commitment to orient monetary policy toward stabilizing the exchange rate, while fiscal policy is used for demand management. Over much of the period since 1993, fiscal policy was used actively to moderate the cyclical upswing in the economy against a backdrop of rising fiscal surpluses, although the fiscal effort has weakened in 1997-98. Some Norwegian commentators believe that without the constraint of moderating cyclical imbalances, pressures for a more expansionary fiscal policy would heighten with adverse consequences for the competitiveness and sustainability of the non-oil sector over the medium term (Dutch disease effects).
70. One way of transferring the responsibility of moderating cyclical imbalances to monetary policy while maintaining fiscal discipline would be to change the focus of fiscal policy from the immediate cyclical situation to a more long-term horizon. The Ministry of Finance has made efforts in this direction in recent years by emphasizing the need to set aside resources to finance rising pension and health obligations that will arise in the future, owing to demographic changes, at a time when oil reserves are expected to be greatly diminished.
71. Even after a decision to adopt an inflation target, exchange rate movements would need to be taken into account prominently in assessing inflation prospects, because the import weight in the CPI is about 40 percent, compared to a weight of about 25 percent in the other resource-based industrial countries. Therefore, if the exchange rate depreciated sharply, Norges Bank might choose to raise interest rates in order to moderate the future effects of exchange rate movements on inflation. Central banks following an inflation target generally accommodate modest, temporary fluctuations in the exchange rate and only change interest rates if the exchange rate movements are expected to lead to permanent effects on the inflation rate, through expectations or wage developments. However, some resource-based countries that rely on inflation targeting make explicit allowance for significant exchange rate depreciation in the event of terms of trade shocks. According to the models developed at Norges Bank and Statistics Norway, a 10 percent depreciation would result in a 2-2½ percent increase in the CPI within one year, rising to 4 percent after three years.
72. Although targeting the aggregate inflation rate is generally understood by the public, deviations from inflation targets are often allowed in inflation targeting regimes in response to supply shocks such as changes in food and energy prices, indirect tax changes, and imputed rental costs. Indeed, cross-country experience suggests that, if and when Norway decided to switch to an inflation target, it would be prudent to consider an inflation target which excluded the effects of temporary supply shocks. In New Zealand the Reserve Bank has identified one-off shocks to prices arising from supply-side developments to which it does not have to react in pursuing the inflation target. These include exceptional movements in commodity prices, changes in indirect taxes, and other government policy changes that directly affect prices. In Canada, the inflation target excludes food and energy prices and the contribution of indirect taxes, and in the United Kingdom, mortgage interest payments are excluded from the inflation target. Although the Swedish Riksbank targets the aggregate CPI, the large effects on the CPI of indirect tax changes and sharp reductions in interest rates in recent years has led to increased emphasis in its inflation reports on the evolution of inflation excluding mortgage interest costs and indirect tax and subsidies.17 Norges Bank periodically reports inflation excluding volatile electricity prices and indirect taxes and work by Bjørnland (1998) indicates that excluding oil prices from the determinants of core inflation (on the basis that it affects long-run output) leads to a smoother inflation series.
73. Finally, an issue which is central to the successful implementation of inflation targeting is central bank independence.18 This is generally understood to mean instrument or operational independence for the central bank to pursue its monetary policy goal free of short-term political pressure from the government; the goal itself is generally set by the government. An important condition for allowing the central bank to carry out its policies independently is that it is held accountable for its actions. This requires a transparent reporting system for the central bank’s policy actions through publications and appearances in parliament.
74. The legal framework in the United Kingdom and in New Zealand provide a flavor of the type of measures that have been implemented to secure central bank independence in these countries. The instrument independence granted to the Bank of England in 1998 provides it with full freedom to achieve the inflation target of 2½ percent but, if the actual inflation rate deviates more than 1 percent either side of the 2½ percent target, the Governor of the Bank of England is expected to write an open letter to the Chancellor explaining the reasons for the divergence from target. The Chancellor can also override the Bank’s decisions on the use of monetary instruments in extremis, but only in an open way. The Reserve Bank of New Zealand has been granted instrument independence since 1989 and its accountability is achieved through a periodic review of monetary policy by the Bank’s Board of Directors who report to the Treasurer. An unfavorable review can lead to the dismissal of the Governor.
75. In Norway the central bank has de jure instrument independence because the government has no legal authority to instruct Norges Bank about interest rate decisions, unless the issue is brought before the King in Council. This has never been tried because of the potential for adverse publicity and damaging asset price fluctuations in the money markets. However, the ability of the government to invoke a meeting of the King in Council is generally seen as providing a means to limit the de facto instrument independence of Norges Bank. If Norway were to choose to adopt an inflation target, the existing central bank laws would need to be amended and new policy statements issued, establishing the inflation target and allowing complete freedom for Norges Bank to implement this strategy.
76. This paper has noted that, as a small open economy, Norway can benefit significantly from the use of a nominal anchor—such as an exchange rate or inflation target—to help guide price expectations. However, it is difficult for a commodity exporting country, such as Norway, to keep its exchange rate stable in the face of large terms of trade shocks. Moreover, Norway has been facing the potential for increased exchange rate volatility in the runup to the third phase of EMU, owing to the small size of its market in relation to the combined financial markets of the EMU participants, with whom it has close economic ties. These considerations suggest that Norway should consider either the adoption of a more formal link to the euro, or a shift to inflation targeting.
77. Adopting a fixed rate against the euro, in connection with formal arrangements that would provide for liquidity support from the ECB, would reduce the problem of speculative attacks. However, in order to receive adequate liquidity support from the ECB to help defend the krone/euro parity, it is likely that Norway would have to join the European Union, a policy which has little political support at present in Norway.
78. The adoption of an exchange rate target implies that interest rate policy is guided primarily by exchange market developments, rather than domestic economic conditions. In Norway this can lead to relatively frequent policy dilemmas, as terms-of-trade shocks affect Norway differently from most of its European trading partners and the Norwegian cyclical position is frequently out of line with the rest of Europe. While fiscal policy can, in principle, be used actively to resolve this dilemma, the experience has been that there are limits. Therefore, the possibility of shifting to an inflation target has received considerable public attention in Norway in recent years.
79. If a decision were taken to base monetary policy on an explicit inflation target, Norges Bank would have to continue monitoring the exchange rate closely because of the large weight of imported goods in the CPI basket in Norway. Moreover, adequate support from fiscal policy would have to be maintained So that policy-induced movements in interest rates would not be too sharp. Norges Bank is well placed to take on the added responsibilities of an inflation target because it already issues quarterly inflation reports and has a fairly well developed macroeconomic model of the economy which could be used for inflation forecasting. However, steps would need to be taken to increase the operational independence of Norges Bank, in order to buttress the credibility of the inflation target. The experience of the Swedish Riksbank—which has a similar institutional set-up—in establishing an inflation target indicates that the Norges Bank could also adopt an inflation target successfully within a fairly short time, provided that it received adequate support from fiscal policy and was given instrument independence which was operative.
R.Amano and S.Van Norden (1995) “Terms of Trade and Real Exchange Rates: The Canadian Evidence”Journal of International Money and FinanceVol. 14No. 1.
BjørnlandH. (1997) “Estimating Core Inflation,”Statistics Norway Research Department Working PaperAugust.
C.Bean (1998) “The New UK Monetary Arrangements: A View from the Literature” CEPRLondon and H.M. Treasury.
S.Cecchetti (1995) “Inflation Indicators and Inflation Policy”NBER Working Paper No. 5161
G.Debelle(1997) “Inflation Targeting in Practice”IMF Working Paper No. 97/35.
A.Dixit (1989) “Hysteresis, Import Penetration, and Exchange Rate Pass-Through,”Quarterly Journal of Economics 104.
D.Gruen and J.Wilkinson (1994) “Australia’s Real Exchange Rate—Is It Explained by the Terms of Trade or by Real Interest Differentials?”The Economic RecordVol. 70No. 209.
HaldaneA. (1997) “Some Issues in Inflation Targeting,”Mimeo, Bank of England.
E.Hansen (1997) “Exchange Rates, Non-Traded Goods, and the Terms of Trade: An Empirical Application for New Zealand,”Oxford Bulletin of Economics and Statistics No. 59
International Monetary Fund (1996) “The Rise and Fall of Inflation—Lessons from the Postwar Experience” World Economic OutlookOctober1996.
LaneT. and S.Van der Heuvel (1998) “The United Kingdom’s Experience with Inflation Targeting,”IMF Working Paper No. 98/87.
LoweP.“Monetary Policy and Inflation Targeting: Proceedings of a Conference,”Reserve Bank of Australia1997
MishkinF. (1998) “International Experiences with Different Monetary Policy Regimes,”Sveriges Riksbank
K.Mourn and T.Eika (1998) “Pengepolitikk som virkemiddel I dagens konjunktursituasjon,”Økonomiske Analyser 2/98 Statistics Norway.
J.Qvigstad“Norwegian Traditions and International Trends,” in A.Christiansen and J.Qvigstadeds. Choosing a Monetary Policy Target Scandinavian University Press Oslo1997
SamieiH.1998“Determinants of the Real Exchange Rate in Norway; Does Policy Matter” in Norway-Selected Issues SM/98/39.
L.Svensson“Exchange Rate Target or Inflation Target,” in A.Christiansen and J.Qvigstadeds. Choosing a Monetary Policy Target Scandinavian University Press Oslo1997
L.Svensson (1998) “Inflation Targeting as a Monetary Policy Tulle,”MimeoSveriges Riksbank.
SyrstadH (1998) “Mål og virkemidler I pengepolitikken-hva kan sentralbanken bestemme?”in Norsk jurudisk tidsskrift number 8.
Prepared by Alun Thomas.
This section is based on J. Qvigstad, “Norwegian traditions and international trends,” in A. Christiansen and J. Qvigstad eds. Choosing a monetary policy target Scandinavian University Press, Oslo 1997
The monetary policy guidelines announced in a Royal Decree in May, 1994 indicated that monetary policy was to be aimed at maintaining a stable exchange rate of the krone against European currencies, based on the range of the exchange rate maintained since the krone was floated on December 10, 1992. These guidelines are still in effect.
The euro differs from the ECU in excluding the Danish and Swedish kroner and UK pound from its basket of currencies.
Over the past two decades the cyclical position in Norway has diverged from the cyclical position in the euro zone, due in part to the importance of Norway’s oil sector (Figure 7).
In addition, in contrast with targeting a monetary aggregate, it is able to deal with sudden changes in velocity because it does not rely on a stable money-inflation relationship.
See in particular, “The monetary transmission mechanism in Sweden” Selected Issues Sweden 1997.
Partly in response to criticism that the actual inflation rate has come below the desired inflation target band in recent months, Statistics Sweden has begun publishing this measure of the underlying inflation rate and is in the process of refining it.