Interest rate projections in theory and practice.

• Author: Bergo, Jarle

1. Introduction: The importance of the expectations channel in the conduct of monetary policy

   The most important task of monetary policy is to provide the economy with an anchor for inflation expectations--a nominal anchor. In Norway, monetary policy is oriented towards low and stable inflation with an annual rise in consumer prices of close to 2.5 per cent over time. Norges Bank operates a flexible inflation targeting regime, so that weight is given to both variability in inflation and variability in output and employment. Flexible inflation targeting builds a bridge between the long-term objective of monetary policy, which is to anchor expectations of low and stable inflation, and the more short-term objective of stabilising economic developments.

   Preventing inflation expectations from becoming entrenched markedly below target was one of the main reasons for reducing the key policy rate to a very low level when inflation fell and approached zero in 2003 and 2004. At that time there was also spare capacity in the Norwegian economy. Between the end of 2002 and the first quarter of 2004 the key rate was reduced by 5 1/4 percentage points (see Chart 1). We indicated that the interest rate would remain low until we saw clear signs of rising inflation.

   Since summer 2003, the Norwegian economy has been in a clear upswing. Low interest rates, high oil prices and a favourable global environment have been important driving forces. Growth is strong in most industries, and profitability in the business sector is solid. Underlying inflation is still considerably below the inflation target. However, several factors point to higher inflation further ahead. We are now normalising the interest rate gradually. Between the summer of 2005 and the beginning of 2007, the key rate has been increased by 2.0 percentage points and there are prospects of further interest rate hikes.

   The shortest money market rates are determined by the central bank via the key policy rate (see Chart 2). But private-sector consumption and investment decisions depend more on expectations regarding future developments in the key rate. To be successful, monetary policy must be able to influence these expectations. The public must therefore understand the central bank's intentions in interest-rate setting. Transparency regarding Norges Bank's monetary policy assessments probably improves the predictability and effectiveness of monetary policy.

   In recent years, we have tried to facilitate the public understanding of our actions. The background material for the Executive Board's monetary policy meetings is published and the assessments underlying interest rate decisions are explained. As from the end of 2005, Norges Bank has published its own interest rate forecast. From using technical assumptions or others' assessments, we have now taken ownership of the interest rate path in our projections. In the Monetary Policy Report (2), Norges Bank publishes the interest rate path that in the Bank's view provides a reasonable trade-off between stabilising inflation at target and stabilising developments in output and employment.

   So far, the experience of publishing our own interest rate forecasts has been positive, it seems that economic agents have understood the nature of the forecasts. Nevertheless, there is considerable uncertainty surrounding the interest rate path, which is why we present fan charts with uncertainty intervals around the forecasts. In addition, the Monetary Policy Report contains several different sensitivity analyses to illustrate alternative interest rate paths that would be preferable should economic developments deviate from the baseline scenario. We also present a kind of "interest rate accounts", where we explain any changes in the interest rate path since the previous Report, for instance caused by actual developments in key variables differing from our assumptions. The specific interest rate path cannot and must not be looked upon as a guarantee, a path to which we unconditionally have committed ourself. To the contrary, should economic developments deviate from the projected path, the interest rate path will also shift. Instead, it can be said that through our communication we commit ourself to a pattern of behaviour, a response pattern. If interest rate expectations can be influenced, it will in many cases be useful for a central bank to commit itself to a predictable response pattern. This kind of commitment can, if it is perceived as credible, enhance the effectiveness of monetary policy.

   [GRAPHIC 1 OMITTED]

   [ILLUSTRATION OMITTED]

   A relevant question is the extent to which our communication actually influences interest rate expectations. Forward interest rates derived from yields at various maturities will in the absence of term premia and other risk premia normally reflect the market's short-term interest rate expectations (3). When Inflation Report 3/06 was published in the beginning of November last year, the forward interest rate was on a par with our forecast for the next six months, but considerably lower thereafter (see Chart 3). Since then, forward rates have increased and approached Norges Bank's interest rate path. It is of course debatable whether it is our communication and actions or new information that has brought about the alignment. It is probably both. Forward rates somewhat further out are still lower than our forecast. The reason may be that market participants have a different perception of the interest rate path that is necessary to stabilise inflation at target and to achieve stable developments in output and employment. Alternatively, the market may have the same short-term interest rate expectations as Norges Bank, but because of extraordinary conditions long-term bond prices are being pushed up and, consequently, long-term bond yields are being pushed down.

   [ILLUSTRATION OMITTED]

2. What is the normal interest rate level?

   When preparing an interest rate forecast, we must have a view of what the normal interest rate level is. Long-term bond yields have been at historically low levels in recent years. From lying in a broad range around 10 per cent at the end of the 1980s, they have fallen to around 4 per cent in the past few years (see Chart 4). Developments in nominal interest rates must be seen in the light of inflation developments. Since 1960, inflation has been relatively low and stable only in the past 10-15 years, i.e. since the first half of the 1990s (see Chart 5). High and variable inflation pushed up nominal interest rates earlier, both directly and via an inflation risk premium. Uncertainty about future inflation generates uncertainty as to the real value of investments and investors may require an extra compensation--a risk...