In 2005, the rise in consumer prices adjusted for taxes and excluding energy products (CPI-ATE) was noticeably lower than projected in 2004. The deviation between the projections and actual developments is primarily ascribable to a stronger-than-expected exchange rate and lower-than-projected wage growth. After surprisingly low inflation at the beginning of the year, the projections published throughout 2005 were closely in line with actual price developments. Capacity utilisation in the Norwegian economy, as measured by the output gap, was somewhat lower than previously projected, but the deviation is small compared to the considerable degree of uncertainty surrounding this projection.
This article evaluates the projections for economic developments in 2005 as presented in Inflation Report 1/04 and subsequent reports. First, we look at developments in output and inflation through 2005. We then analyse in greater detail the deviations between the projections and actual developments. Finally, we compare our projections with those of other institutions, both for 2005 and over a somewhat longer time horizon.
The operational objective of monetary policy is 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. Monetary policy affects the economy with a lag and policy must therefore be forward-looking. Thus, projections for inflation and future economic development are an important basis for monetary policy decisions. Norges Bank continuously works to improve the basis for the projections. Analysing deviations between actual developments and projections is part of this work.
It is important to evaluate previous projections with a view to further developing analysis and projection work. This holds true whether the projections prove to be close to or far from the mark in retrospect. If the projections are not in line with actual developments, this does not necessarily mean the analysis underlying the forecasts was wrong. Conversely, a projection may be on the mark even though it is based on an analysis that proved to be of inferior quality.
When evaluating the projections, it may be useful to distinguish between the various causes of projection errors.
Uncertainty about the current situation
There is considerable uncertainty surrounding the actual state of the economy when projections are made. Such uncertainty is due in part to the fact that it takes some time before statistics are published, and in part to the fact that statistics are often subject to considerable revision at a later date. Norges Bank's view of the current situation is summarised in the estimate of the output gap. The output gap is the difference between the economy's actual output level and the output level that is consistent with stable inflation over time. The output gap is thus an expression of inflationary pressures in the Norwegian economy. The output gap is not an observable variable, which implies that historic values must also be estimated. Our estimate of the output gap is based on technical calculations and our assessment of various indicators. The estimate of the output gap may change if national accounts figures are revised or if more information emerges and new methods are developed that provide a basis for reassessing potential output and capacity utilisation in the economy.
Random disturbances and errors in assumptions
The projections are based on several exogenous assumptions. If these assumptions develop differently than we had expected, this may lead to projection errors. Among the most important assumptions are developments in GDP growth, inflation and interest rate developments among our trading partners. The price of oil and developments in petroleum investment are other important assumptions in addition to the activity level in the central and local government sectors.
Up to Inflation Report 3/05, Norges Bank's projections were also based on technical assumptions regarding developments in interest rates and the krone exchange rate, based on developments in forward rates. In the first two Inflation Reports in 2005, forward rates were adjusted upwards somewhat towards the end of the projection period. Analyses indicated that long-term interest rates might be pushed down by temporary conditions and did not therefore reflect actual expectations concerning the future interest rate level. (2)
Since Inflation Report 3/05, Norges Bank has based its projections on the Bank's own forecast for developments in interest rates ahead. Consequently, the interest rate path can no longer be viewed as an independent projection but is instead the result of a simultaneous process with the projections for capacity utilisation and inflation. The interest rate path should provide a reasonable balance between the considerations Norges Bank is to emphasise in interest rate setting, and thus becomes both a reaction to and a basis for the other projections.
Random events can also affect the economic variables that we forecast. Such random disturbances will naturally be unexpected and therefore lead to projection errors. One example is the introduction of new VAT rates, which had some bearing on price developments in 2005. Another example is regulatory changes that probably led to a considerable drop in sickness absence through 2004, which increased the supply of labour.
The functioning of the economy
The projections are based on our understanding of the functioning of the economy, which is based on theory and empirical analyses of history. Structural changes in the economy's functioning can be difficult to capture and take into account before they occur. Moreover, even though we are aware of emerging structural changes, it can be difficult to assess their implications. Further, it is often difficult to determine the duration of various changes that arise. An example of this is the shift in trade towards low-cost countries which has led to a steady fall in prices for imported consumer goods in re-cent years. Initially, Norges Bank considered the shift in trade to be a transitory phenomenon affecting only certain groups of goods, but evidence suggests that this trend may persist for some time and affect a wider range of goods.
2 Inflation and output through 2005
The economic recovery that has marked the Norwegian economy since the beginning of 2003 continued in 2005. As estimated by Norges Bank, the output gap was probably positive at the beginning of 2006. Low interest rates, increased petroleum investment, strong international growth and an improvement in the terms of trade have been the main driving forces behind the upturn. Low interest rates contributed to high growth in private consumption and residential investment. Growth gradually became more broad-based, and in 2005 exports and mainland business investment expanded at a brisk pace.
Compared with previous upturns, however, it took longer for employment to pick up. A sharp drop in sickness absence throughout 2004 led to a relatively rapid increase in the number of person-hours worked (see Chart 1). Enterprises may thus have increased their supply of labour without having to hire new staff. The supply of additional resources as a result of lower sickness absence, combined with low wage and price inflation, is one of the main reasons why Norges Bank assumes that potential growth in the Norwegian economy was somewhat higher than normal in both 2004 and 2005. As a result, the economy has been able to grow at a faster pace without giving rise to bottlenecks and cost inflation. Towards the end of 2005, employment also picked up as a result of continued strong growth in output and demand.
While capacity utilisation in the economy recovered from the relatively moderate cyclical trough in 2003, inflation also picked up from its very low level in 2004 (see Chart 2). Consumer price inflation adjusted for taxes and excluding energy products (CPI-ATE) was 1.0 per cent in 2005. Through 2005, the decline in prices for imported consumer goods varied between -0.5 and -1.5 per cent. A higher rise in prices for domestically produced goods and services contributed to a sharper rise in consumer prices during the first six months. Towards the end of the year, however, both lower domestic inflation and an accelerating fall in prices for imported consumer goods pushed down CPI-ATE inflation. Other measures of underlying inflation showed somewhat stronger inflation than the CPI-ATE for 2005 as a whole (see Chart 3). The difference was, however, less pronounced than in the two previous years. Partly owing to increased VAT rates and high energy prices, CPI inflation has picked up.
3 Deviations between projections and actual developments
Table 1 shows central assumptions and projections for 2005 in the Inflation Reports published since autumn 2003. In the box entitled "Changes in the Projections" (pp. 101-102), a more detailed account is given of the changes made to the projections in the various Inflation Reports.
The output gap
Estimates of the output gap in Norway in 2005 have not been substantially changed, in view of the considerable uncertainty surrounding this variable. Norges Bank's current assessment is that the output gap was close to a normal level in 2005 as a whole, but positive towards the end of the year. Even though this is somewhat lower than projected in the Inflation Reports, the projections have been based throughout on the assumption that the output gap would gradually close and become positive in 2005.
The estimate for the output gap further ahead reflects a combination of three uncertain variables:
* Estimated current output gap or the current economic situation.
* Projected potential growth in the economy; i.e. how much output can...