Forecasting in Norges Bank.

Author:Kloster, Arne

Norges Bank's forecasts of economic developments form an important part of the basis for monetary policy. The projections of economic variables and Norges Bank's interest rate forecast are interdependent. Current information about economic developments, judgment and various economic models are all employed in our forecasting work. The current economic situation and developments in the next few quarters are assessed first, and then we determine the long-term impact of driving forces in the economy. This article describes the work on forecasting.

1 Introduction

Norges Bank's forecasting forms an important part of monetary policy. In the work on forecasting, we analyse the current economic situation and the driving forces that will influence the economy in the period ahead. Since Inflation Report 3/05, demand, output and inflation forecasts have been based on Norges Bank's own interest rate forecasts. The forecasts of the interest rate and other economic variables are interdependent: the interest rate is influenced by developments in output and inflation, while interest rate developments must be assessed against the background of output and inflation forecasts. They indicate whether the interest rate forecast strikes a reasonable balance between the objectives of monetary policy. The forecasts of the interest rate and other variables are therefore made simultaneously in a multiple iteration process (2).

The purpose of this article is to provide insight into Norges Bank's forecasting work, and to describe the methodology underlying the projections published in the Inflation Report. The article discusses the various stages of the work and the tools used. The main emphasis is on the procedure. (3)

The structure of the forecasting work is illustrated in Chart 1. The projections of developments ahead are based on two premises in particular. The first is ah assessment of the current economic situation and short-term forecasts. The second is forecasts for exogenous variables--those that have to be determined outside our model. On the basis of these premises, we use our macroeconomic core model to produce ah initial set of projections for developments in output, inflation, the interest rate and the exchange tare. The forecasts in the Inflation Report cover a broader set of economic variables. They are produced through an iteration procedure between projections based on the core model and a system of smaller models surrounding it. The use of judgment plays a decisive part in shaping the economic outlook presented in the Inflation Report.

The analysis of the current economic situation and short-term developments is described in more detail in section 2. Section 3 describes the procedure for projecting developments further ahead.


2 Analysis of current situation and short-term developments

Current situation

The projections in the Inflation Report are based on an interest rate path that in the view of the Executive Board provides a reasonable balance between the objectives of monetary policy. In order to project future economic developments in a way that provides the best basis for these assessments, it is decisive that we have a reliable analysis of the current economic situation. The analysis of the current situation is based mainly on current statistics and other information about economic developments. However, short-term statistics are often uncertain, and there may be a long lag between the measurement and publication of new figures. Information from Norges Bank's regional network therefore provides an important supplement to current statistics. This network consists of companies, organisations and municipalities throughout Norway. Five rimes a year, business and community leaders are interviewed about developments in their sectors, and the impressions they provide form part of the basis of our assessment of the current and near-term economic situation. Preliminary surveys indicate that the network provides reliable information about developments before it becomes available through official statistics. (4)

An important part of the work consists of analysing the driving forces behind the current economic situation. The analysis of the current situation culminates in ah assessment of capacity utilisation and inflationary pressures in the economy today, and forecasts for developments in the next few quarters (see Chart 2).

The output gap

The estimate of the output gap expresses our assessment of total capacity utilisation in the economy. The output gap is defined as the difference between actual output and potential output, which is the output level that is consistent with stable inflation over time. The output gap has to be estimated, as potential output is ah unobservable variable. In order to make a correct assessment of economic pressures, and hence inflationary pressures ahead, it is important to that our assessment of the output gap at the time in question is correct. If the initial level is incorrectly estimated, the error will be diffused over the forecast period. This will impair the basis for assessing which interest rate path can best contribute to achieving the desired developments ahead. The desired developments are illustrated in Chart 3.



Our estimate of the output gap is the result of an overall assessment of available information concerning resource utilisation in the economy. Norges Bank's output gap estimates are based on several different methods and data sources, which may help to reduce the uncertainty of the estimates. (5)

* We base our assessment of the output gap on technical calculations of trend growth in the Norwegian economy. There are a number of methods for estimating the output gap. The most commonly used methods generally provide the same picture of cyclical fluctuations over time (see Chart 4) (6). At certain times, however, the different methods may result in fairly different estimates of the output gap level. Our estimate is based on trend growth as calculated using a Hodrick-Prescott filter, but the results of other methods are also included in the assessment.

* Our estimate of the potential GDP growth rate is adjusted if we have information about extraordinary factors that influence developments. In the past, for example, we have made adjustments for the increase in the number of vacation days in 2001 and 2002, and the rapid decline in sickness absence in 2004. These are examples of changes that are captured by technical methods after a period, but which would have to be adjusted if we have information on sudden shifts.

* In our assessment of the output gap we also take into account other indicators that provide direct or indirect information regarding the utilisation of resources in the Norwegian economy. Among other things, we use information from Norges Bank's regional network to develop an index of average capacity utilisation in the Norwegian economy which can be compared with the output gap (see Chart 5). Statistics Norway's business tendency survey for manufacturing, and in particular the capacity utilisation index, may also function as a cross-check.

* The situation in the labour market provides important information about the output gap. The unemployment level fluctuates with the business cycle, and forms an important basis for assessing capacity utilisation. The unemployment level that is consistent with stable price and cost inflation is uncertain, however, and has to be estimated. The level of unemployment that is consistent with normal resource utilisation in the economy may also vary over rime, among other things as a result of structural changes in the labour market. At the same time, the different unemployment statistics may in periods give different signals about labour market tightness. Historical experience plays an important part when these factors are assessed.

* Developments in the labour force, the number employed and person-hours worked also provide information about the extent to which available labour resources are being used. These variables normally fluctuate with the business cycle, and can provide a more nuanced picture of resource utilisation than that obtained by looking at the unemployment rate in isolation. One approach is to consider developments in unemployment, person-hours and the labour force as deviations from calculated trend levels, as shown in Chart 6.

* Developments in financial variables can also provide information about capacity utilisation in the economy. Some financial indicators appear to correlate closely with the output gap, and in some cases can predict developments. See Gerdrup et al. (2006) for a more detailed account of these indicators.

* The link between real and nominal variables is largely found in the labour market. Wage developments provide ah indication of how the social partners assess the tightness of the labour market. A considerably stronger increase in real wages than in labour productivity implies strong competition for labour, high capacity utilisation in the economy and inflationary pressures.




The output gap can also influence prices directly through profit margins in the enterprise sector. When demand for goods and services is high relative to production capacity, prices tend to rise to a further extent than implied by firms' costs. The extent to which margins increase in such a situation will depend especially on the competitive situation in the various industries. An assessment of the competitive situation in the economy is therefore important for estimating the effect of higher demand on inflation.

Owing to increased cross-border labour mobility and increased trade in services, assessing the output gap is more complicated than earlier...

To continue reading