Norges Bank's regional network: fresh and useful information.

• Author: Brekke, Henriette

1. Introduction

   Norges Bank established its regional network in 2002. Regular interviews with contacts in Norwegian industry are used to obtain information about enterprises' view of the current situation and outlook before other official statistics are available. This contact also results in a better understanding of the dynamics within and between sectors and a better insight into what is driving current economic developments. This insight from the enterprises in the network, together with the regular reports on economic developments, makes the regional network an important factor in decision-making by Norges Bank's Executive Board in its conduct of monetary policy. The economic developments reported are published regularly and made available on Norges Bank's website. Summaries of the national and regional reports are also published here.

   Kallum, Sjatil and Haugland (2005) reported in their article on how the regional network is organised and what information is collected. Their article also includes a comparison of information from the regional network with official statistics. Since that article was published, the Norwegian economy has experienced a strong economic upswing and a financial crisis. It is therefore now appropriate to re-evaluate the information from the regional network in order to assess how well the network has captured these economic developments. This article also presents a broader variable set than previously published. It should therefore be a useful reference for those following the regional network's regular published reports.

   Section 2 provides some general information about the regional network. To explore how well the network captures and predicts developments in the economy, in Sections 3 and 4 we compare the series from the regional network graphically with relevant series from other available statistics, and calculate correlation coefficients. We use series for the national average, sector averages and regional averages. Section 3 looks at the network's retrospective series for reported developments, while Section 4 covers the network's prospective series. Section 5 summarises the most important results.

2. What is Norges Bank's regional network?

   The regional network (RN) consists of around 1 500 enterprises, organisations, local authorities, hospitals and other public bodies right around the country. We refer to these as contacts or contact enterprises, and each one is contacted once or twice a year. We have divided the country into seven regions: Inland, Mid-Norway North, North-West, South-West, South and East. Information is normally collected through four ordinary contact rounds a year. (2) We hold meetings with 40-50 contact enterprises in each region, making a total of almost 300 per round. The regular themes for contacts in the private sector can be divided into four categories: output and utilisation of production capacity; investment plans; employment and labour supply; and costs, prices and profitability (see Appendix A). The regular themes in the public sector are: investment; employment; labour supply; and costs (see Appendix B). Besides these regular themes, a topical special theme is chosen for each round, such as the use of foreign labour or planned action in response to the financial crisis.

   Contacts are divided into the following sectors: manufacturing, building and construction, retail trade, services and public sector. Manufacturing contacts are categorised into export industry, domestically-oriented manufacturing and suppliers to the petroleum sector, according to where the bulk of their output is aimed. Contacts in the service sector are divided into household services (B2C) and corporate services (B2B). In the public sector, contacts are based at local authorities, regional authorities and hospitals. The regional network does not have contacts in oil production or overseas shipping, because the focus is on the mainland economy. Nor does the network include contacts from agriculture and some other primary industries, because these industries are deemed to be regulated and are guided more by factors other than the business cycle. We do not have contacts in central government, but obtain information periodically from central government bodies, such as the Norwegian Labour and Welfare Service's regional offices.

   Information-gathering focuses on what the contacts believe the developments in their sector to be. The information received is used to produce both quantitative statistics and qualitative reports. The quantitative statistics are compiled by weighting the information received to reflect developments in each sector in the various regions. The regional series for the individual sectors are then weighted to create regional averages, national sector averages and national averages for each theme. The national average for a theme, compiled by weighting the various national sector series, is referred to as the aggregated series. Martinsen and Wulfsberg (2009) provide a detailed description of the principles for this weighting process.

   One of the aims of the regional network is to obtain fresh information about developments in Norwegian industry. The time at which the data become available is therefore very important. The quarterly national accounts (QNA) are available roughly 50 days after the end of a calendar quarter. The regional network supplies data for the same quarter at a much earlier point. For example, the contact round carried out in November captures developments in October and part of November. The regional network signalled as early as December 2008 that output in the mainland economy had fallen in Q4. It was more than two months (19 February 2009) before the QNA for Q4 were released and showed the same.

   So that the information is as up-to-date as possible when presented at the Executive Board's rate-setting meetings, it is often collected across calendar quarters. Many of the official series against which the regional network data are evaluated are quarterly series, and so we also construct quarterly series. The quarterly values are a weighted sum of the two contact rounds covering the quarter in question.

3. Current situation

   One important aim of the regional network is to provide a good picture of the current situation in Norwegian industry. In the following, we compare the results for the series for output growth, capacity utilisation, labour supply, employment growth and price changes with other available statistics.

   3.1 Growth in output over the past three months

   We ask the contacts in the regional network about seasonally-adjusted growth in their production volumes over the past three months relative to the previous three-month period (Question 1.1 in Appendix A). We also ask them to consider whether their volume growth is representative of the sector as a whole. We categorise the responses on a scale from -5 to +5, where +1 corresponds to annualised quarterly growth of 1-3 per cent, +2 corresponds to annualised quarterly growth of 3-5 per cent, and so on. An annualised quarterly decrease of 9 per cent or more is categorised as -5, while an annualised quarterly increase of 9 per cent or more is categorised as +5. This information is used in the preparation of Norges Bank's projections of GDP for mainland Norway. It is, for example, included in one of the models in Norges Bank's System of Averaging Models (SAM), where it has demonstrated solid explanatory power. (3)

   In principle, the questions relating to output are designed to capture developments in gross output. However, as the network's output series are used for projections of value added (net output), it is also interesting to see whether the network's output series correlate with value added. (4) We have therefore compared the regional network's output series with the annualised quarterly growth in value added for mainland Norway (see Appendix C). We have used a trend-adjusted series from the QNA where the seasonal component and irregular components have been removed. Chart 1 plots this series against the network's aggregated output series and the network's output series for manufacturing and services.

   [GRAPHIC 1 OMITTED]

   Table 1 shows the correlation coefficients between the network's output series and the annualised quarterly growth in value added for mainland Norway.

   Table 1 reveals a strong correlation between most of the network's output series and value added for mainland Norway. The correlation is strongest for manufacturing, which shows stronger correlation than the aggregated series. Chart 1 shows that both the RN series for building and construction and the aggregated RN series for 2008 reported a fall in output before the QNA reported a fall in value added.

   We have also compared the regional network's aggregated output series with annualised quarterly volume growth in gross output in mainland Norway from the QNA (see Appendix D). Again we have used trend-adjusted series from the QNA where the seasonal component and irregular components have been removed. The RN series is expected to show strong simultaneous correlation with this QNA series. Chart 2 plots the series against one another, and the top line of Table 2 shows the correlation coefficient.

   [GRAPHIC 2 OMITTED]

   Both Chart 2 and the correlation coefficient of 0.91 show that the aggregated RN series closely mirrors developments in gross output for mainland Norway in the QNA series. In particular, the RN series and QNA series move almost identically from 2008 onwards. When the downturn in the economy struck in autumn 2008, the network reported a fall in overall output for mainland Norway in the last part of 2008 in December, whereas the QNA data, which also showed a decrease in output for the same period, were not published until February 2009. The network also provided information in mid-May 2009 that the downturn was easing, which was used for internal analyses in the Bank. In August 2009, the QNA...