Wholesale and retail sales statistics

Turnover: Includes income from taxable and tax-free sales of goods and services. Financial income, government subsidies or value added tax are not included.

Name: Wholesale and Retail Trade Sales Statistics
Theme: Wholesale and retail trade and service activities

Structural Business Statistics Division

Concerns bi-monthly data only: Regular publication of data disaggregated by counties in addition to data published at the national level. Other geographical distributions can be ordered.

Bimonthly figures and quarterly indices.

The bi-monthly turnover statistics are initially published approximately seventeen weeks after the expiry of the statistical period. The revised data is then published two and four months respectively after the initial publication.

The quarterly indices are initially published approximately sixty days after the expiry of the quarter. They are published only at the two-digit NACE sector level (NACE 45 and 46). Revised data is released with the next release, ie three months after the initial release.

Published data is revised up to two calendar years back. This is explained in the table below, with quarterly indices as an example. The superscript T refers to the current year, while the subscript 1-4 refers to the relevant quarter.

The quarterly indices are communicated to Eurostat, both in raw data and in data corrected for working days, sixty days after the end of the quarter.

Primary data and compiled statistics are stored electronically in SAS programming language files.

The purpose of the statistics is to identify the level and development of wholesale and retail trade. The statistics have been produced since 1995.

From 1995 to 2006, only fortnightly data were published at the national level. The quarterly indices have been published by Eurostat only. From the first quarter of 2007, the quarterly indices are also published at national level.

Users include public and private sector agencies and organizations. Statistics Norway’s National Accounts statistics rely on the timely production of wholesale and retail trade statistics. Other users include the research department of Statistics Norway.

Similar statistics are being compiled for all dutiable industries. The trend in turnover growth is compared to the monthly retail trade index. Current developments and level figures are compared with the annual structural statistics for wholesale and retail trade.

Statistics Act, paragraphs 3-2.

EU Council Regulation No. 1165/98 of 19 May 1998 concerning short-term statistics.

The population is made up of all units subject to VAT in the wholesale and retail sectors, repair of vehicles and household and personal goods included. Agency trade is not part of the population. This means NACE 45, 46 (except 46.1) and NACE 47, according to the Standard of Industrial Classification (SIC2007). Units belonging to the state and units belonging to the social security administration or to local or regional governments are not part of the population. The observed unit is essentially the same as the unit subject to compulsory VAT, ie enterprises with annual turnover exceeding NOK 50,000.

The VAT register

Wholesale and Retail Trade Statistics represent full census survey data for the entire population.

The data is collected from the VAT register.

Turnover is checked against previously received data. If the discrepancy is significant, the individual VAT unit is contacted. The information in the VAT register is checked against the central register of establishments and enterprises and the central coordinating register of legal persons.

For fortnightly statistics, the data is added together. When a VAT unit consists of several active enterprises, the data is broken down according to the turnover recorded in the last annual data file of the central register of establishments and enterprises.

Quarterly indices are calculated from fortnightly statistics. It is therefore necessary to divide some of the fortnightly data into two. For division, the number of working days in a month is used. The model for calculating quarterly data is described below:

The variable Q refers to quarterly figures, B to fortnightly figures and wd to the number of working days in March, April, September or October. The superscript T refers to the actual quarter, while the subscript T refers to the current year. The index j indicates that the data are broken down according to the most detailed NACE level possible, that is to say the three-, four- or five-digit sectoral level. The h subscript refers to the two-digit sector level.

If a table cell contains fewer than three observations, the data for the table cell itself is not published. This is necessary because it would be possible to identify the individual data of the real companies. This is especially important for regional data.

In the period 1995 to 2002, the statistics were published according to the standard industrial classification (SIC1994). Since 2002 it is published according to the standard industrial classification (SIC2002) and since 2009 it is published according to the standard industrial classification (SIC2007)

Changes in statistics: Prior to the first biweekly in 2001, statistics only included dutiable sales. Since the first quarter of 2001, the statistics include both dutiable and duty-free sales.

Measurement errors (the respondent provides incorrect data) and processing errors (misinterpretation of the numbers and letters entered on the paper forms during optical reading) can occur. These types of errors are attempted to be avoided by control routines, as explained in chapter 3.5. Another possible error can occur if outdated distribution formulas are used to divide the turnover in cases where a VAT unit consists of several companies.

Errors in the estimation of quarterly indices may occur if the evolution between the first and second quarters, or the fourth and fifth quarters, of the bi-monthly statistics is significantly different last year and this year.

The County Revenue Office is contacted in cases where large businesses have not responded. For the first and second publication, total non-response is imputed. At the third and last publication, the total non-response turnover is set to zero.

Non pertinent

Other errors in the VAT register, under declaration included. This is difficult to assess, but it is assumed that it does not represent a substantial source of error.