Cross border trade

Name: Cross border trade
Topic: Wholesale and retail trade and service activities

Division for Business Cycle Statistics

National and regional

Quarterly and annual, published eight weeks after the reference quarter.

Annual reporting to the Statistical Office of the EU (Eurostat).

Collected and revised data are stored securely by Statistics Norway in compliance with applicable legislation on data processing.
Statistics Norway can grant access to the source data (de-identified or anonymised microdata) on which the statistics are based, for researchers and public authorities for the purposes of preparing statistical results and analyses. Access can be granted upon application and subject to conditions. Refer to the details about this at Access to data from Statistics Norway.

To sample the span of Norwegian households’ transactions abroad, the travel survey covers the transactions made when nights have been spent abroad. For day trips, the cross-border trade surveys the transactions.

The statistics for cross-border trade were established in 2004 to replace the central bank of Norway's currency statistics, which fell away from 2005. From January 1st 2023, Statistics Norway changed the method for collecting data for the cross border trade survey. The new data collection includes more detailed information regarding the cross border trade. The new survey is based on monthly data collections which are compiled into quarterly estimates. The previous survey was based on quarterly data collection.

The statistics are used by public authorities, trade organizations and international organizations like Eurostat. In addition, media is an important user. Internally, the national accounts are the main users.

No external users have access to statistics before they are released at 8 a.m. on ssb.no after at least three months’ advance notice in the release calendar. This is one of the most important principles in Statistics Norway for ensuring the equal treatment of users.

The data on cross border trade and the travel survey data give the total spending figures for Norwegian household spending abroad. These figures are used as input into Statistics Norway's balance of payments statistics. The number of same day visits are reported to Eurostat.

The Statistics Act § 2-1 (voluntary)

"REGULATION (EU) No 692/2011 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 6 July 2011 concerning European statistics on tourism and repealing Council Directive 95/57/EC."

Statistics covers a sample of the Norwegian population in the age group 16 - 79 years, and their expenditures on same day trips abroad without accommodation, independent of the purpose of the trip. The observation units are persons, and a sample of 4,000 people receives the survey each month. The results from this sample are inflated to apply to the entire population.

Expressions such as Norwegians, we, ours and similar that are used in publications and in other analyses of travel habits have nothing to do with the persons' citizenship. It refers to persons who reside in Norway and who have been removed from the Statistics Norway Population Register (BeReg).

The Population register in combination with telephone interview/CATI.

The travel survey is carried out among 2 000 Norwegians between 16 and 79 years of age. New sample is drawn every quarter.

The main source for the cross border trade statistics is monthly representative samples of people. We collect the information in interviews conducted online.

Residents of all the country's municipalities are included. The sample consists of 48,000 people in one half-yearly draw and randomized in a 12-month sample of people aged 16-79 per 1/1/2023. The sample is drawn as a self-weighting probability sample from Statistics Norway's demography and population database Population Register (BeReg). To secure the most up-to-date sample as possible, it was drawn as close to the start of the interview as practically possible.

Collection of data

The respondents receive an email where they are asked to log into Altinn where they follow a link to an online questionnaire. Before 2023, a telephone interview (Computer Assisted Telephone Interview CATI) was carried out. Data collection is monthly and starts on the first working day after the month for which data is to be collected. Most responses come in at the beginning of the month.

There are several checks in questionnaire to prevent registration errors when filling in the self-administered questionnaire online. On some occasions, the respondent receives warnings when registering an answer with extreme values. In other cases, controls have been added when the stated values ​​do not add up. If extreme Statistics Norway receive extreme values, these are treated separately in the processing of the collected data.

There is some variation in the response rate between months, but usually the response rate is around 45 percent.

Editing

Editing is defined here as checking, examining, and amending data. The production system for cross border trade is programmed in Python. Extreme values ​​and outliers are either discarded or the results from these are not inflated. The missing values for expenditures, ​​are imputed and expenditures stated in local currency, are converted to NOK. The results from the responses are inflated to apply to the entire population based on age, education, and region.

Estimations

To estimate the total values in cross border trade we must blow up the numbers from our sample. To do this we estimate a factor tied to each individual respondent. This factor tells us how many people this person counts for in the final estimates. We use population totals at the regional level, such that the sum of all factors in, for example, the Innlandet region, is the total (relevant) population in that region.

In addition to population totals we weigh the factors using demographic markers. It varies to what degree a group responds to our survey, and also what they respond when they do. If, for example, we have a group that is more likely to respond while also spending less money on cross border trade, we would underestimate the totals if we didn’t make any corrections. The demographic markers we group and weigh by are sex, age and educational attainment.

Not relevant

Interviewers and everyone who works at Statistics Norway have a duty of confidentiality. Statistics Norway has its own data protection officer. Statistics Norway does not publish figures where there is a risk of identifying individual data about persons or households. More information can be found on Statistics Norway’s website under Methods in official statistics, in the ‘Confidentiality’ section.

The cross border trade statistics were first published in October 2005 with figures from the first quarter of 2004 to the second quarter of 2005. Outdated time series can be found under «Closed time series” in the Statbank. The number of same day visits and expenditure, by destination, was expanded in 2010 from three to six shopping destinations. The entire time series back to 2004 was updated and figures can be found at the six shopping destinations back to 2004.

The number of cross-border trips and expenditure, by region of residence, broke between 2019 and 2020 after the regional reform was introduced in 2020. Under the outdated time series, figures from before the reform can be found.

From January 1st 2023, Statistics Norway changed the method for collecting data for the cross border trade survey. The new data collection provides more detailed information on cross border trade. With the new data collection, there will be no breaks in the time series for same day visits, expenditure, by destination region of residence or destination.

For the first time, cross-border trade is published broken down by product group. This means more detailed statistics that not only tell how much money Norwegians spend on cross border trade, but also what is being purchase. Figures for cross border trade with product categories will be published quarterly and wit extended product categories annually.

Information is also collected on which mode of transport has been used for cross border trade and the costs of transport. The purpose of the same day visit abroad is collected in the new survey; whether it is a trip in connection with work, commuting or in your spare time. For now, this will be published annually and there are no previous time series.

Coverage errors

We cannot expect to response from everyone who has been selected to participate in the survey. A large part of the sample does not want to participate in the survey, approximately 45 percent of the gross sample, responded to the survey. Other reasons for not participating in the survey are that the respondent was prevented by illness, language problems or the like. The coverage error rate does not vary significantly by gender and area of ​​residence. With regard to age, we see a clear tendency for the dropout to be greatest among young adults (20-29 years) and then fall evenly to the 65-69 age group where the dropout is lowest. The dropout also falls according to level of education. For the group with education at primary school level or below, as well as those with an unknown educational background, the dropout is respectively greatest. With regard to marital status, we see that the dropout rate is lowest among the married and slightly variable among the other groups. In particular, the apostasy accounts for two-thirds of the unmarried. The highest dropout by family phase is found among single people aged 25-44. I There are small differences in dropouts by region. The highest dropout is found in Western Norway, followed by Trøndelag. For the other parts of the country, there are only small differences.

Errors of non-respondents

Sampling bias in relation to one characteristic does not necessarily mean that the net selection is skewed in relation to other characteristics. On the other hand, good agreement between the distributions in the net and gross sample for one or more characteristics does not guarantee that the selection is not biased by other, and possibly unknown, characteristics. Such biases can occur both in the selection process and because some groups respond to a greater extent than other groups. It is more likely that deviations between the net sample and the gross sample than between the gross sample and the population will create bias. The deviations between the gross sample and the population are due to random sample variance, and one can expect that the individuals who are selected in each population group do not differ systematically from those who are not selected. In comparisons between the net sample and the gross sample, on the other hand, there is a greater risk that the people in a population group who have actually participated (the net sample) differ systematically from the people who have not participated (the dropout) because the differences are not random.

In this survey, the results are weighted, showing differences in the composition of the gross and net sample divided by gender, age, education and region. Some groups in the population are over- or under-represented in the net sample. The differences between people living in different parts of the country or densely/scattered settlements, as well as gender, are quite small and have little significance for the results. The differences according to gender, age, education and region have been found to be so different that weight is given to missing answers from these. In particular, we see that people who are young adults, unmarried and without children are underrepresented. These characteristics are largely coincident. Furthermore, we see that people with low education and people with a rural background from a country other than Norway are also underrepresented.

When some groups are underrepresented, there are also some who are overrepresented. In particular, this applies to people over 50, those with education at college or university level, married people and those with a rural background from Norway. See more details on committee representation in Appendix A.

The cross border trade survey is a sample survey, which means that there is uncertainty associated with the published figures. This sampling uncertainty can be calculated. For example, cross-border trade in the first quarter of 2023 is estimated at NOK 1.89 billion. Calculations of the sampling uncertainty show that the value we would have obtained if we had asked all Norwegians in the age group 16-79 years, instead of a sample, with 95 percent certainty would have been within the interval: NOK 1.43 to 2.35 billion. With a doubling of the sample from and including March 2023, lower uncertainty is expected in the estimates from and including figures for the 2nd quarter of 2023.

A revision is a planned change to figures that have already been published, for example when releasing final figures as a follow-up to published preliminary figures. See also Statistics Norway’s principles for revisions.

Revision is not relevant for the cross border trade statistics.