This case study explores the project to regionalise the Swedish Tourism Satellite Account (TSA) using transaction data. The project includes the calculation of domestic tourism spend using data from Visa and Mastercard. This approach has overall resulted in cost savings in Sweden due to lower costs associated with surveys on tourism spend.
Using transaction data to measuring regional tourism in Sweden
Abstract
The Swedish Agency of Economic and Regional Growth is the Swedish national statistical office responsible for producing accommodation statistics and has for the last decade also published the Tourism Satellite Account (TSA) on a national level. In 2021, the Swedish government provided extra funding to the tourism sector in response to the COVID-19 pandemic. The extra funding covered both the development of tourism statistics as well as tourism promotion. It was in this effort that the agency decided to explore the potential of developing a regional TSA (RTSA).
Both transaction data and mobile positioning data were considered and purchased. When exploring the data, transaction data proved to be a better fit for purpose, as it contains actual transactions for specific postcodes. From this, it was possible to derive a usual environment. A person moving outside their usual environment on a non-regular basis will be categorised as a tourist in the model.
Since 2021, transaction data from the card companies Visa, Mastercard and the acquirer Nets has been purchased. This data did not only help to regionalise the TSA but has also improved national numbers. The regional data from 2019-2023, as well as updated national numbers for the same years, will be published in 2025. The Nordic Council of Ministers had a similar support package for tourism around the same time that this initiative started, and it was decided that this initiative would be spread to all Nordic countries. The Nordic RTSA project runs in parallel to the Swedish, and it is the aim that all Nordic countries in the future will have fully comparable tourism statistics at the regional level.
The project started in Sweden in 2021, when the agency received extra funding for tourism statistics, and the RTSA for Sweden will be published in 2025. The Nordic RTSA project has a plan to continue until 2026 and bring the other Nordic countries on board to have fully comparable Nordic tourism statistics at the regional level. The project is run by the Swedish Agency of Economic and Regional growth, but the model has been developed in co-operation with Visa, Mastercard and Nets. The Nordic part of the project is funded by the Nordic Council of Ministers and runs in collaboration with VisitDenmark and Centre for Regional and Tourism Research, with all Nordic countries as members of the project.
There was a high initial cost for the transactional data which included the cost of setting up the model with the card companies. This was funded by the extra funding from the government. Once the model was set up, the cost for buying updated data was substantially lower. For Sweden, the yearly cost of buying data is roughly EUR 80 000. Meanwhile, the costs of surveys have been reduced, as transaction data can be utilised instead of survey responses to estimate the tourism spend. Overall, the Sweden is saving money with this approach.
The main purpose of tourism statistics is to measure the economic impact of tourism on the national economy. There is also a wish from regional and local destination management organisations and actors to have statistics on a local level. To produce regional or local statistics using the current survey data, a generalising input-output model would be necessary. Alternatively, a new survey which would take this geographical dimension into consideration could be conducted, however, this would be time consuming and costly, as Sweden consists of 290 municipalities. Therefore, access to transaction data is an important component in the regionalisation of tourism statistics. However, the credit card data comes with its own limitations and issues to be addressed, and more work still needs to be done, especially when it comes to online purchase, accommodation platforms, and travel expenditure.
The consumption of international tourists in the national TSA is considered of high quality, as this conforms with the Balance of Payments. For this tourist type, a top-down approach is used, dividing the national figures by the municipal proportions from the credit card data. Analysis indicates that the credit card data confirms well with the national TSA data.
The data for domestic tourism spend (Swedish tourists within Sweden) used in the national TSA was inadequate and unstable, as it relied on national surveys and other data sources. Instead, bottom-up approach is taken, compiling the national figures by aggregating the numbers for each municipality to a total. The greatest level of uncertainty for the bottom-up approach comes from the assumptions regarding cash-payments, online purchases and person-to-person payments.
The spend is compiled using data from the card companies Visa and Mastercard and a third party working with payment infrastructure, the acquirer Nets. The data from Nets is used to enumerate for company cards and other card types not included in the Visa and Mastercard data. Finally, payments that are not part of the transaction data, such as person-to-person payments, cash and online spend, is added. Additionally, in some product groups, the consumption at travel agencies is added, as the travel agencies buy/sell accommodation and tickets for travel and culture. The calculation of the total domestic tourism consumption can be written as below.
Where is the consumption in numbers, stands for person-to-person payments, and ∂ is the partial enumeration. The partial enumeration depends on the product category.
The credit card companies have information about the card number (a unique identifier), when a transaction was made, and the terminal number (the merchant’s number). The credit card companies can follow the transactions of a specific card number over time. Several assumptions have been made to identify tourism transactions in the data. Visa and Mastercard provide data at aggregated level, split on municipalities and product groups. Only private cards are included, as the usual environment for company cards is more challenging to establish. In many cases, company cards are primarily used when the credit card holder is on business trips, in other words, for tourism. The Nets data are used to enumerate for company cards and other card types (such as America Express or Diners Club).
Every merchant within Visa and Mastercard’s card schemes are assigned a merchant category code (MCC). This code indicates the area of business the company operates within. There are codes for example for airline companies, restaurants, grocery stores. The MCCs have then been linked to the product categories valid for tourism statistics, such that only merchants within a certain tourism connected MCC are included in the data. The MCCs can be linked to ISIC and NACE classifications.
The Visa data are subject to a compliancy rule concerning the level of detailed that can be provided. This is called the 5/50 rule: if there are less than five companies, or if one company has more than a 50% share of the market in one municipality, data will not be received at this level of detail.
Identifying tourists from usage patterns
Following the International Recommendations for Tourism Statistics, a trip to a main destination outside the usual environment, for less than a year, for any main purpose is considered tourism. The usual environment is defined as the geographical area (not necessarily an administrative unit) within which an individual conducts their regular routines. Most economic activities associated with tourism occur while visitors are outside their usual environment. Nevertheless, the TSA also includes some consumption that takes place within the usual environment, for instance, services delivered before the trip and clearly related to the trip (for example, vaccinations, passport services, medical controls, travel agency services) and products acquired before a trip that are intended to be used during the trip (specific clothes, medicines) or bought as gifts. The timing is also relevant, as items such as transportation and accommodation often are booked and paid for before being consumed. The corresponding payment might also happen after consumption when paying off a credit card or a special loan taken specifically for the trip.
The usual environment in this project is defined as the municipality or municipalities where transactions are made on a day-to-day basis. It is a place that is visited frequently, for example the home or workplace. In a metropolitan area, it is likely that the usual environment consists of more than one municipality, for example one could live close to the border of another municipality, or the workplace can be in a municipality different from the home municipality. In both examples, transactions can take place on a regular basis outside the home municipality. This consumption should be part of the usual environment and is not tourism consumption. Trips to secondary homes are defined as tourism trips regardless of the frequency of visits. Two separate models have been set up with Visa and Mastercard to define the usual environment.
Visa
Visa defines the usual environment as one or two municipalities, the home and work municipalities. The home municipality is defined as where the majority of the face-to-face transactions occur. The work municipality is defined as where the card holder makes most transactions at MCCs such as grocery stores, supermarkets, restaurants, transportation, and fuel during weekdays at business hours (9 AM to 5 PM). The home and work municipality can be the same. If there is a tie between the two, then the municipality with the biggest amount of spending, is considered the home municipality. Data on all transactions by credit cards in Sweden has also been received from Visa, which allows for the compilation of the tourism share of spending in each product group.
For about 21% of cards, Visa could not define a usual environment from the model. However, those cards represent a negligible number of transactions, as 99% of all transactions were made with a card with an identified usual environment. The cards where the usual environment could not be defined were primarily used online or had less than 20 transactions over 3 years. A work municipality could be defined for 88% of active cards, and 18% of active cards have a work municipality that differs from their home municipality (two municipalities as their usual environment). It is worth noting that around 35% of Swedes work in different municipality from where they live, however, that does not necessarily mean that they spend a lot of money there.
The Visa data was also used to investigate commuting across the border to Sweden. 11% of international cards had made at least one transaction in Sweden per month for three months. These cards on average appear in Sweden 5 times per year. It is assumed that excluding these cards would exclude important border trade and frequent trips to summer houses in Sweden. Therefore, all inbound card transactions in Sweden are defined as tourism transactions.
Mastercard
Mastercard’s model is different to Visa’s. Two different models were set up by choice to investigate the best fit. The conclusion for now is that a combination of the two models works better than relying on one. Mastercard defines the usual environment as where most transactions occur, but allows for up to eight municipalities, using the number of trips as an indicator of whether to include an additional municipality in the usual environment. A municipality is defined as part of the usual environment if a card has made trips for more than 21 days or has made more than 8 independent trips to the same municipality over a year. This model is less rigid than the model set up with Visa. The model may include vacation homes in the usual environment, however, as 35% of Swedish workers commute over municipality borders, this model could also better reflecting commuting statistics. 52% of cards have one municipality as their usual environment in Mastercard’s model, 29% have two, and the rest have three or more.
Enumeration for other types of transactions and accounting for non-card spend
Although private Visa and Mastercard account for a majority of payments made by tourists in Sweden, some further adjustments are made to the transaction data. First, there are other types of card payments such as business cards and other card schemes than Visa or Mastercard. This was addressed by acquiring data from Nets and using this data to enumerate the credit card data per municipality. These corrections have been relatively small. Business cards are used for less than 5% of all spend, and other card schemes, are used for less than 1% of spend. This is true for all types of categories except accommodation and air passenger services. This is one of the reasons that accommodation and airport statistics are used for the spend distribution in municipalities for these categories rather than credit card data.
Cash and person-to-person payments are also addressed using rough estimates. In the Nordic countries, the use of cash is decreasing. Person-to-person payments such as Swish (Sweden), Vipps (Norway) or MobilePay (Finland and Denmark) are relatively frequent and have partly taken over from cash as easy direct payments. It is also a frequent transaction method for small businesses and companies, especially for small sum transactions. As with cash, person-to-person payments are untraceable in the transaction data. Recent surveys from the national bank of Sweden indicate that cash roughly accounts for 8% of all payments, however, as cash is primarily used for small transactions, it is expected to account for less than 8% of the value of transactions. This correction is not applied to all product categories due to differences in payment habits. More information on cash transactions is needed and further development of national and border surveys could help fill this gap.
Online tourism consumption cannot be distinguished from regular/everyday consumption and is therefore treated separately. In some product groups, online purchases dominate the transactions, especially for product groups such as accommodation, travel and culture. Mastercard provides the total amount of online transactions by Swedish cards in Sweden and the rest of Europe. This is divided per product category relevant for the TSA. Visa provides the total amount of online purchases in Sweden for each category split by Swedish and international cards. A product-by-product approach is used: for the products where online purchases are more prevalent, such as accommodation and train-tickets, the total spend has been calculated using other sources. In other product groups such as restaurants, groceries and fuel, it is assumed that no online purchases are made by tourists.
Transaction data has been useful to compile the regional and municipal TSA in Sweden. The results from the transaction data have been contrasted against the national total from the existing national TSA. For some product groups, top-down data has been used as the transaction data are not deemed to include all tourism spend. This is the case for product categories where most spending is done before the trip online, such as airline tickets and hotels. In other product categories, such as restaurants, it is estimated that the transaction data are better than surveys. Even when the top-down method has been used, the transaction data have been used for the regional breakdown.
Several assumptions have been made to calculate the RTSA. These assumptions may be subject to change over time as new information becomes available. Each product category needs special consideration depending on its nature. This is especially true at regional and municipal level. Assumptions were also made to estimate missing values as a result of data protection rules. The main takeaway from this project is that transaction data can provide an efficient and timely way to compile the RTSA. In the Swedish case, using transaction data also resulted in cost savings.
For further information please contact:
Maria Wiberg, Analyst, Swedish Agency for Economic and Regional Growth, Maria.Wiberg@tillvaxtverket.se, Turismräkenskaperna - Tillväxtverket
This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.
© OECD 2025
Attribution 4.0 International (CC BY 4.0)
This work is made available under the Creative Commons Attribution 4.0 International licence. By using this work, you accept to be bound by the terms of this licence (https://creativecommons.org/licenses/by/4.0/).
Attribution – you must cite the work.
Translations – you must cite the original work, identify changes to the original and add the following text: In the event of any discrepancy between the original work and the translation, only the text of original work should be considered valid.
Adaptations – you must cite the original work and add the following text: This is an adaptation of an original work by the OECD. The opinions expressed and arguments employed in this adaptation should not be reported as representing the official views of the OECD or of its Member countries.
Third-party material – the licence does not apply to third-party material in the work. If using such material, you are responsible for obtaining permission from the third party and for any claims of infringement.
You must not use the OECD logo, visual identity or cover image without express permission or suggest the OECD endorses your use of the work.
Any dispute arising under this licence shall be settled by arbitration in accordance with the Permanent Court of Arbitration (PCA) Arbitration Rules 2012. The seat of arbitration shall be Paris (France). The number of arbitrators shall be one.
This document was produced with the financial support of the European Union. The views expressed herein can in no way be taken to reflect the official opinion of the European Union.