This case study explores the use of payment statistics to compile tourism accommodation expenditure in Austria. The initiative aims to leverage payment data for tourism statistics and balance of payments for travel. The study combines payment data, collected by the Austrian National Bank, with survey data to address issues related to the accrual principle, distinguishing touristic from non-touristic expenditure, and various forms of payment. The payment data provides comprehensive coverage and high granularity, while the survey data offers long-term availability and comparability. The combined approach aims to improve the accuracy of the balance of payments and survey results. The study highlights the strengths and challenges of integrating payment statistics with survey data to enhance tourism expenditure calculations.
Using payment statistics to compile tourism accommodation expenditure in Austria
Abstract
The initiative is part of the travel workshop, a working group aiming to make best use of payment data for tourism statistics and for the balance of payments for travel. The travel workshop is an interest group that was set up in 2010 consisting of representatives from national banks, national statistical institutes and Eurostat to offer members the opportunity to exchange ideas and experiences. The case study acts as a country case to demonstrate a way of calculating tourism expenditure by payment statistics data. The payment data are available to most EU member countries due to Regulation 2020/2011 of the European Central Bank that made it mandatory for national banks in the European Union to compile cross-border card-based payments transactions via resident payment service providers.
The initiative makes use of both payment data and survey data. This combined approach offers favourable conditions to reduce issues in connection with the accrual principle (expenses should be recorded in the period when the service was rendered, independent of the time of payment), the distinction between touristic from non-touristic expenditure, different forms of payment, package tours and bookings via online platforms. In combining these datasets, the strengths of both methods remain intact, while major drawbacks can be minimised.
The payment statistic is collected and aggregated by the Austrian National Bank, that makes it available to Statistics Austria to compile the travel item in the balance of payments. The main strengths of payment data lie in its comprehensive coverage on a global scale combined with a high granularity regarding the kind of goods and services purchased. As a result, it is not restricted to limitations caused by sample surveys. This high granularity is achieved through the categorisation of transactions by so-called merchant category codes (MCCs), which first became available for data relating to 2022. The payment data contains no single transactions but aggregated transaction values by country, MCC and quarter of the year.
The survey data originates from the demand side survey stated in Regulation No 692/2011, which obligates EU member states to conduct demand side surveys to compile the touristic activities of their respective residents abroad. The main strengths of the survey are the long-term availability in many countries with comparable methodologies and well documented metadata, supporting the comparability between countries.
It is envisaged that the results will be used to improve the accuracy of the balance of payment and the quality of the survey results, for example, by identifying significant outliers. The aim of the initiative is to develop a reproduceable way of calculating touristic expenses abroad and is divided into milestones. The first milestone - the calculation of accommodation expenses - was reached in the first quarter of 2024. The next milestones are ways to calculate expenses for food serving services, local transfers, and goods and services. Depending on how well the data can be checked for plausibility and cleaned from non-travel related expenses, these are expected to be reached during 2025.
The development of this project started with the first granular data containing MCCs becoming available in the course of 2022 and gained momentum in 2023, when the travel relevant MCCs were broadly defined by a subgroup of the travel workshop. To clarify the advantages and possible uses of the new payment data, five subgroups were formed. Each of which was entrusted with sub-aspects and recorded their findings in "technical notes". This initiative developed as part of the work by the technical note 2 subgroup, which is entrusted with reviewing the possibilities of integrating payment statistics into the balance of payments travel item and other tourism statistics.
The main contributors to the initiative are Jürgen Weiß and Jörg Grissenberger from Statistics Austria and Patricia Walter from the Austrian National Bank. In a broader sense, the subgroup members who contributed to the creation of the technical notes on travel related MCCs, integration of payment statistics, accrual principle, travel agencies and residence issues have also contributed to the initiative, since the work is based on findings of those subgroups.
Because the Austrian National Bank provides Statistics Austria the data free of charge by email, no extra costs were incurred in connection with the initiative, besides the value of hours of work invested. Since the payment data are connected to an EU regulation, the long-term availability is very likely.
An underlying assumption is that there are MCCs that are strongly connected with tourism and travel. One of the declared aims of the travel workshop working group is to find and define this set of codes. Although this may not be a one-off solution and will require adjustment from time to time, there are many codes, which are quite directly linked to travel and even clearer to tourism.
The first step was finding MCCs that are typical for touristic expenses (for example accommodation, restaurants) and leaving aside those, that are clearly not typical (for example door to do sales, concrete work contractors). In most cases, this distinction was clear. These are defined as the "touristic MCC set", which consequently is a subset of the MCC set for travel. The complete list of codes attributed to tourism can be found in Table 1.
Source: Statistics Austria
It is important to understand that this MCC set still contains non-tourist expenditure, for example from border workers or students. Hence the logical second step is clearing the data from non-touristic expenses. Conventional methods to compile non-tourist expenses were already in existence but needed some transformations to meet the demands of the method. This was achieved in two phases. First, by deducting expenditure items that cannot be found in the touristic MCC set at all (like housing rents or tuition fees) from the seasonal/border worker or student expenses. After that, a credit/debit card share of the remaining expenses had to be determined by using administrative data, ECB data and assumptions. The resulting values are then deducted from the payment statistics data for travel, resulting in touristic expenses from Austrian residents.
A serious and obvious disadvantage of payment statistics is that not all forms of payment are covered. Cash payments are not visible at all, and bank transfers are not visible in sufficient detail. Another drawback is the risk of distortions when using data for credit card payments via online transactions, as opposed to transactions conducted on site. The data can show wrong destination or origin countries, for example because of online booking platforms. It is also challenging to determine pre/post trip spend and to allocate it to the quarter when the trip took place.
The last two drawbacks have been greatly reduced by using only face to face transactions that were done on site via card readers. Of course, this procedure further reduced the absolute payment value and the informative value with it. And it does not solve the problem with cash payments and wire transfers.
The role of the demand side survey in this combined approach consists in its ability to provide a reliable total amount for tourism expenditure. Since it follows the accrual principle, is completely based on residents, contains all forms of payments and all types of trips (including package tours) it is a well-suited complement to payment statistics, where all those are weak points.
The basic idea was to use the global amount for tourism expenditure for accommodation, local transport, food serving services and goods and services. Each total from the survey is then allocated to individual countries based on the percentage distribution according to the payment statistics. In this way, the missing amounts (cash, wire transfer) are included in the survey results, and no estimates for the totals are needed. The approach in its current state implicitly assumes that Austrian residents are equally likely to pay by credit card in every country. This simplifying assumption can lead to certain distortions in countries where the acceptance rate of either cash payment or credit/debit card payments is particularly low. For this reason, it is planned to integrate additional questions into the demand side survey from 2025, which will be used to examine payment behaviour when traveling abroad. Provided that plausible weighting factors can be derived, additional weighting steps will be implemented in the future to minimise the inaccuracy.
At the same time, the survey benefits from the fact that expenditure for less popular travel destinations can be (more accurately) reported since the payment statistics is not subject to confidence intervals. This also makes it easier to identify outliers, for example when the survey shows a few trips to a country with unusually high (but still possible) expenditure. This can lead to an obviously overestimated extrapolation of tourist expenditure in a country. In such a case, the payment statistics provide a corrective factor.
It was helpful to be able to use an existing MCC list for travel as a starting point for the MCC list for tourism. Even though there may be shifts in the future and individual countries may have different definitions due to national circumstances, this travel list represents a largely harmonised basis. In addition, the technical notes of the travel workshop subgroups provided good indications of the difficulties that needed to be considered. In conclusion, the critical success factor for the initiative is manifested in the exceptionally good fit for many destinations between payments statistics, demand side surveys, and the nights spent at tourist accommodation establishments by country of origin of the tourist published by Eurostat. A comparison between the demand side survey and payments statistics is shown in Figure 1.
Note: (1) Incl. Vatican and San Marino, (2) Incl. Andorra and Gibraltar, (3) Incl. Monaco
Source: Statistics Austria
This can be taken as an indication for the validity and reliability of the method. However, because tourism is a multi-layered phenomenon, several constraints and obstacles had to be addressed during the project.
All travel MCCs were unambiguously assigned to one of the four aggregates by a subgroup of the travel workshop: accommodation, food serving services, local transport, goods and services. However, this assignment did not always match the way the categories were formed in the survey. For example, payments at supermarkets in the payment statistic was agreed to be attributed to "food serving services" by the travel workshop members, but it can be found under "goods and services" in the survey. The expenditure aggregates of the survey had to be linked with the aggregates of the payment statistics to compare the expenses with each other.
Another issue was to find a reliable method to consider package tours and trips booked via a travel agency. Travellers usually book a trip via a travel agency in the country they live, resulting in a domestic transaction that is not part of payment statistics. Travel agencies pay hotels and service providers either abroad or in the same country usually without a card involved (for example by wire transfer), which also makes this payment invisible. This issue was solved by calculating package travel shares and travel agency shares that indicate the relative importance of package travels and travel agencies for each country separately. The shares were derived by analysing survey results from three previous years in aggregate.
A similar approach was used to distinguish expenses for holiday trips from business trips, since only the survey offers information to this topic. Therefore, the survey findings were used to split the total expense per aggregate and country into holiday and business spend.
The other three aggregates were tackled in 2024 and put the method to the test. For example, it was found that neighbouring countries can behave differently, especially for gastronomy, goods and services, as day trips and frequent border crossings might account for a significant proportion of expenditure, although they are not (sufficiently) covered in the survey. For this reason, special calculation methods were developed, and the distribution of expenditure was decoupled from the rest of the world.
Concerning expenditure for local transport, it proved helpful not to concentrate solely on the distribution method, but also to take into account a link with accommodation expenditure.
An appropriate solution needs to be found for assigning expenses to countries that do not show payment transaction because of political or other reasons (such as Russia). If these countries show travel expenses in the demand side survey, these results may have to be imputed as a proxy. If not, other sources should be identified to fill the gap. Work is done to develop solutions, and this area offers room for improvement.
Initially, it was considered to also include mobile phone data indicating the presence of Austrian SIM-cards abroad for the country allocations of spend. This idea was later discarded as the available data was too imprecise. The data held no possibility to distinguish between different types of travellers (such as tourist vs transit), did not contain any information on length of stays and was quite noisy for neighbouring countries. If more precise data becomes available in the future, it could be suitable for this purpose.
The milestones concerning expenses for the aggregates for food serving services, local transport, and goods and services, will pose further challenges. In general, a lower share of card payments is assumed within these aggregates compared to accommodation, making a weighting per country more important. MCCs describing local transport also contain expenses for arrivals and departures. The aggregate goods and services (defined in the touristic MCC set) is comprehensive and shows unexpected high totals in the payment statistics as compared to the demand side survey. As such, further investigation is still needed to reach the milestones as planned.
For further information please contact:
Jörg Grissenberger, data scientist, Statistics Austria, Joerg.Grissenberger@statistik.gv.at;
Juergen Weiss, Head of Tourism and Travel Statistics, Statistics Austria, Juergen.Weiss@statistik.gv.at
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.